Marine Magnet Dispatch Service Centre

Top 10 Product, Production and Performance "Digital Twin" Sets Consist of Multiple Virtual Models

1/13/2019

The concept of “Digital Twins” is beginning to expand beyond just the notion of a virtual product. The idea of the digital twin is that all of product data is contained together, from initial design files through simulation and materials data. Finally, data about the product’s performance in the field will also be added. Lately, the term is also including the product’s manufacturing data.

The twist that makes the Digital Twin concept important is that computers have greater predictive skills that humans working with physical prototypes. “The digital twin a good definition of what we do. It’s a virtual representative of a physical product or operation.

The point is the precision we can reach. A lot of our technology acquisitions have been in simulation, test, and electronics. If we can improve the precision of the models we’re creating, we can predict performance.”

Prediction is the key to the Digital Twin. If a computer can calculate and predict how materials perform, that computer can create the optimum configuration for the product. System level simulation is predicting how we expect that system to perform. Design requirements can be improved and move into the design phase with a higher set of requirements.”

As the data is moved through design and into test, the original design files remain attached as part of the Digital Twins data. “The Digital Twin comes with CAD and tool development. That gives a lot of information to the downstream processes.. If we understand the language of the product from its inception, we can see if it is meeting the desired threshold and how it will perform in the future.”

The goal is to create a model of the product that not just downplays the need for a physical prototype but actually surpasses what a prototype can provides. We can see the Digital Twin as many twins – the 3D simulation through the product in the field. All of that feeds back into the Digital Twin model. “If you look at the physical design, it’s generative design. Let the computer look at weight or aerodynamics. The physical analysis, the tech systems create the shape based on the constraints.”

The Digital Twin in the manufacturing process brings together two virtual worlds, the digitised product and the digitised shop floor processes. Beyond that, data relating to the supply chain is also part of the mix. “Using the Digital Twin of the factory, we can see how the factory is performing by looking through all of the data. We can look at future orders and create a model of how the factory would need to perform to meet the projections and thus predict the future.”

Take aerospace as an example of how a Digital Twin can assist with the performance and maintenance of a product. One application is to create a Digital Twin of every plane the comes down the line. We can understand the precise configuration that operates at any given time and create a Digital Twin to support the maintenance of each plane.”

The Digital Twin technology can benefit a wide range of industries, but it is particularly well suited to high volume, high mix products. “All of the discrete industries are benefiting from Digital Twin technology now. The industries benefiting the most are those ones what put a large emphasis on physical design like aerospace, To create a high-volume production run, you have to have high reliability in the model. Precision level is your confidence level. The Digital Twin has to be accurate.

A virtual reality function in the Digital Twin package may help in replacing the physical prototype, since it will give users a powerful sense of the product. “We use augmented and virtual reality to impact the feel of the aircraft. In the near future, there will be more confidence in viewing a product in virtual reality. They’re doing that in the test phase. In aerospace, physical prototypes were the name of the game, but we were going to get more accustomed with looking at the product in the digital world rather than as a prototype.”
　
Others define a Digital Twin as digital replicas of different assets, processes and systems in your business which can be used a number of ways. This generic definition is basically correct. However, a Digital Twin is more accurately described as an integrated set of digital replicas or models driven by a rich information model called a digital thread.

A true Digital Twin is not a single model of the asset, even though it’s referred to as a singular “Twin.” It consists of many mathematical models and virtual representations that comprehend the asset’s entire lifecycle – all the way from ideation, through realisation and utilisation – and all its constituent technologies, including electronics, mechanical, manufacturing and in-service performance.

We refer to a set of Digital Twins: Product, Production and Performance. Each of these Digital Twins consists of multiple virtual models appropriate for the given product and production system.

Accurate Digital Twins comprehend the interactions between all aspects of the product and production system. Digital Enterprise Tool was built on our Innovation platform, enables the creation of the most accurate, comprehensive Digital Twins possible.

Product Digital Twin. A Digital Twin for products is typically created using our Systems Driven Product Development tech, which drives the creation of intelligent 3D models. Technologies like Convergent Modeling, Generative Design and Predictive Analytics are the basis for these intelligent models.

Key enablers for System Product develop tech include our comprehensive, semantic digital thread data model, as well as a complete set of integrated tools to help you create model-based system representations and accurate Digital Twins of the product, and these Digital Twins are able to comprehend the impact of design changes on the production system.

A product Digital Twin will typically include electronics and simulations applications, finite element structural, flow and heat transfer models and motion simulations.

Product Digital Twins also rely on predictive engineering analytics, which applies multidisciplinary engineering simulation and tests with intelligent reporting and data analytics. These capabilities lead to digital twins that can predict the product’s real-world behaviour throughout its lifecycle.

Predictive engineering analytics includes tools manufacturers leverage to expand traditional design verification and validation into a predictive role that supports system product development. The ultimate goal of implementing a predictive engineering analytics strategy is delivering innovation for complex products faster and with greater confidence.

Production Digital Twin. The Digital Twin for production uses many of the same tools and techniques to create a twin of the production system. Production Digital Twins, like product Digital Twins, can include 3D models and predictive analytical models as well as production engineering specific models that use tools for conceptual design and virtual commissioning of smaller Production Systems and Machine Tools used to engineer and virtually commission large production lines.

Performance Digital Twin. The Digital Twin for performance is based on tools enable insight discovery, analysis and monitoring from in-service products and production systems. Performance Analytics quickly identifies product issues disrupting the supply chain, manufacturing process or customer experience.

The performance Digital Twin may also include data analysis to discover hidden product issues before they occur; graphical displays to clearly identify potentially problematic configurations; and automated data monitoring to fine-tune operations and provide insight for improving your products.

Virtual Product Development is used to develop a product heavily – but not exclusively - relying on digital representations throughout the whole or a part of the project’s life cycle. The term “virtual” expresses that it does not yet exist as real hardware, that in fact it is not touchable.

Virtual development makes use of many tools for layout, geometry generation, calculation and analysis, test and evaluation including the management of digital data.

This suite enables to concurrently anticipate a lot more considerations sooner for product and process design than in earlier days. Although mostly relying on and working with approximated data, calculations and simulations today come close to verification and qualification test results.

The Digital Mockup is a core element in Virtual Development, as it is the culmination of the design intent that gives the product an “early face”. If Virtual Product Development is the unifying concept and approach, Virtual Prototyping actually is the process of digitally testing and evaluating the virtual representations of the product in all aspects of developmental and operational life. So the Digital Mockup is embedded in a spectrum of prototyping activities, which aim to mature the product as fast, as less costly and as reliably as possible

Construction of Hardware Mock-ups has been an integral part of any complex product development. A Mock-up is traditionally a hardware or physical model of a component, an assembly or an entire product. It can be full-size or a scaled model of metal or actual production hardware. All military airframe integrators use Hardware Mock-ups to evaluate and verify the design, train personnel or present them to customers.

Major Hardware Mock-ups activities involve the assessment of space allocations, detailed part fitting checks, interference and clearance studies, part installation and removal and assembly verification. A Hardware Mock-up as primary tool for determining lengths of electrical harnesses and for fitting tubes, hoses and other routings into a densely packed structural space. Maintenance checks could also be run with them. Mock-ups for marketing purposes still represent an important means in the commercial business and are likely to do so for quite some time.

Hardware Mock-ups, though having been standard and successful industry practice for a long time have a number of shortfalls that are better addressed with their digital substitutes. In fact, they are only an assurance that everything will fit together, in former times based on drawings.

No “pre-mock-up” investigations can be run. Update with modifications or duplications are expensive, as always a physical component has to be produced. The response to changes is relatively low except for minor ones like drilling holes and there is an inability to reflect real-time configurations.

So the mock-up is usually representative only for one aircraft, and is of very little use afterwards e.g. after certification. Last but not least, maybe the major issue, it is quite costly. There are design, labour and maintenance hours to be paid for, not to forget tools and all the materials. In addition, it consumes precious factory floor space.

The majority of Digital Mock-up applications, as the classical substitute of Hardware Mockups, cover the geometrical and functional areas. The closer one tries to assess the behaviour and interactions of the product in its environment the more will efforts shift to the right end of the spectrum.

Commercially available Digital Mock-up, Digital Simulation tools and Virtual Manufacturing/Digital Factory tools have matured as they cover most requirements for geometrical and functional assessment, and, to a lesser extent, operational constraints.

Shortfalls have to be compensated either with specific tools e.g. for tolerancing, tool developments or with physical specimen. To fully cover the whole spectrum requires all design tools to efficiently communicate with each other so that results obtained individually can be compressed to an overall view of the Virtual Prototype.

The majority of Digital Mock-up applications, as the classical substitute of Hardware Mockups, cover the geometrical and functional areas. The closer one tries to assess the behaviour and interactions of the product in its environment the more will efforts shift to the right end of the spectrum.
　
1. Baseline

A baseline is an agreed set of 3D and non-3D data at a certain time during development that is used as the reference for all design activities. It represents a preliminary status and is the starting point for the next iterations. It is a configuration of a product or status of product data, formally established at a specific point in time, which serves as a reference for further activities.

2. Change

Term used to identify a definition progress over time with reference to a basic or technical definition of a product managed by means of a modification system.

3. Dual-use

These are technologies, manufacturing facilities and products that have military and commercial applications. Commercially produced items, hardware or software, that can therefore be used, with or without adaptations, for military equipment.

4. Effectiveness

The effectiveness of a system is a quantitative measure of the degree to which the system’s purpose is achieved. Effectiveness measures are usually very dependent upon system performance.

5. Iteration

A repetition and rework activity that encompasses multiple passes for the design to converge to suit an array of sometimes conflicting specifications.

6. 3D Master Model

A Master Model is a set of digital 3D data-- surface or volume used as the basic reference for design and/or manufacturing. It progresses during development and when fully detailed it becomes the input for production, documentation and verification activities.

7. Modification

Any controlled change by the Modification system to the definition of the aircraft or equipment whose introduction affects airworthiness/certification, operational serviceability, customer or own company contractual/financial considerations.

8. Pilot

A pilot is a near term demonstration project. It can be a proof-of-concept and a preoperational assessment. Done in laboratories or usually in small-scale business units it serves to find out flaws in the behaviour of the system under operational or near operational conditions.

9. Prototype

A prototype is the first or original example of something that has been or will be copied or developed. It is a model or preliminary version.

10. Version

A specified customised definition of an allocated aircraft within a given production standard/model.

0 Comments

Top 10 Stages in Internal Design Team Build of Sketch Models/Prototypes Create Run Samples

1/13/2019

0 Comments

When utilising Design Tools, one of the common requests is why can’t it be more automatic. There are many people that would like to simply load a model, and have the tool determine the best way to mill it, then automatically program it.

While this “Fantasy World “ of design programming may not exist for all types of parts, that precisely is one of the goals of Automatic Feature Recognition. To sort out common elements of the 3D model and using rules and templates, create the code on the part.

Solid models have all the necessary information to describe a parts shape. Users should not need to waste time re-creating part data just for the CAM system. Most prismatic parts are comprised of standard features such as holes, pockets and bosses. Where other CAM systems may require users to define geometry, create boundaries, and then specify cutting operations, automatic feature recognition eliminates that. Typical 2D parts can often be programmed in just a few minutes, versus an hour or more.

Users can typically run a wizard to automatically find and sort the different features within their part. Typically input the stock material size, give it an initial tool orientation and then the tool analyzes the solid model and creates machinable features for machining.

These features will then be listed and automatically sorted for logical machining operations. Alternatively, users that wish for fine control may select the features themselves interactively, and sort them manually. Usually the best approach is a combined approach of utilising automatic feature recognition with some interactive selection or sorting of features for their specific preferences.

Once features are defined, they are sorted for machining efficiency and then they can be machined with strategies ideal to the properties of the individual feature. For example, an open pocket may be machined with a different strategy than a closed pocket. Or shallow bosses will be milled differently than tall ones.

the features are found or selected a three-step process or wizard can be used to program parts utilising the features. Users first select from a list of features so you can input information on shapes and sizes. Next users select a preferred machining strategy based on the information provided, or go with the recommended strategy. The programmer can accept the strategies suggested based on the type of features for rough, semi-finish, and finish passes, or make changes to fit the machining needs for that particular part.

The combination of feature dimensions, stock material, and cutting strategies are analyzed allowing the tool to recommend the most efficient cutting tools, toolpaths and feeds and speeds for each cutting operation. The programmer can choose the recommended tool or search for another tool in the library. Users can accept or change the recommended feeds and speeds.

Although automatic programming provides a good starting point, users should still retain control over how the tool generates its CNC code. Users can set machining preferences ahead of time for the CAM software to apply to future jobs. Although it recommends tools, feeds, speeds, etc, the users can override it with their own preferences at any time.

Automatic feature recognition can extend beyond using just the tool axis. Utilise multiple setup orientations to find all features on the part, regardless of the orientation. These will then be sorted by workplane, and feature type.

Users can then decide the order of various orientations they prefer to use, and program the features accordingly.

Based on your manufacturing knowledge, automatic feature recognition intelligently makes decisions for you. It automatically selects your tools, stepover, stepdown, and more, providing programming consistency, utilising parameters out of the box, or those that you customise for your own operations. These parameters, among others, form part of your operation.

As you create multiple features, the tool dynamically updates your process planning. Providing an optimal machining order, based on what you want to achieve.

Part changes are an inevitable fact of working in a job shop. If a part has already been analyzed and programmed, than the part geometry changes, one can simply compare the updated part to the original. Feature list will automatically update for the changes, and the generated code will update.

This strategy is not only useful on part changes, but when milling similar parts, or a family of similar parts. Simply utilise automatic feature recognition to sort and program one part, tweaking with your personal preferences, then apply all of those same strategies to a similar part, automatically.

Automatic feature recognition allows for standardisation of the entire machining process. The best expert machinists could set up the machining templates; and all programmers, regardless of experience, would be able to utilise their knowledge and experience. This allows for consistency in your finish, tool life and overall quality of parts produced.

Automatic feature recognition and feature based machining can simplify the machining processes of certain parts typical to the job shop, allowing for programming to be completed in minutes instead of hours. Knowledge gained during the process can be retained for future jobs, for your individual preferences

Previous workflows required engineers to create multiple sketches linked together to form a part that was organic in nature. Direct modeling allows engineers to simply push, pull, twist, etc. to manipulate a part. It meets the needs of the fast-paced design industry and allows engineers to create without the traditional boundaries of parametric modeling.

Many engineers in modern design operate mainly in the direct modeling that is not just about creating cool designs, it also is about having the power to edit your parts without neglecting physical characteristics of the design.

Direct modeling allows you to adjust a design organically without worrying about voiding certain necessary parameters set earlier in the design. This ultimately means that when a designer needs to alter a product per the client’s request – or any reason – it becomes an easy process.

Pretty much all of the modern design programs have direct modeling integrated into them. It’s a workflow that is just too essential to the modern engineer. You can use direct modeling techniques to move, rotate, resize, or scale features from imported geometries. The easy push/pull controls allow designs to be modified for quote. All of this functionality ultimately leads to faster iterations between simulations and directly editing the geometry.

The aerospace industry has always been on the forefront of technological developments and tends to be an early adopter of new technology. Even if these new technologies are only used initially on their concept vehicles or in high-performance racing vehicles. The massive manufacturing advances being made across the board point to a future of energy and cost efficiency, and generative design has placed itself firmly in the middle of this revolution.

Generative design is machine-learning-assisted design and is used to optimise a given design based on a set of user-specified parameter and iterative design solution based on external factors. It must not be confused with topology optimisation which is focused on improving an existing design.

Generative Design takes manufacturability into account when creating all the different configurations so you don’t end up with a good-looking design that would be completely impractical to manufacture. Another advantage is that homogenous parts can be created that replace sub-assemblies of multiple different components.

Multiple configurations are presented to the designer and each of these configurations meet all the criteria in different ratios. The designer can then choose the best configuration. However, if additive manufacturing techniques are being used then it is not necessary to do any modifications on the part as it can be directly printed.

Generative design has massive potential since there are thousands of components that make up a vehicle, and many of these can be revised using this system. There are many benefits of using generative design for vehicle engineering, such as cost efficiency since components use less materials, less energy and less time.

Usually components are designed and put through extensive simulations to verify the design, thereafter the part is tweaked and simulations are run again. With generative design, the simulations are run during the design phase and each of the thousands of iterations created have already passed a strength test.

Design Modeling tools have advantage of reducing time spent on the development of conceptual designs since 3D model is generated as part of the process, freeing up the designer to focus on more top-level design concerns. Once the designer decides on the final configuration, it can be slightly modified to meet designer preference.

As a designer, you need to look at what the factory does with the CAD and the interactions between the design team and factory, along with the creation of the instruction manual and sales & marketing material.

The differentiation between designer and CAD modeler varies depending on the size of the organisation or studio undertaking the design in question. In some cases a designer may take their designs to the CAD stage by themselves.

In other cases a specific CAD modeler may be employed to take on this part of the process. This can vary depending on work load, skill set or even the specifics of an individual design. In any case, the designer will remain plugged into any decisions and deliberations.

Maintaining design integrity is often the most important skill required. A CAD modeler is often asked to demonstrate some 2D concept capability, in addition to their core skills, before they're employed to make sure they're capable of following the design language into 3D.

There are many stages for a design to go through before it ends up in the hands of end-users. After a two-dimensional idea has been explored and resolved, the next fundamental stage in the process is the use of CAD. You will learn the differing uses of 3D in the design process, and look at how the role of the designer and CAD modeler comes into play as they work through the stages of getting a sketch into a form that can be used to generate a finished product.

Turning a sketch into 3D is a fairly simple process; all you have to do is add the third dimension. Doing it really well and maintaining the designer’s intent, however, is an entirely different proposition.

At this point in the process, you now have a solid design foundation to base all CAD upon modeling. You have detailed drawings and illustrations that depict several angles of the chosen design, along with specific details and technical features being noted and explored by the design team.

The factory will take over the CAD and truly engineer the design for most of the above points. The factory will also identify areas to strengthen in order to avoid failure during use and drop testing, consider production issues, such as the number of parts, and assess the production technique in order to minimize costs.

often have a habit of taking the most direct route to a 3D solution, which tends to be a straight line. Straight lines rarely exist in good design. So before the designs are handed off to the factory, the CAD modeler and the designer work in tandem to get as close in 3D to the original 2D design as possible.

1. Keeping the Curves

This stage is set of curves moved into the correct three-dimensional positions, known as a wireframe with no surfaces being built. This is a ‘quick and dirty’ process, much like a first rough pencil sketch, which can feed back into more illustrations from the designer if they’re not happy with certain aspects of the form.

2. Fluid and Flowing

Many tools give you the ability to create fluid nonlinear geometry. They're generally the tools of choice for most studios, as they allow designers to stay true to their sketches before applying engineering constraints. Model sets will often need rebuilding by an engineer or factory team to get to a ‘toolable’ state. Of course, the more experienced the modeler or designer, the less change required.

3. Tight and Constrained

Some tools allow for a high level of engineering from the very first stages of modeling and is often used by factories to create final CAD for tools. To this end, the more technical studios prefer their designers to work within these design spaces to reduce lead time to production. This can, however, result in some loss of some desirable properties since constraints within the tools make it more difficult to create flowing geometry.

4. Organic

Some modeling options are applicable to designs that are of a far more organic nature. Organic surfacing is difficult in most applications. However, despite being able create great organic designs and allow for 3D printed prototypes, the data is not viable for tooling.
.
5. Modeling Process

Designer and the modeler will sit down together and run through the final illustrated design document to highlight the salient features of the design. Turnaround/elevation drawings and specific features required will be discussed, and any questions the modeler has can be further sketched and defined. Depending on time and complexity of the design, the modeler and designer will take the CAD through several different stages to get to the final 3D.

6. Sketch CAD

The first kickoff point can often be a set of elevation drawings in 2D CAD created by the designer or, for the non-CAD proficient designer, a set of elevation drawings. Once the 2D drawings are into the 3D package and arranged in the specific views, a sketch model is produced. Relevant internal components will also be arranged in space to make sure they’ll fit inside the design. This model is a first pass of very basic surfaces to make sure internals can be housed and to get an idea of how forms and surfaces might interact with one another. Objects may overlap and intersect, with the main purpose to get a volume to look at.

7. Mockup CAD for Prototype

In product development there’s often a desire to get something physical from the process as early as possible. If there are mechanisms to test an early prototype is often used, but lacks nearly all of the details and design of the final product, serving as a shell to place mechanical components within and will take the form of a relatively cheap 3D print. Problems can be identified at this early stage, such as space constraints, assembly issues, visibility of sensors, rotation restrictions can be identified to give an early look at potential manufacturing issues later down the line.

8. Refined Prototype

Once the rough prototype has been explored, feedback will be taken and fed back into the CAD process. The model will have evolved into something that looks a lot more like the designer’s intent along with improved mechanical constraints. Considerations for how the product will be assembled will come into play, with accurate parting lines being added, along with the addition of draft angles to allow parts to come out of the mold, and internal structures being added to support components and facilitate assembly, Usage can now be tested properly with engineers continuing to feed into the design, highlighting production issues or material constraints as well as updating the factory with new CAD or providing duplicates of the prototype.

9. Refined CAD

Once the prototype has been thoroughly tested and any issues identified, the CAD is refined further with new ideas being encompassed into the model. Relocation of components or internals will be assessed and practical considerations will be implemented, such as part assembly and interaction sensors or buttons.

10. Visualisation

The final task to be undertaken on the CAD is to create renderings of the design. There are samples at this stage that can be shown to select buyers, but they’re few and far between and relatively expensive to create in numbers so illustrations are still a necessity. Renderings are now used to populate and update marketing documents and sales material, along with illustrations for packaging. In particularly complicated products or ones with very deep interaction or play, the CAD model can also be used to create an animation to further demonstrate the product in various user scenarios and further support sales efforts long before fully functioning samples of the product are produced.

0 Comments

Top 10 Generative Design Benefits Using Digital Simulation Tool Creates Optimal Outputs

1/1/2019

1 Comment

Generative Design is new approach to engineering set to transport the potential of manufacturing over the next decade. It’s a process that enables engineers using digital tools to define an engineering problem, and solve it over and over again by an adaptive artificial intelligence program, yielding different results each time.

if you pool enough computing horsepower, your network is capable of completing many iterations of very complex tasks very quickly. Questions about the potential for many designs used to be fairly straightforward. Design teams have a finite amount of time and money at their disposal, and can’t afford to prototype more than a few designs—let alone thousands.

But what are these constraints costing you, in terms of the ideas that never get tested and the solutions that never get prototyped? Are there unexplored methods of building your product that might be lighter, faster or cheaper?

“There’s really not one solution for every problem; there are many. This is where the power of parallel computing is going to help us test more ideas and look at more concepts in a shorter *period of time.

The “secret sauce” is an AI platform that has been “trained” to create solutions to engineering problems. Unlike human engineers—who first design a solution, then determine how to build it, then prototype and test its properties—generative design is capable of carrying out all three of those steps simultaneously.

The process requires human engineers to define the problem by using CAD systems to lay out basic specifications for the component that needs to be designed. “You don’t start with the geometry. You start where it attaches to other components and how.”

After that, engineers further refine the design parameters, specifying load requirements, deflection, rigidity, material preferences, cost of production, weight requirements and even manufacturing methods, attention to detail is critical during the process of defining problem.

“If you input junk you’re going to get back junk, essentially. “But if you actually specify the requirements early and set up to solve the right problems, that’s when you’ll see the benefits of the tools.

“So then, we can push the magic button, hit generate and send it to processing, and it will give you a bunch of different answers.

Working its way through the predefined parameters of the problem, the AI platform solves it over and over, employing a new approach each time. the program sketches out new CAD designs, tests them in simulations and learns from its mistakes and successes.

What results is a collection of hundreds, even thousands, of computer-generated designs, catalogued by the degree to which they meet various criteria. Some designs bear similarities to the components they’re intended to replace, while others look like nothing that’s ever been manufactured before.

“You can even watch as it solves it over and over and over, right in front of you,”

“The bone-like, geometry takes some getting used to. It would have been a nightmare manually modeling this from scratch. “Once you see it, though, it’s impressive, and there’s nothing like it out there in the market. After it has generated solutions for a problem, generative design tool performance tests them in simulations for buckling, fatigue and failure points. It can even account for the wear and tear of specific manufacturing processes.

Faced with a nearly unlimited supply of generative design solutions, engineers will have the responsibility of sorting through the performance characteristics and combining top-performing features from various designs to meet the specific needs of the product in question.

“Using VR, you can start to move and articulate these things, re-shaping your design in real time. “When you’re able to see how a component fits into things on a one-to-one scale in real life, it’s always better.”

Ultimately, engineering is about problem-solving, and the more data engineers have to work with, the more money they can save, the faster they can get to production and the lighter or stronger they can build a component. With generative design and the power of new AI tools, design teams will be able to tap into a data trove that’s deeper than any they’ve seen before.

“There’s really no one solution to any problem. “There are tons of right solutions, and we want the engineer and designer to get the information they need to make better decisions earlier in the design process.”

Generative design takes an approach to engineering that we’ve never seen before in the digital realm. It replicates an changing approach to design, considering all of the necessary characteristics. Couple this with high-performance computing and AI, and you’re left with capabilities that engineers never thought they would have.

The way in which engineers design is being brought into question with new generative design tools. If you’re an engineer and haven’t seen your workflows altered yet, prepare for the coming future.

The onset of practical AI tools has enabled the possibility of mainstream generative design tools. That means engineers can create thousands of design options inherent to their digital design and choose which design meets their needs to the fullest. From here, you can solve manufacturability constraints and ultimately build better products.

It allows engineers to hand the reins off to their CAD tools to organically find the best solutions to a given set of constraints. Through generative design, collaboration with technology can be organic and flowing. It results in ideas that are better than what you could come up with on your own, and products that are lighter and accomplish their directives better. It simply results in better engineers.

Generative design is a tool that uses machine learning to mimic a design approach similar to designs in the real world. It interfaces with engineers by allowing input design parameters to problem-solving. If you have loads in certain locations, you need to maintain certain material thicknesses, or even keep certain costs, all that can be fed into generative design tools.

After you press run and let the tools do their thing, you’re left with generative designs that meet your input criteria. From there, you can cycle through, pick which design is the most optimised for your design end goals and modify from there. In essence, it takes you down a digital shortcut to optimising the perfect design.

The advantages of generative design become apparent when you consider just what it takes to get started with any design. You approach problems with a general understanding of what your design needs to do, but you’re left to your own creative devices to find a solution. Instead of starting a design based on the idea you have in your mind, you can start by offloading that data into computing tools and allowing it to kickstart the design process.

One of the best examples of how this methodology and thus generative workflows can be practically implemented is by examining how to build a flight cockpit. Instead of starting with some sketches, creating various designs, and picking the best one, you can start by feeding a computer some constraints. Input the cost, the weight it needs to support, and what material you’d like your cockpit to made out of. Then the computer can deliver thousands of design options that take into account manufacturability for you to select from. This is what generative design offers to the modern engineer.

True generative design tools use the power of machine learning to provide sets of solutions to the engineer. This is in stark contrast to tools we’ve seen before, such as topology optimisation, latticing, or other similar CAD tools. All of these previous tools improve existing designs, whereas generative creates a new design.

Generative design is also different from other existing CAD tools in that it can consider manufacturability. If you’ve ever used tools like topology optimisation, you’ve probably been left with an end product that looks great on paper but isn’t easily manufacturable in the real world.

Coupled with this account for manufacturability, true generative design takes into account simulation throughout the entire design process. On the front end, that means taking into account your manufacturing method, and the tools will take care of simulating a given design’s feasibility. This only yields designs that meet necessary simulated criteria and is manufacturable.

Computers that creatively come up with ideas on their own are the heart of generative design. In generative design, you share your goal with the computer, tell it what you want to achieve, as well as the constraints involved, and the computer actually explores the solution space to find and create ideas that you would never think of on your own.

Project that lets designers describe the forces that act on an object and then lets computers go off and make it. These forces can be structural loads or even manufacturing methods. You start by sharing the goal with the computer, telling it not what you want it to do, but what you are trying to achieve. You describe your well-stated problem, and then, using generative methods, the computer creates a large set of potential solutions, automatically putting them together.

But here’s the key: In the time it would have taken you to do just one design, the machine has done all of them. Its design proposals are delivered back to you in an explore tool, and you can then start steering through the various designs and understanding the tradeoffs between various solutions. This process may enable you to find something interesting, helping you redefine the problem so you can repeat the loop, but, ultimately, you’re going to select one of the computer’s designs to fabricate.

Good results and bad results that have a lesson in them all disappear, and the computer treats the next question as if it had never seen anything like it before. But what if there were a machine-learning system in the loop so that every time you analyzed something for its aerodynamics, the computer got an impression of the connection between cause and effect? What would the results be if that happened over and over again?

Eventually, you won’t need a deep analysis to arrive at an opinion, because you will have deep-learning system that can tell you its hunch. It can always ask if you want a more serious investigation, based on tried-and-true analysis code.

Engineers like having that second opinion available, but you can be satisfied with the quicker one sooner: an basic understanding what aerodynamics means. Then you can show the computer a brand-new thing it had never seen before, and it could give me an opinion about whether or not it was aerodynamic.

it’s remarkable that computers are going to start having opinions. But here’s the really interesting part: Imagine if the computer were doing this in its off hours. What if it were generating new forms on its own, putting them through analyses, and seeing the relationship of cause and effect? The systems that understand those linkages will start getting hunches about things that a company is working on.

Pretty much every company offering new tools is exploring the potential of generative design , and they all have really pretty pictures of prototypes that have been designed with their various tool packages. But is there more to the technology than fancy prototyping? When are we going to see these products appearing on our shelves? Is generative design even suited to mass production?

After all, generative design has been enabled by the growth of additive manufacturing, which currently still has a long way to go before reaching production run levels equal to those of the traditional manufacturing techniques used in mass production.

Additive manufacturing has been a powerful driver behind the recent surge of interest in generative design, and given that additive manufacturing is not exactly well suited for mass production yet , you could be forgiven for thinking that generative design’s dependency on additive manufacturing is hindering the breakout of these products into mass production.

“Physics-driven design can be tailored to produce designs that can be made with traditional manufacturing techniques, such as extrusion, forging, casting, etc.,” “While often not as optimal as designs driven without traditional manufacturing constraints, the unit costs can make optimisation viable for lower cost/higher production run components.

“We are seeing this already and have been driven by customers to enhance tool capabilities so that techniques like casting, extrusion, etc., are viable alternatives for a physics-driven optimal design.”

So, it’s clearly not a case of traditional manufacturing not being up to the job. It seems to be a case of how you adapt the generative design tools to be used with traditional manufacturing techniques. After all, if you use less material and still manage to reduce the number of tooling operations, the unit cost will still be reduced. Generative design is providing designers with new ways of looking at old problems.

New generative design tools are leading to productivity gains and higher-performing designs, and some technology only recently become available, so we haven’t yet seen many examples of commercial products. However, we believe customers are well underway in adding and benefiting from generative design in their product development workflows.”

What is it going to take for designers to migrate these designs from prototypes into actual large-scale production? There are technological hurdles to overcome first.

“The outputs of generative design tools, at first glance, can be unusual and non-intuitive to engineers accustomed to more traditional-looking design. But our customers are beginning to appreciate that the technology allows them to explorer huge amounts of practical, manufacturable solutions by simply specifying the engineering problem they aim to solve unlike traditional tools such as topology optimisation that merely enable incremental improvements to traditional designs.

We are observing that once engineers have performed comprehensive validation using traditional simulation tools, they build trust in the outputs of generative design. This growing trust along with an increasing rate of adoption and ever-decreasing costs of advanced manufacturing techniques means that generatively designed parts will soon be in production. We’ll also see an increase in production parts created generatively once our technology starts creating designs suited to traditional, lower-cost manufacturing processes.”

So, we are seeing some sort of consensus among the CAD people—lowered costs in manufacturing will enable the move from expensive prototypes into cheap mass production. And slowly but surely, engineers and consumers are becoming accustomed to the strange new forms that generative design allows.

Therefore, which industries will be the first to bring these generative-designed parts to market?

“The adoption of generative tool design is happening most quickly in industry segments where a premium is put on high-performance and uniqueness of designs. the most practical cost benefits are realised when multiple parts are consolidated into one unique and original component. Part consolidation simplifies the customers’ supply chain and reduces downstream assembly costs.”

Generally speaking, when we think of parts being consolidated like this, we tend to be speaking of parts manufactured with additive manufacturing. The printing of multiple parts at one time means a reduction in assembly time, inventory and fastenings. But is the growth in generative design driven exclusively by additive manufacturing? This was definitely the case in the beginning, this is no longer necessarily the case.

Generative design tools produce geometrically complex designs that are manufacturable and practical with additive manufacturing. “With every passing year, additive manufacturing is becoming more mainstream and accessible as costs come down and more materials are supported. And, as we expand to producing designs that are suited to processes like machining, casting, etc., we will enable more and more customers to mass produce their generatively designed products.

“Generative design is inherently about understanding requirements, from performance to materials to manufacturing processes, and providing engineers with a variety of practical, manufacturable options to make informed trade-offs between time to market, cost and performance. “Ultimately, our goal is to enable designers and engineers to produce more, better, with less—more new ideas, products that better meet the needs of users, in less time and with less negative impact on the environment. generative design will not be tied to one manufacturing method in the future.

That’s good news for traditional manufacturers. Generative design is not just for those with high-end metal printers, but for anybody with a CNC-enabled workshop and access to generative-designed tools.

“As soon as generative design starts creating designs that are suited to multiple manufacturing methods, we expect mainstream adoption to grow rapidly. “We also see large potential in industries where high-value, low-volume products are produced, , such as those in aerospace.

“What you’re seeing with 3D printing is they prefer to take traditional manufacturing out and put 3D printing in without changing anything around that, and this approach isn’t going to work. These new technologies will only be allowed to work if you allow the context around it to also adapt, and it’s not as simple as people want it to be sometimes. It may take a little bit of time.

There are multiple reasons for low production adoption working at the same time. One is the time frame we are working at with the development of designs and the construction of large aerospace products. If we have an idea now, you will only see that built on-site in say several years.

We work in a complex value chain where no one has complete control over the final results, so it is a lot of different people and a lot of different companies that have to move at the same time or at least following in a series, and what you see in these processes is that they all start with great ideas and throughout the years and the phases they become victim to planning pressure and costs so a lot of new ideas and challenges are just designed out.

“So, its not to do with 3D printing not manifesting itself—more it is to do with change being difficult to establish because change is difficult, change might be costly, and change has to always be better in every single way than what we have now—and that is something that is very difficult to foresee and to guarantee, so it’s little steps.”

Its less to do with how the parts look and more to do with how they perform and at what investment. For each application, we will evaluate cost and performance benefits as compared to a more traditional manufacturing process.

We may choose to invest more to 3D print a part if the added value in terms of performance merits the additional cost. If a traditional manufacturing process is less expensive than 3D printing and there is not a substantial performance benefit, then traditional manufacturing processes will be used.”

Great potential for generative design combined with additive manufacturing processes to enable part designs that are lightweight with the performance criteria we expect. Metal 3D printing costs are still relatively high in comparison to traditional manufacturing methods.

“Generative design can be used to design parts for some traditional processes. For example, casting processes could be used to manufacture some of the generatively designed parts.” Additive manufacturing is about to become even more critical to manufacturers in highly competitive industries thanks to generative design capabilities.

Generative design leverages tools based on previous design knowledge and high-performance computing to autonomously generate or modify design geometry based on requirements for or constraints on product performance.

Using generative design, engineers specify parameters, such as weight-to-strength ratios, efficient material use, and temperature, pressure and force ranges. The generative design engine creates sever design options through an iterative. Engineers then evaluate and select from among the generatively designed options—more options than would be humanly possible with traditional design tools, and likely many options that the engineers would never have considered.

Together, generative design and the additive manufacturing processes give engineers powerful tools that speed prototype, and finished part and tooling development and production; as well as parts with unique characteristics or functionality.

Both generative design and the 3D printing process make possible entirely new geometries and topologies, including more organic designs than ever could be realised with traditional manufacturing. Many shapes coming from generative design are difficult or impossible to produce using traditional methods, while additive manufacturing processes can directly print these shapes.

In addition, by pairing generative design with additive manufacturing, companies can build products with optimum functionality. For example, many parts for aerospace industry require materials that are heat resistant, which are expensive.

Additive manufacturing processes, when paired with generative design, will become even more critical as design and production processes compress to meet customer demands for faster delivery of more complex, often customised products at lower cost.

The power of generative design and additive manufacturing processes is compounded when adopted as part of an end-to-end product lifecycle management strategy that integrates design, engineering, manufacturing and maintenance through unified solutions and data.

Manufacturers must frequently and quickly deliver new and innovative designs of complex products, such as those in aerospace, industrial machinery and heavy equipment, and make sure you are moving quickly to adopt them.

Generative design both uses and adds to product knowledge in the digital thread, as simulation results are stored, ready to be applied to optimise future designs. Generative design, Convergent Modeling and additive manufacturing have become critical tools that support digital and team-driven workflow.

1. Increased Customer Satisfaction

Being able to generate multiple designs at a faster pace means more satisfied customers. Since the delivered designs are also of higher quality and meet all the customer’s requirements, it is a great way to increase customer loyalty and build an solid reputation.

2. Large amounts of Design Concepts at the Touch of a Button

Thanks to powerful artificial intelligence tools, designers no longer need to think up designs the old-fashioned way. For example, when designing a coffee mug for a fighter jet, the designer can input desired parameters such as the weight, material, and volume and the tool will deliver all possible designs that meet those criteria.

3. Rapid Approach to an Optimal Solution

Since more designs can be created within a shorter time frame, the optimal design solution can also be found quickly. This is because designers can compare and contrast all the different designs generated by the tool before selecting the best one.

4. Customised Constraints

Using the designer’s inputs and artificial intelligence, the latest design tools can produce highly customised build plans based on preset parameters. After the initial design is produced, engineers can then adjust the tool creating different designs that satisfy certain criteria such as the build size and cost.

5. Increased Productivity

An increase in productivity is to be expected as a result of the many design variants available at the touch of a button. Instead of taking precious time to come up with the various possibilities of a design, designers can use this time on other projects.

6. Consolidation of Multiple Parts

The ability to consolidate multiple parts into a single part is another benefit of generative design. This is because highly complex information can be processed at a rate that is not possible for workers. As a result, a single part can now be created to replace assemblies of two or more separate parts.

7. Decreased Manufacturing Costs

Due to the consolidation of multiple parts into a single part, decreased manufacturing costs can be expected because the supply chain will be simplified due to the elimination of unnecessary parts, reducing the overall manufacturing cost of the product.

8. Reduced Material Consumption

This is another benefit resulting from the consolidation of multiple parts. By creating models that require fewer parts, less materials are also required. This helps reduce material costs.

9. Avoids Expensive Manufacturing Rework

Manufacturing rework is a costly process that can reduce production output significantly. Rework requires extra time and energy to coordinate and complete. With the help of simulation and built-in testing functions, most rework can be eliminated and a final design can be reached within a shorter amount of time.

10. Reduced Weight

Reducing the weight of manufactured parts is another benefit and a real game changer for aerospace industries. In one recent case, engineers used generative design to produce a new bracket that combined many components into one, resulted in large weight reduction and strength increase compared to the original design.

1 Comment

Top 10 Considerations Review Component "Digital Twin" Designs In Product Engineering Develop Phases

1/1/2019

0 Comments

Digital Twin prototyping is great for exploring design possibilities. You can try several different prototypes, and even if they don’t work for you it’s a useful exercise. It enables you to visualise concepts you have not yet created and gives you new design perspectives.

Neither digital or physical methods will give you 100% accurate feedback because you’re not using the actual materials or processes that the end product will use. If that’s the case then which approach should designers choose? At early stages of a project digital is by far the better choice because it’s much faster and offers much more flexibility.

It makes sense to start the prototype process in digital work space. That means sketch and explore design concepts in 3D from the very beginning. It’s not the speed of carving a 3D digital model that is the advantage; it’s the amount of extra information you get for the effort.

Rapid prototyping allows engineers to use computer-aided design CAD to generate 2D and 3D models of parts or assemblies, and hit the production floor running. “These models can be assessed and tested through 3D visualisation and simulation, and once cleared the prototypes can be made by 3D printers and other machines using the original files for fast production.

The great thing about the digital workspace is you can explore sketch prototypes in 3D. This allows to visualise the real amount of space that is available, the true proportions that implementing a notion may put on the final form. Manipulating geometry in 3D space presents hard to see opportunities very early; before everyone has committed to the first reasonable idea.

As other new technology like VR and AR become more and more embedded into design practice too, the designer already working in a 3D work space is much better equipped to embrace that technology. Designers who can sketch and draw in 3D and then understand how the same work can be used to create videos, animations, VR and AR realisations of their work will be more and more in demand.

Digital Twins are learning digital models of physical assets, parts, processes and even systems. The purpose of the Digital Twins is to relay data about the performance and properties of a physical counterpart. With this information, Digital Twins will achieve complete repeatability of a 3D printed part, and greatly improve process reliability.
　
Consider launch of a new “Digital Twin” Process Simulation solution for predicting distortion during 3D printing. When metal parts are 3D printed and the layers build up, residual heat from fusing the layers can cause parts to warp inside the printer, causing various problems, from structural issues within the part itself to print stoppage. Simulation of the printing process can solve many of these problems.

The Process Simulation solution uses a Digital Twin to simulate the build process prior to printing, anticipating distortion within the printing process and automatically generating the corrected geometry to compensate for these distortions.

We tried many tools to get to the final iteration. You can see in the picture we show two design iterations—the middle one that is still quite similar visually to the original node was our first iteration. And you can see version 2.0, the second iteration is where we really made the big step with the freedom of form.

Using generative design, engineers specify parameters, such as weight-to-strength ratios, efficient material use, and temperature, pressure and force ranges. The generative design engine creates several design options through an iterative approach. Engineers then evaluate and select from among the generatively designed options—more options than would be possible with traditional design tools, and likely many options that the engineers would never have considered.

No one group has complete control over the final results, so it is a lot of different engineers have to move at the same time or at least following in a series, and what you see in these processes is that they all start with great concepts but throughout the phases they become victim to planning pressure and costs so a lot of new ideas and challenges are just designed out.

Now we have a digital representation of what the designer/customer wants, we have the actual part that we can touch and feel and also a Digital Twin of that actual part. In 3D printing we can only work in the digital world with a 3D digital model of the desired component. Now we can build the part, according to the 3D model, take that physical component and carry out our own 3D scan, creating yet another 3D model.

Digital Twin of the actual part can then be sent back to the designers and he can compare what we have manufactured to what his model wants, and even use the actual part model to simulate its impact, digitally, in the final design.

In the case of a 3D printer, we’re building a Digital Twin of a build process and recording the slightest defects, deviations and other build characteristics. With Digital Twins, models will continually be updated with each new build and become ever smarter in recognising and troubleshooting any potential issues that might arise.

Not only will there be a Digital Twin of the component, showing the internal and external requirements, but also a Digital Twin of the process that made that part; the process parameters, how long did the build take, how many layers were built, were there any issues.. all of these aspects building a digital picture of the part enabling further analysis and confidence in final applications of components.

Next generation of Digital Twins incorporate information from other sensors monitoring the 3D printing process, such as the shape of the pool of metal rendered molten by the laser. In addition, this smart, real-time quality control will not function in isolation.

The power of Digital Twins is their ability to share insights with each other. So you can imagine many 3D print machines sharing unique build insights with each other that makes them each more informed about what to watch for during a build process.

Through the Digital Twin process, you can accelerate the production of mission-critical equipment. Using Digital Twin technology, we’re aiming to rapidly speed up the time that parts could be re-engineered or newly created using 3D printing processes.

The key challenge with 3D printing is being able to additively build a part that mirrors the exact material composition and properties of the original part that was formed through subtractive measures. With operation of mission-critical parts there is no room for deviations in material performance or manufacturing error.

Properties and serviceability of 3D printed components are affected by their geometry, microstructure and defects. These important attributes are currently optimised by trial and error because the essential process variables can’t currently be selected from scientific principles.

A solution is to build and validate a Digital Twin of the 3D printing process capable of predicting of the spatial and temporal variations of metallurgical parameters affecting the structure and properties of components.

In principal, the Digital Twin of 3D printing process , when validated with accurate with experimental data would replace or reduce expensive, time consuming physical experiments with rapid inexpensive numerical experiments. In the initial phase, the Digital Twin would consider all the important 3D print process variables as input and provide a transient 3D model.
　
What Process Was Used To Make the Prototype?

Metal 3D printing and vacuum casting are great for making prototypes. Each process can be a little slow per part, but that’s not really an issue if you just want one or a few. But when scaling up to low-volume production for a new product launch, different manufacturing solutions should be considered. In the case of a vacuum cast prototype, for example, it makes sense to choose plastic injection molding for larger quantities.

Injection molding differs from vacuum casting in many important ways, both in how the process is done as well as the results obtained. First, it requires that a hard tool be made from aluminum or steel. And injection molding rewards close adherence to design rules that don’t come into play in vacuum casting. These can include the dimensions of ribs and bosses, the use of gussets, minimal wall thicknesses, draft angles, the positions of gates, runners, ejector pins and many other considerations.

So product developers must ensure that their plans account for the additional cost and time-to-market that a transition from one process to another will involve.

Will The Results Be The Same?

A single part made by CNC machining will not be identical to a counterpart made via pressure die casting, and the differences can be both cosmetic as well as mechanical.

Each process introduces its own set of variables that can affect quality. For instance, die cast parts must contend with porosity, and porosity might limit the part’s strength or affect the surface finish – which is not true with a CNC machined part made out of the same raw material.

Understanding this in advance will help a product developer to calculate costs, development lead times and engineering and testing protocols for the finished part.

What Material Was Used?

Another consideration is the choice of material. A prototype, by its nature, can be made of most anything. When ramping up from prototype to new product introduction, it’s best to avoid hard-to-find, expensive or hard-to-work-with materials. The cost and time constraints of production favor the use of materials that can be purchased readily in the commercial market in sufficient quantities without supply interruption and which can be processed efficiently.

This could compromise the look and feel of the final part compared to the prototype, but this can perhaps be compensated for by using alternate finishing techniques.

What Was The Finishing Process?

A prototype that was carefully sanded, polished and hand painted with a custom color no doubt looks great. But is that practical on a large scale? Elaborate finishes tend to require a lot of attention to detail and careful hand work. This is simply not practical for volume production unless the intention is to deliberately target high-end clients. For most products on the shelf, it’s necessary to find solutions that reduce hand work to a minimum.

Processes that can be automated are one way. Another possible solution is to stick to one kind of finish rather than multiple finishes, each of which would need to be handled separately and which would represent a much larger investment in time and money.

How Many Components in the Build?

As with complex hand work, it’s also best to limit the number of individual components to a minimum for volume production. Fewer parts require less assembly, for one thing, but also represent less of a risk for breakage, loss, out-of-tolerance specifications and potential variation from one production batch to another. For volume production, minimising every variable is the best way to both save money and maintain consistent quality.
　
Keep it Simple

Production places a premium on repeatability and consistency, minimising costs, automating processes, and using simplified materials that are easily available. The engineering and design skills needed to fulfill these criteria may be different than those used to come up with the initial great idea. You can contract out to offer advice and alternative solutions when you upload your CAD drawings for a free quotation.

Rapid prototyping helps companies turn ideas into realistic proofs of concept, advances these concepts to high-fidelity prototypes that look and work like final products, and guides products through a series of validation stages toward mass production.

Engineers and designers have been creating hardware prototypes for decades, but the tools, materials, and methods used to create those prototypes have made tremendous progress. With rapid prototyping tools like 3D printers, product development teams can create prototypes directly from CAD data, and quickly execute rounds of design revisions based on real-world testing and feedback at a substantially lower cost than ever before.

Prototyping with 3D printers, however, can be quite different than with working with other traditional tools or outsourcing to machine shops and service providers. Cost factors, efficiencies, and design rules often don’t directly translate.

In this guide, we collected ten insights to help you optimise your 3D printing rapid prototyping workflow to be as cost and time efficient as possible, from choosing a technology to practical design tips.

1. Prototype In-House

For any business involved in prototyping, one of the first questions that comes up is whether to order prototypes from service contractors and machines shops or to purchase equipment to prototype in-house.

Rapid prototyping stops being rapid when an outsourced part takes multiple days or even weeks to arrive. Outsourcing can quickly become expensive when a project requires dozens or more iterations. On the other hand, purchasing a variety of machinery to produce all the different parts in a single product often requires substantial investment, a dedicated location, and expertise to operate.

The answer is not always clear-cut, but the best practice for most companies is to bring the most frequently used prototyping tools in-house and outsource larger parts, and parts that require non-standard materials or complex machinery.

Smaller desktop or benchtop 3D printers can cover many of the prototyping needs for most companies. They’re fast, easy to use, can operate in small shops and require minimal training. Depending on the number of parts and printing volume, investment in a desktop 3D printer can break even within months and save weeks or months of lead time over the course of development.

2. Choose the Right Technology and Machinery

To find the right prototyping materials and equipment, first, consider what you need from your prototypes. Do you need prototypes for visual demonstration only or for testing the mechanical attributes of your product?

Understanding these needs will help you choose the right technology. For example, for basic concept models the only requirement could simply be speed—finish and details may not matter. Looks-like prototypes, however, may require technologies and materials designed for fine details and high-quality surface finishes, while functional prototypes might need to withstand mechanical stress or have specific properties, such as optical transparency.

3. Automate Post-Processing

Post-processing is an often overlooked, but potentially time-consuming aspect of prototyping with 3D printing. Some technologies require less post-processing than others, but all 3D printed parts require a certain degree of post-processing. Some aspects of post-processing can be automated to reduce labor time and costs.
.
4. Assemble Large Parts From Multiple Prints

3D printing large parts can be a costly and lengthy process, often requiring outsourcing parts to service providers with large industrial printers. But, just as assemblies consist of many individual building blocks, splitting a model into smaller parts is a great solution to creating objects larger than what fits into the build volume of a 3D printer.

You can add features to your design that will allow the prints to align themselves, or simply split the parts with straight cuts, requiring you to align them during the fastening process.

When selecting a bonding method, your primary consideration should be the strength of the bonded joints, which is dependent on the ultimate use case of the parts such ask a bonding agent for art, scale models, and complex shapes that are not meant for functional use and to sustain impact.

5. Make Parts Hollow

By default, most 3D printers create fully dense parts. When you’re not printing functional parts that require a certain strength, hollowing out large and bulky designs can be a great way to save a considerable amount of material and printing time.

6. Adjust Layer Height

Adjusting the layer height is a great way to reduce printing time. On some systems the difference between parts printed with 50 and 100 micron layers is often barely noticeable, but reduces printing time by 50%.

7. Optimise Schedule

There are a few methods for optimising your printing schedule to get the highest throughput possible, printing close to 24 hours a day. Best practices for optimising schedule include Batch multiple parts into one build, printing small, shorter runs during the day and large builds overnight, using multiple printers to distribute the workload and increase same-day throughput and using Dashboard to receive alerts when a print finishes and to manage and watch multiple printers remotely.

8. Reduce or Eliminate Support Structures

A poorly oriented part can result in excessive support structures. Excessive supports use more material, increase printing time, and require more post-processing time. Depending on your design, a part can often be printed with limited or without any support structures. Most print preparation tools allow you to experiment with different part orientations and check how different setups affect overall print time and material usage before printing.

A manifold prototype printed directly on the build platform with limited support structures for overhanging features. Some technologies might also be better suited to your designs than others. Some printers often require excessive support structures for designs with complex shapes, angles, and overhangs. Support structures on printers are easy to break away and support requirements can be reduced with smart tools. Some machines do not need support structures at all, as the powder acts as a support for the parts while printing.

9. Optimise the Design

While 3D printers offer a high degree of design freedom, a bit of time spent on optimising part geometries goes a long way to ensure efficient printing of high-quality parts. When designing a part for 3D printing, make sure to follow design guidelines for the specific technology or printer.

Common optimisations include: Maintaining wall thicknesses at or above minimum specifications, Eliminating or supporting angled walls and steep overhangs, Adding drain holes for hollow designs, and using lattice structures to achieve an ideal balance between part strength, material usage, and print speed.

10. Prevent Failures

Failed parts and broken machinery wastes expensive engineering time and can set development cycles back by days or even weeks. Fortunately, 3D printers have developed tremendously since the first desktop printers entered the market a decade ago and professional 3D printers today are tools that companies can rely on.

As a rule of thumb, you can reduce failures to statistical insignificance by following some simple rules: Work with reliable machines and companies that provide training and technical support, Keep your machine and workspace clean, Take the time to set up your prints properly, Print only with reliable, tested materials, Check the expiration date of materials before printing and Carry out regular service and maintenance as specified by the manufacturer.

0 Comments

Top 10 Virtual Reality Training Simulation Application to Field Activity Scenario Prototypes

12/11/2018

1 Comment

The future of Marine aviation is complex: aircraft are growing more technologically advanced, pilots face a proliferation of high-end and low-end threats, military budgets are squeezed and demand for Navy forces around the globe is growing.

So how will Marine aviation training keep up? In part, with fielding of tech advanced simulators.
Joint Terminal Attack Controllers using the simulator can coordinate with pilots in the air to identify and mark targets for air strikes from the ground

In a feat that combined live training and simulator training, we conducted a live, virtual and constructive demo. We took equipment that’s already on the aircraft that broadcasts the aircraft’s altitude, airspeed, position in real time, and we put a transmitter or receiver unit on the top of the building.

We were able to tap into that feed, and what that did was it took that feed of an actual aircraft on the range, and we piped it into simulator, and it was accurately recreated in the virtual workspace.

The aircraft is actually flying on the range and is properly displayed in the simulator with very low to minimal latency in a real-time altitude, airspeed, and attitude. So what that provided for is a real-time control of that aircraft with the ability to see the aircraft as well as have the ability to achieve visual recognition.

Marines are able to look up and actually assess the attitude and profile of that aircraft and then provide the clearance to essentially employ munitions on the desired intended target.

Before we had the simulator, we were really slow in the first few days on the range because that’s the first time operators did it. But now getting some practice time in, you get better control and better performance on the range with the live assets, so it makes it more efficient. So the simulator is really useful, it’s invaluable as far as getting Marines ready to go.

It has been harder and harder to get fleet aircraft that can support training due to a high operational tempo and due to challenges in keeping the aircraft ready to fly. The more training Marines can get on the range, the better they are when they actually get to an actual aircraft.

“So they’re not stumbling on Day 1, they’re already semi-proficient or trying to get there, whereas in the past before they had this simulator you’re a mess your first several times, so it’s good training for you, but for the guys airborne, they’re holding for a half hour just to get a bomb off because the guy on the ground is learning what to do.

The simulator has created a dramatic improvement in the first pass drop and the communications on the radio and everything. Marines work everything out here, so by the time that they’re on the range it’s just the real-life stuff that hits you. … A lot more first-pass drops, which is the whole goal of close-air support.

Want Marines will eventually be able to do is put this into a guy in a aircraft simulator and they’ll be running this simulator, talking to the guys in this simulator, and doing all their controls to get their currency requirements to satisfy their trainin while taking their targeting cues from other Marines in their own simulator.

A next step towards achieving that vision of connecting multiple simulators spread across the battlespace is the integrated training facility to house, all under one roof, simulators for pretty much anything in the carrier strike group.

We’ll be able to integrate them all together. Eventually we will be able to pipe in feeds from live aircraft out on our range – that’s the live part, and then vice versa hopefully we can pipe what’s being seen in the simulators, or what’s being constructed in the simulators, out to the live aircraft as well.

What we want to be able to do in the future, and this facility is the first step, is machine-to-machine data gathering. And that will allow us to gather large amounts of data – so not just necessarily how they did on that event, in the actual actions they took on that event, but we can also gather historical data on the aircraft, its system, how well the systems have held up.

We can look at, automatically, machine-to-machine, look at the pilot to assess proficiency, and see how much flight time was received recently, helping us build that bigger picture so we can inform leadership with the best information we can give them.

Despite the focus on high-end warfare technologies, aviators could face, equally dangerous less expensive threats like shoulder-launched anti-air missiles so we invested in Surface-to-Air Missile simulators to help ensure that pilots cycling through training events are aware of the threats they face on the ground and are flying with tactics that would keep them safe.

Though the SAM simulators aren’t connected to the planes in the air – so the pilot didn’t know in real-time he had been “shot” at with the simulator – the simulator logs video of the encounter. That video is incorporated into the pilot’s debrief after a training event, with the instructors explaining to the pilot whether his flight profile would have kept him safe or put him at risk to ground threats.

This is how you prove to Marines, you’re reachable, you need to be careful and you need to know what you’re doing, get your tactics right. Everything that’s out there is beatable, you’ve just got to know what you’re doing, but you’ve got to get your tactics right.

Readiness Tool allows top brass to determine which battalions and gear are most prepared for battle.

Marine Corps is experimenting with artificial intelligence to improve the way it deploys its forces and spot potential weaknesses years in advance.

The Marines built a tool that crunches data on personnel and equipment to measure how prepared individual battalions are for combat. The tool could ultimately help top brass deploy some 186,000 active-duty Marines and countless pieces of military hardware.

Allocating the service’s resources is an imperfect science. Leaders map out deployment strategies years or even decades in advance, but situations will invariably arise that throw a wrench in those plans.

Planners are constantly forced to “reshuffle the deck” as crises flare up in different places and figuring out which units to move around is a complicated process. Numerous factors—training, deployment history, equipment readiness and others—affect how prepared a group is for a given situation.

Today planners rely on spreadsheets, whiteboards and basic applications to track readiness and manage forces, but artificial intelligence can offer them a better understanding of the resources at their disposal and the long-term effects of the decisions they make.

The tech crunches both structured and unstructured data from multiple force management applications to create a real-time image of how prepared each unit is for combat. The tool specifically aims to build a five-year management plan for the Marine infantry battalions.

Tool has two primary functions: It flags the units that are most ready for action and explains why others come up short. Armed with that knowledge, commanders can proactively train and invest in less prepared groups before they fall even further behind.
　
“A lot of times Mairnes only invest more when the problem arises. Now they can see it ahead of time and say ‘OK, we’re going to take action now to prevent that from occurring.’”

The tool sheds light on how deployment decisions will affect forces in the long run. By analyzing historical trends along with real-time data, the tool could show how a unit’s readiness would change if it were, for instance, moved to a new location or given additional resources.

Marines are also building a separate AI system that ranks course of action plans based on those extrapolations, which could one day be merged with the readiness system.

“You integrate that all together and you get a full view of readiness across your force. Now you can really make some data-driven decisions.”

The next stage of the effort will include parts of the Marines’ aviation and logistics units, bringing about half branch into the purview of the program. With that additional data, the AI would further refine its processing rules to deliver better results.

So artificial intelligence is tasked with managing the particular deployments of troops in battle, moving them around in new and unexpected ways.

One way that future might manifest is by looking at a place where AI already manages workforce inventory-- like a warehouse stocking system, a process where items are unloaded wherever there is space in a warehouse and then scanned into a computer system than can track where the item is located.

When it comes time to retrieve an item for delivery, the same computer system directs warehouse workers to the most efficient route for finding the item, which could be stowed throughout the warehouse.

When modeling the warehouse system, it is interesting to consider how AI, given the same objectives as a commander, might organise and direct forces to achieve them.

“Why would an AI allocate forces in distinct areas of the battlefield? It could intermingle them and manage them at a granular level. Its categories are way more numerous, in the way that a warehouse AI manages categories at the shelf level.

Instead of distinct groupings of armor, air support, infantry, and artillery, a system run by artificial intelligence and managing a battle could coordinate a single helicopter with a pair of howitzers and an infantry platoon, directly grouping each in the same way that a warehouse worker finds an assortment of items to place into the same package.

“Anytime we’re on the road, our job, maintenance wise, is to provide safe and reliable jets for the pilots to accomplish their mission. Every new location presents a different challenge in how we get the job done, but the end goal for providing a safe jet for a pilot never changes. What does change is the environment in which we operate in.”

“Every exercise you go on is different, and it can be hard to start off. It could be not having the parts we need on hand, or not knowing how the base operates to get the support we need. Over time you figure out how to acquire some of that on site, what to bring along yourself and how to solve a problem before it becomes one.”

Here we consider how AI systems could be useful to a typical work order job of launch and recovery of aircraft, engine maintenance and servicing of life-saving equipment-- just a few of dozens of tasks Troops are expected to accomplish within a full day.

“We learn to operate in new environments, out here we’ve adapted our operations to give the best support possible. Maintenance is maintenance, our job never changes, but how we execute the mission does.”

“Our main mission is to enable successful sorties by generating aircraft parts, ultimately maintaining our full spectrum readiness. Our team encounters new repairs that force changes in direction and orders, but they all adapt and constantly find ways to make sure the job gets done.”

Maintaining the aging aircraft can be challenging as some parts are no longer commercially produced and the Fabrication Flight must collaborate and innovate to construct parts on their own.

“We all need each other in order to complete a task and make sure operations are done correctly. “Everything revolves in a circle – sheet metals technicians hand over parts to metals technicians who follow their technical order before sending to nondestructive inspection to make sure the piece is good for use on an aircraft.”

To display the teamwork necessary, the Troops walked us through the Fabrication Flight process.

Sheet metals technicians , kick off operations by receiving technical orders for aircraft repairs. Troops survey the technical order and pulls a thin, malleable sheet from their collection. The sheet is then cut to the specific measurements and handed off to a metals technician to be heat treated in a large oven.

"On our side we handle breaking the metal down and then crafting it to match the technical order for the specific part. When completed, the piece is hauled over to nondestructive inspection where tests are conducted to ensure the part is compositionally sound and aircraft ready.

"With the resources we have here, we are the final stop on a part's journey to an aircraft,” "If anything is wrong with the part, it's flagged and sent back to the workshop to either correct the issue, or start the operations all over again."

Accuracy in fabrication is essential in getting aircraft back up flying. When the part has completed all processes and is cleared for use, it is installed onto the aircraft, restoring readiness of the aircraft.

Fabrication flight Airmen gain a sense of accomplishment by witnessing their work come to fruition each time an aircraft takes off.

“Having combatant commands and other mission partners on base only adds to the importance of mission success. We take pride in the work of the flight, seeing the aircraft out there completing missions thanks to the maintenance here is an amazing feeling.”

By creating a virtual representation of an asset in the field using lightweight model “Digital Twin” visualisation, and then capturing info from smart sensors embedded in the asset, you can gain a complete picture of real-world performance and operating conditions. You can also simulate real-world scenario conditions for predictive operations.

Advances in virtual prototyping spaces has made possible the capability to simulating visual fidelity to a very high level. The next big challenge for virtual prototyping teams is simulating realistic interaction. Virtual prototyping, sometimes referred to as digital prototyping, is widely adopted by industry to simulate visual appearance and functionalities of production.

But conventional virtual prototyping techniques lack the simulation of the physical properties of a real interaction between user and product. Force feedback is based of development of virtual prototyping.

Virtual Reality tech creates an alternative reality in which worlds, objects and characters can be experienced that may not yet be available in reality so stakeholders are allowed to not only see the future product- achieved with concept sketch or mockup, but also experience the product and the interactions with its use context.

Simulation models as used in virtual engineering during development of training systems can be used during operation phases as well. In order to fully benefit from this, the simulation model must be connected to the physical system and other business operations In this way, information regarding past operation and current status can be fused with information regarding possible future operation, explored through virtual scenarios.

The overall result can be used for decision support in for instance operational planning or service and maintenance. In this way, simulation serves as a tool for arriving at a situation in which the future scenarios are perhaps not completely known, but in which one can readily address the most likely scenarios in an adequate manner.

Artificial intelligence can play a role in virtual manufacturing by improving simulation models or by offering better decision support. Extending the use of simulation models from the design phase to the operation phase also has advantages when new products are to be introduced or system needs to be reconfigured.

Virtual Reality is an attractive option since it offers the user a sense of being immersed in information where objects have a sense of ‘presence’ and allows them to interface with information at full scale if required. A design begins with an image or idea and the concept is disseminated via diagrams and descriptive speech.

Typically, information sources for conducting various virtual reality activities are not one single specific source, but instead all the different tech training information systems that are used in DoD The integration of these sources is not usually out-of-the-box-solution but most often highly customised solutions, engineered by specialists.
　
"Digital Twins" Provide Line of Sight to 3D Print Part Builds Previously Not Visible.
Digital Twins are learning digital models of physical assets, parts, processes and even systems. The purpose of the Digital Twins is to relay data about the performance and properties of a physical counterpart. With this information, Digital Twins will achieve complete repeatability of a 3D printed part, and greatly improve process reliability.

Now we have a digital representation of what the designer/customer wants, we have the actual part that we can touch and feel and also a Digital Twin of that actual part. In 3D printing we can only work in the digital world with a 3D digital model of the desired component. Now we can build the part, according to the 3D model, take that physical component and carry out our own 3D scan, creating yet another 3D model.

Digital Twin of the actual part can then be sent back to the designers and he can compare what we have manufactured to what his model wants, and even use the actual part model to simulate its impact, digitally, in the final design.

In the case of a 3D printer, we’re building a Digital Twin of a build process and recording the slightest defects, deviations and other build characteristics. With Digital Twins, models will continually be updated with each new build and become ever smarter in recognising and troubleshooting any potential issues that might arise.

Not only will there be a Digital Twin of the component, showing the internal and external requirements, but also a Digital Twin of the process that made that part; the process parameters, how long did the build take, how many layers were built, were there any issues.. all of these aspects building a digital picture of the part enabling further analysis and confidence in final applications of components.

Next generation of Digital Twins incorporate information from other sensors monitoring the 3D printing process, such as the shape of the pool of metal rendered molten by the laser. In addition, this smart, real-time quality control will not function in isolation.

The power of Digital Twins is their ability to share insights with each other. So you can imagine many 3D print machines sharing unique build insights with each other that makes them each more informed about what to watch for during a build process.

Through the Digital Twin process, you can accelerate the production of mission-critical equipment. Using Digital Twin technology, we’re aiming to rapidly speed up the time that parts could be re-engineered or newly created using 3D printing processes.

The key challenge with 3D printing is being able to additively build a part that mirrors the exact material composition and properties of the original part that was formed through subtractive measures. With operation of mission-critical parts there is no room for deviations in material performance or manufacturing error.

Properties and serviceability of 3D printed components are affected by their geometry, microstructure and defects. These important attributes are currently optimised by trial and error because the essential process variables can’t currently be selected from scientific principles.

A solution is to build and validate a Digital Twin of the 3D printing process capable of predicting of the spatial and temporal variations of physical parameters affecting the structure and properties of components.

In principle, the Digital Twin of 3D printing process , when validated with accurate with experimental data would replace or reduce expensive, time consuming physical experiments with rapid inexpensive numerical experiments. In the initial phase, the Digital Twin would consider all the important 3D print process variables as input and provide a transient 3D model.

1. Systems design: Design before you build with a visual, simulation approach.

2. Asset-based system of system design: Specify, publish, find, and reuse organisation simulation systems,

3. Product-line engineering: Design product platforms and variants quickly and efficiently, and make better trade-off decisions.

4. Systems model review: Improve product quality and model consistency through early design reviews within a systems modeling tool.

5. Systems model simulation: Validate complex behaviour earlier in the design life cycle, and establish predefined standards and best practice–based process.

6. Establish an open, flexible simulation system: Such a system is necessary to incorporate information sets from multiple engineering domains and quality control

7. Align combat engineering teams for better collaboration: Disconnected combat engineering teams across mechanical and electrical systems working in their own workgroups must collaborate as needed-- utility of systems-level view of products must be evaluated

8. Balance vitality and stability: Balancing vitality of innovation with reuse and predictive stability during establishment of an innovation platform for simulation and during product design and engineering.

9. Unify simulation connected systems optimisation: A single view of cross-domain system, product, and process is required for successful simulations

10. Incorporate quality with design and development: Achieving high level of product quality is why simulation virtually validates systems-level view. Assuring Incorporate/embed quality information from the early-stage design through subsequent product phases is key so simulations can more easily flow from system designs into product attributes.
　

1 Comment

Top 50 "Digital Twin" Virtual Prototype Questions Combine Connected Product/Network Platform

12/11/2018

0 Comments

The powerful utility of Digital Twins in Prototyping enables you to answer many fundamental questions without any physical prototype or testing. Everything is digital. So you are getting smarter about the operation of this product without spending lots of money to build anything physical.

With respect to developing products linked to networks, we’re in the early stages. But soon, some companies are going to want to sidestep development problems they’re experiencing. They’re going to realise that building and testing five rounds of prototypes is unacceptable. They’re going to realise months of delay completely undermines their competitive position in the marketplace.

As a result, those companies are going to want to adopt more proven and standardised practices. Virtually prototyping with Digital Twins is not there yet. But given the rush of companies toward network solutions, you should expect the demand for this practice to only increase.

The concept of a Digital Twin is mostly applied to the case where more insight is obtained from a physical operating product and a network platform. But in the case of virtual prototyping, the Digital Twin concept is applied to an old practice in mechanical hardware.

One approach to developing a smart connected product, one that hooks up with a network platform, is to just build it. Just piece it together. Throw some sensors on a product. Wire that to some kind of embedded system. Wire that to your antenna. Start sending data to an network platform. You and your organisation can actually learn a lot from going through that exercise.

While that needs to be done, you will quickly run into limitations on the experiments you can conduct with physical prototypes. Swapping out a sensor isn’t easy when it’s soldered in place. There might not be room, physically, for the sensor you really need for accurate measurement. You might run into too much electromagnetic interference for the antenna you planned to use.

Working through these issues is new to some organisations as they transition traditionally mechanical products to smart connected ones. However, the problems associated with resolving issues through physical prototyping isn’t new. In fact, that is an old concept when it comes to hardware. Long ago, mechanical and electrical engineers figured out that modeling and simulating a design virtually means you are more likely to get it right the first time when you get to prototyping and testing.

The benefits of an approach utilising virtual prototyping with digital twins are many. You have fewer rounds of prototyping, saving money and time. You have fewer change orders. You stay on schedule. You stay on budget.

So while virtual prototyping is new to some organisations, this approach has advantages when applied to the development of linked smart, connected products and networks platforms. Digital Twins are a key enabler.

How exactly can Digital Twins be used to virtually prototype smart, connected products and network platforms? You first need to set up the digital model component of a Digital Twin with one of the following:

Numerical Models: These models use machine learning and artificial intelligence tools. These applications or agents either extrapolate that data and/or correlate data to existing events. Both are an effort to predict future behaviour.

1D Simulation: These models are a combination of flow diagrams with equations or formulas behind the blocks that simulate the performance of embedded tools or multi-disciplinary engineering systems. These models can provide deeper insights into ongoing operation

3D Simulation: These models, often in the form of multi-body dynamics, are commonly used to predict the dynamics and structural performance of products. These models can provide deeper insights into ongoing operations.

For scenarios based on engineering physics or asset operation, no prototype or operating product exists. As such, there is no sensor data to feed this digital model. However, the model can be fed historical sensor data from prior products or even from past physical tests or operational data. In the worst case, a set of inputs can be modeled using statistics or even a higher level simulation, such as a multi-body dynamics model. This creates a set of input data that can be fed to the digital model.

The combination of that digital model and the input is enough to get started. You run the model as a simulation, generating data from virtual sensors, which are points of measurements from the simulation.

In this application, that virtually generated data is used instead of physically recorded data from sensors. That output can then be fed to a network platform as if it were receiving streaming data from a running product. Only, in this case, there is no physical product. There is only a virtual product that is running in a simulation.

In this scenario, you overcome many issues that you might experience when trying to physically prototype a smart, connected product.

You can change anything related to the sensor configuration, including placement or type.

You are not limited by network bandwidth other than the limitation between the compute resource running the Digital Twin and the one running the network platform.

You can change the product design in terms of mechanical or electrical hardware, embedded systems and more.
There is a tremendous amount of flexibility with this approach.

So now you have this concept of using a Digital Twin to virtually prototype a smart, connected product that is linked to a network platform. What does that get you? Interestingly, it allows organisations to answer a set of serious questions.
　
1. Is the systems configuration right for this product?

2. Do we need to use a physical sensor to capture this data, or can it be a virtual sensor?

3. Is the data we want to flow to the network platform limited by bandwidth?

4. Is edge processing required for the sensor data?

5. What data should be processed on the product versus being fed to the network platform?

6. Are there changes that should be considered to improve placement of sensors?

7. Are there changes that should be made to avoid interference?

8. How will connected product and network platform work to fulfill requirements?

9. What conclusions can be drawn from the data once it is in the network platform?

10. What data trends are precursors to events critical to the connected product?

Top 10 Operating Mechanisms of “Digital Twin” Architecture Framework Represent Flow Between Each Builder Module and Sub-System

To ensure digital networking of production systems and the optimisation of material-specific requirements, we need to measure, assess and replicate the changes in material properties in a process where "Digital Twins" of materials are created.

The materials digital space has laid the groundwork for this process. When a finished part rolls off the production line, this is one of the first questions always asked: "Does this component have the properties we want?"

Often, even the tiniest of variations in the production environment are enough to alter a part’s material properties – and throw its functionality into question.

Manufacturers avoid this by close inspection of samples throughout the production process. Breaking down the samples into their composite parts and measuring them separately is an extremely time-consuming process.

"The outcome of the sample testing process branches out into an array of different subsets, each with their own specific measurement results. While experts may be able to keep an overview of the complex interrelationships in their heads, until now there has been no way to take the diversity of resulting data and portray it in a coherent digital format."

Now, for the first time, a proof of concept has been developed demonstrating that it is possible to digitally represent many such material processing cycles with a materials data space for test specimens produced using additive manufacturing.

"The data space concept allows us to integrate any type of material information into a digital network – a really valuable tool. We want to use the materials data space to automatically generate a digital twin of each material that will mirror the current state of the physical object under examination."

Data spaces can be used to integrate all types of materials information into digital networks. The advantage of the materials data space is that it provides an overview of all relevant parameters at a glance, whereas formerly data on different material parameters was scattered among numerous data repositories in many different formats.

But the real promise lies in the future. "In the years to come, the materials data space has the potential to become the production command center. Whenever component quality isn’t up to the expected standard, you can compare it with information on previous components stored in the materials data space to determine whether the present component can in fact be used or whether it must be rejected.

In the future, these results could be automatically integrated into industrial decision-making processes: whenever component quality dips below the required standard, production automatically comes to a halt.

Creating the data space –and managing the diversity of materials data – calls for a corresponding information model. "In this case, the model reflects the natural material world, in which material states and properties are assigned to defined categories.

The best way of thinking about it is in terms of a social network where each user is a node in the network. And in turn, these nodes have their own subject matter associations. What we do is to create semantic relationships between the individual material objects and their associated processing steps.

Then there are also interrelationships among these communities. What would be a “follow” on social media is represented in the materials data space by details on the chronological sequence of production or work steps, for instance "leaving the additive manufacturing process" or "this laser is part of the 3D printing process".

The new demonstrator for additively manufactured metal components has the capacity to generate samples, characterize the materials they contain, conduct subsequent data analysis and determine material properties.

Thanks to the logic underpinning the model, users can make extremely complex queries of the data space that simply wouldn’t be possible with the same degree of flexibility in the case of a conventional database.
　
1. Creation of quality prediction models: Collected sensor/process network data crunched with AI techniques to build quality prediction models

2. Storage of quality prediction/response models: Response plans/times are entered into model repository according to operator experience

3. Creation of production schedules: After production simulation schedules include machine/production target outputs specified to timetable and sent to reference coordinator builder to lead into quality/productivity detector

4. Request for reference performance indicators: Coordinator delivers the production schedules to performance indicator simulation sub-system to be used in production-monitoring standards

5. Transmission of reference performance indicators: Reference models created based on simulation results serves as the criteria for manufacturing execution monitoring criteria to include target quality and production per unit time per process

6. Quality analysis and prediction: Quality and productivity detector predicts the quality and productivity in real time using quality prediction model and reference performance indicator

7. Request for analysis of response plans: Coordinator requests model repository for analyses of response plans for irregularities and receives corresponding information

8. Request for future performance indicators: Predict future performance indicators caused by responses to irregularities and current schedule sent to the performance indicator simulation sub-system.

9. Request for new schedules: Coordinator requests new schedule when difference between change in initial/future performance due to irregularities is significant

10. Transmission of visualisation information: Progress of entire systems is simultaneously sent to dashboard in the form of information visible to the user through graphical user interface.
　
Top 10 “Digital Twin” Engineering Team Strategies Define Agent-Machine Interaction Focus Configuration Programme Goals

1. Formalise planning development, integration, and use of models to inform enterprise, programme decision making, support engineering activities to digitally represent the system of interest

2. Ensure models are accurate, complete, usable across disciplines to support communication, collaboration and performance and decision making across lifecycle activities

3. Provide enduring, authoritative source of secure authentication with access/controls to establish technical baseline, product digital artifacts, and support reviews for accurate decision making

4. Incorporate technological innovation to establish end-to-end digital engineering enterprise and foster conditions for productive step advances towards goals

5. Enable end-to-end decision making using advance human-machine interactions

6. Establish mature supporting digital engineering activity infrastructure to perform activities with connected information networks

7. Develop, mature, and implement technology tools to realise digital engineering goals and share best practices using models to collaborate with stakeholders

8. Improve digital engineering knowledge base, policy, guidance, specifications, and standards and streamline contracting, procurement and business operations

9. Lead and support digital engineering transformation efforts, vision, strategy, and implementation to establish accountability to measure and demonstrate results across programmes

10. Build and prepare workforce to develop knowledge, competence, and skills with active participation and engagement in planning and implementing
　
Top 10 Observations of Day-to-Day Work/Field Experiences Identify Trends/Challenges of “Digital Twin” Simulation Utilisation

Specific sessions with engineers, technical/simulation managers, R&D, quality managers were performed. Surveys were conducted in a typical inductive approach assess key practices, processes, tools and data associated with simulation and product/process development.

1. Domains/application: depth and completeness of engineering simulation areas

2. Methods: engineering simulation tools utilization

3. Level of integration within driver/follower processes

4. Process gates and decision criteria: definition, completeness, visibility

5. Documentation of simulation process and decision-making criteria/milestones

6. Level of adoption/dissemination of engineering simulation in extended organisation

7. Depth/completeness of specific skills of engineering simulation

8. Organisation: relationship and integration between engineering simulation teams and the rest of product development teams

9. Data lifecycle/workflow: modelling, capture, revision, access/control

10. Infrastructure: central/distributed computational capacity, support/availability, post-process remote/local capabilities

Top 10 “Digital Twin” Implementations

1. 3D model of a generic part/product

2. 3D model of a specific part/product

3. 3D model shows live information

4. Live set of physical database assets

5. Process simulation validate manufacture potential

6. Human Machine Interface process plant

7. Live icon representation e.g. display indicator

8. Live numerical value graphic

9. Simulation of a machine e.g. rotating

10. Simulation responding to real time data

0 Comments

Top 10 Structure Assessment Tool Requirements Provide Evaluation of Status Update Framework

12/1/2018

0 Comments

Current readiness systems only include commander’s best estimate for equipment status. Estimates have traditionally been utilised usually for overall equipment assigned to the unit and not individual pieces of equipment.

Military Services use systems to maintain records of equipment under service, but records do not include any information about what units it is assigned to.

Central to the work presented here was the development of a tool, the Marine Air-Ground Task Force Equipment Structural Assessment, which was loosely based on previously developed plan

Inputs to the system consist of a MEU equipment list, the tasks identified through the mission deconstruction process, the measures and metrics used to define equipment capabilities, and the set of linkages between tasks and equipment.

What equipment is available to the MEU to accomplish mission tasks and subtasks? A diverse set of factors affect the types of equipment aboard a MEU, including not only space available but also risk trade-offs made by commanders and expectations about the mission on deployment.

Since there is no standardised table of equipment for a MEU, the study team obtained a list of equipment assigned to a recent MEU, which included information on what was embarked and what was left behind.

What measures and metrics should be used to assess the capability of selected equipment? The loading list provided the set of available equipment. We then used equipment manuals and sponsor input to define the capabilities of each piece of equipment in performing designated tasks. This information is displayed to the user when a piece of equipment is selected

We identified the measures and metrics, or “planning factors,” needed to assess the capability of each piece of equipment in the loading list. In our initial construction of the tool, we identified which alternative equipment might accomplish a task, but not as effectively.

We concluded that either we had it wrong or that other equipment might do just as well or better than what we selected. Consequently, equipment selection is now up to the user of the tool. The upgrades proceeded in parallel with these activities.

We have not received official Marine Corps approval to run our application at Job Sites. The programme is a valuable materiel readiness information tool designed and Tested for Marine Corps.

Users training on existing systems, even though we have designed logical architecture and operation modes, we have found concerning lack of training on logistics information tools throughout Marine Corps.

Tool supports this objective by asking the user to define mission-specific characteristics and allowing the user to tailor equipment lists, equipment priority, and task priority as appropriate.

The approach used in this report is for the user to use the tool to facilitate the development of planning factors by the user and use the tool to assign equipment to tasks.

This process provides a framework that MEU commanders can use to develop mission plans and understand where equipment shortfalls are likely. The process consists of simple steps that translate mission requirements into tasks, subtasks, and military activities, each of which is linked directly to the types of equipment needed for completion. It also highlights key parameters that may affect the types of equipment needed or the execution of key tasks.

Site Visit Executive can first look at broad readiness, but can also look at readiness levels of subordinate units to provide the ability to control, distribute, and replenish equipment and supplies in assigned areas of operation, to receive supply support from and provide supply support to other services

Readiness Terms are used in different contexts/processes. Operational gaps in systems used by Marine Units must be closed so exchange is seamless. Capability to link information as it is processed by Units must be built.

Aggregated information provided to Commanders must be traced/linked to operational systems used to rollup information. But no Marine Site Executive has yet stood up to identify functions spanning across process and write terms required to support processes

If Site Visit Executive has better overview of equipment status, resources will be allocated/pooled more efficiently so greatest potential for operational readiness is realised.

Information from readiness systems is required to determine number of pieces of equipment available for deployment. No Site Executive has created an easy way to link equipment information available from readiness and Services systems.

Technological advances in production and distribution can strengthen the Navy and Marine Corps aviation parts supply chain. Improved spare parts logistics systems and 3D printing will increase flight availabilities and decrease costs. 3D printing is the headline of how far we’ve come with efficiencies, both at the fleet readiness centers and out in the field.

The entire spare part logistics system has the potential be sped up with the use of 3D printing. With forward deployed forces, addition of 3D printing increases availability and save costs by quickly producing small replacement parts onsite instead of waiting for the supply chain to send equipment far off.

In addition, 3D printing as a way for the industry to quickly manufacture the parts needed by aircraft maintainers without necessarily having to sink money into new machinery to make specialised components not frequently requested.

Ultimately, this on-demand manufacturing will help companies control their costs. The only limiting factor is the ability for 3D printers to create air-worthy parts. “We’re at the front end of this. There are parts that require airworthiness for approval and the non-air worthiness, the non-airworthiness are easier to do.“

Maintenance portion of an aircraft program is of equal importance as new acquisitions in keeping costs down. “We got to operate it, and sustain it, and fly it for the lifecycle. So understanding your supply chain and making sure it’s robust is key.”

A new logistics sustainment system Marine maintainers are trying will help both the service and industrial base adjust their ability to purchase and manufacture replacement parts. The new system prioritises how to allocate replacement parts to aircraft based on how quickly it will return to service after the part arrives.

Consider the fate of two aircraft from different squadrons. Both are grounded, and each requires the same replacement part, but one of the aircraft needs additional other work done to get back in the air.

Under the current system, the part goes to the maintainers who request it first, even if this aircraft needs additional work resulting in being grounded for weeks. Meanwhile, the aircraft that only required the one part could’ve been ready sooner, but remains unavailable while waiting for part delivery.

“We’re now using supply optimisation tools that are taking a look across a base, and not only a base but across a type, model series. As an example, a long lead-time part is coming in, so what airplane benefits most from that? That’s one area where we’re using agent learning to make decision making.”

We got the chance to learn about the latest technologies showing potential for Marine Corps deployment. We are covering this expo as a team who has never been to one, so here is our perspective on the experience.

Even though we had never been to a Marine expo before, we had a pretty good idea of just what it will entail. We were expecting multiple companies to be set up in theatre showing off their latest tech advancements and best manoeuvre practices.

These expos seem like a great way to learn about Marines and connect with one another for possible future partnerships or work. In a nut shell, we believe this advanced expo is going to house a lot of tech and experienced Marines and we should be able to learn a lot from it.

When we walked through the doors and entered this expo, it seemed like the number of Marine suppliers was infinite. Everywhere you looked there was something new to look at and learn about. The first supplier that grabbed my eye was a growing company from who had designed a sorting system for small parts.

The group realised their designs potential for larger parts and how it could be applied to multiple different industries. The group upscaled their design and went from one machine designed for extremely small parts, to an array of multiple machines each having the ability to sort different sized part much larger than the first design. It was incredible to see how one design could be changed only slightly and have so many different applications for Marines.

There were no limitations when it came to suppliers and the number of different AORs present at the expo. There were solutions for warehouse storage, automation, sorting, milling bits, and even 3D printing. One of our favorite booth we visited was run by a company who had all the 3D printing solutions you could ever need, even for new areas of the field.

These suppliers were displaying their new desktop metal 3D printer and its ability to print in metal.3D printing in metal was something we had heard about before, but imagined it had only been done by a very small number of Marine companies.

Right there in front of us this printer was creating quite incredible parts all in metal. We were told these prints had similar strength qualities as cast metal parts and were printed using a type of metal powder mixed with wax.
This was incredible to hear if you had only ever experienced 3D printed plastic parts before. Looking around the booth it was easy to see all the Marines gears turning as they were able to see all the applications this desktop metal printer could be used for their own field operations.

Going into this expo we didn’t have much knowledge about Marine manoeuvres. After speaking with suppliers showcasing part sorting/packing systems, we were surprised to learn a significant amount about sensor sorting systems. These machines simply take images of the sorting bed and use tools to tell which items to grab or re-sort.

We saw how easy 3D printing in metal and its benefits compared to plastic. With this machine being able to create custom metal parts that no other manufacturing process can create, it is easy to see the endless possibilities and applications for this process.

An expo is a great place to satisfy our curiosity. With all kinds of new technology and processes you can check out all the booths for hours finding the answers to all your pressing questions about the Marines.

These expos don’t need to be just for Marines. We were welcome to learn a little more about the processes and machines used to make products. Overall, going to this advanced expo was a great experience and we will be attending many more in the future.

Even while the groups and organisations hyping artificial intelligence solutions popped up everywhere at the expos with promises to create the next battlefield advantage using next generation weapons, gear, or satellites. The term artificial intelligence splashes the headlines with promises that we’re moments away from revolutionising the battlefield.

It’s frustrating. The special AI “task forces” and their massive budgets are great, but it’s time to get honest about the rest of the military.

Ask any Marine their opinion of how things run on a daily basis and you will hear complaints about lost orders, broken gear, and outdated technology.

Bottom line: all those flashy AI applications being touted as perfect for Marines use are not going to run on the outdated infrastructure on which a majority of the military still operates.

That doesn’t mean that AI isn’t a good fit or shouldn’t be pursued. But it does mean that AI success requires a force readiness approach. First, AI isn’t new and it isn’t new to the military.

Marketing hype around the term has experienced a surge lately but the fact that something wasn’t tagged as artificial intelligence historically does not take away the fact that it was actually AI.

Despite the hype, AI is simply a field of science that trains systems to perform some human task through learning and automation. There are varying degrees of sophistication but most of the mining, network assessment constructs and mapping technology used over the past decade or more have all been forms of AI.

Weapons systems and combat vehicles have been leveraging AI for many years as well. So don’t let the noise change the focus from the mission need.

There are varying degrees of sophistication but most of the data mining, network engineering and mapping technology used over the past decade or more have all been forms of AI. Weapons systems and combat vehicles have been leveraging AI for many years as well.

Marines on the front lines need their supporting forces to be trained and armed with the appropriate technology to support the advances being operationalised on the battlefield. If we look specifically at the intelligence arena, the vast majority of military intelligence analysts are still using the same products and systems from 10-15 years ago.

Efforts around collecting intelligence are ripe for sophistication, but what about the Marines that have to sift through and make sense of that additional data? How has their training changed to account for a more technologically advanced battlespace? How do products and solutions integrated requirements and workflows with real time information truly augment their efforts?

The majority of data mining and visualisation tools on the market have flashier interfaces than we saw a decade ago, but the true sophistication of what the vast majority of Marines have been offered doesn’t really reflect the decade of advancements seen in the commercial market.

You don’t have to be a part of a high profile AI initiative to find value in the science for nearly all areas of the military. We need the whole force to have the technical advantage on the battlefield and that means AI must become a force readiness initiative.

It’s all about augmenting human efforts across battalions, regiments and divisions to raise the readiness levels of the entire force. Marines inside the wire should have the knowledge, technical skills and agility to support all of the operations and technology our troops outside the wire are running.

Then there’s the applicability across all military systems.

An “AI watchman” could prevent ships from colliding with one another since the computers are “constantly looking at sensor data and is making sense of the environment and the situation.”

“There is that safety aspect of using artificial intelligence to augment the level of capability and intelligence available on ships, on tanks, in aircraft, all over, where you almost have an embedded AI technician be part of every military asset.

“That is a capability and it leads to benefits that are tremendous. And the possibilities may be endless. There’s an easy answer to the question: Where can AI be applied? “It can be applied literally everywhere.

“The biggest part of the problem of artificial intelligence is: they build these incredibly long algorithms with all of these gates to go through. They push all of this machine learning and data through it. Frankly, we are not entirely sure how all of that works, all the time.

1. Use language processing and machine learning to automatically classify and match incoming data to indicators and warnings being monitored. Provide alerts on trending topics, keywords or themes that may indicate emerging tactics, techniques and procedures.

2. Display both geographic and temporal representation of multi-intelligence data with a natural language generated summary of the data. Include the ability to break data into individual entities as needed and internalise analyst annotations into the automated summary.

3. Automatically map finished intelligence products to the priority intelligence requirement to help answer with automated caveat classification of documents tied to user permissions. Include smart search capabilities to that repository so analysts can find relevant products more efficiently.

4. Monitor human developed courses of action beside computer generated courses of action including the criteria for suitability, feasibility, acceptability, uniqueness and completeness. A machine will see information differently than its human counterparts and may identify behaviour discrepancies present in data that human analysts may miss due to the sheer volume and complexity of reporting that an intelligence analyst is presented with.

5. Capture workflows and product development in a shared space so knowledge gaps are reduced between shifts or rotations. Use automation to track knowledge gaps and alert users to update analysis and finished products when significant knowledge gaps are filled to tag/map intelligence gaps as new information comes in and alert users to the new information.

6. Measure impact of operational intelligence and associated collections or requests that contributed to that intelligence by automating inputs and processes serving as operational measurements.

7. Add cognitive search into the massive data repositories analysts are required to sift through to move beyond keyword search and enable contextual search at an enterprise level.

8. Provide in-depth training on AI systems and set standards on how technology augments the human analytic process without replacing the analyst behind the screen. In short, tie the technology into existing workflows and adjust workflows to account for technological innovation.

9. Conduct initiatives in parallel with operations to ensure force efforts are complimentary and requirements are aligned with discrepancy alerts or gaps in the operational plan and the intelligence needed to execute it.

10. Invest in both garrison and tactical systems and infrastructure that are capable of running and sustaining the increased computing power that comes with training and deploying AI programmes

0 Comments

Top 10 Blockchain Rules Consider Work Order Consensus Agent Cooperation Bucket Brigade Routes

12/1/2018

0 Comments

We set up experiment involving Digital Twin” robots learning to work together: one robot ideally handing off items to the other, which in turn carries them to a final destination.

Bucket brigade systems are a tool to build robust simulated robot control systems. This choice is sufficient to achieve adequate levels of performance for a variety of behaviours. The parallel implementation of the bucket brigade system would speed up the training process and implement robotics controller.

Bucket brigade systems provide guidance shortening the number of cycles required to learn task rules using only a few training examples starting with classifiers that were randomly generated.

The Remote Access Nondestructive Evaluation system is a snake-like robotic arm tool that fits into small spaces of an aircraft to perform inspections. Maintainers usually have to remove whatever hard-to-reach component and crawl inside the small area.

The robotic arm can manoeuvre through access ports with small as diameter and serve as an agents eyes as it moves around inside the aircraft, saving time and eliminating the need to take the aircraft apart. The robot is currently a prototype and is ready for the programme office to request exact specifications for use on certain aircraft.

New Platform Creates Autonomous Robots Create Map of Workspace Using Sensors so Users Choose/Teach Route

Platform integrated into commercial service robots so autonomous navigation stack doesn’t have to be built from scratch. This is interesting stuff that is potentially game-changing in terms of the cost able to command volume pricing on sensors that we can’t. And we are engineering things to fit together nicely.

Getting tools right is especially challenging for robots that will be autonomously navigating in complex scenarios such as airstrips and other high-traffic workspaces. These areas often have tight spaces and continuously changing obstacles that require complex routes. The challenge is creating tool to handle these issues with the end-user in mind.

The robot must involve minimal training for operators, no battle space setup, single-shot learning by demonstration, and productivity reporting.

Variables impacting autonomous navigation are not limited to physical obstacles crowding a robot’s work space. Feature-less work space, and even time of day, add complexity to autonomous navigation. Many of these types of hurdles are edge cases that do not present themselves until after tools have been developed and the robots are tested in a live mission space. Edge cases are the punch you don’t see coming.

For example, being able to navigate in a cluttered, dynamic workspace with a lot of troops moving around is difficult because there’s a lack of features. There’s nothing, really, to anchor or tie into when you’re building your map.”

Success is contingent upon getting your robot, and the tools that runs it, into many different workspaces early on. Functional autonomous navigation systems are not developed in a lab. It’s fine to begin development there to create a demo, or to get funding. But those stages are the limit for lab testing.

It won’t work until you’ve been in a number of scenarios because the problems that you’re going to experience in theatre-specific deployments cannot be replicated in a lab. You can’t solve or anticipate every edge case your robot will encounter in the real world. Keeping the troops involved in the installation process and giving them the tools to troubleshoot issues in real-time can improve your robot’s efficiency.

Examples include when robot mistaking light from a reflective surface as a physical object or infrared heaters disrupting the robot’s path. Essentially, the more edge cases you can solve, the better your navigation solution.

The key to designing a robot with autonomous navigation is creating a system that has precise and accurate motion control.

“For a lot of robots, you’re just taking a robot from point A to point B, but with field applications like equipment distribution machines, they need to drive as close to an edge, as close to a wall, as close to an obstacle as possible to maximise the floor coverage provided.”

Highly accurate motion control is imperative if you want your robot to be able to handle complex, tight spaces. That’s something you can’t do with a robot that has much larger footprint. Designing a system that is as tight and accurate as possible give you much better capabilities to navigate complex spaces.

Detection by actual troops is crucial for expanding end-user applications. If your robot can’t tell a set of troops from a package on the floor, you’ve hamstrung your unit manoeuvre before it starts.

“If you’re developing your own tools, if you are looking for navigation systems to use in your robotics project, then having a system that can recognise Troops as different to obstacles is essential.

Unless you’re going to clear everybody out in the space the robot works in, which limits the applications, you really need to solve the workforce element. This is one of the biggest problems you need to solve.

But being over cautious also has its problems. In running some initial pilot tests with the equipment distribution machine, the robots were checking, pausing, and analyzing for the sake of safety so often it made Troops less comfortable around them. People thought the robots weren’t intelligent, thus making them feel uncomfortable around the robots. It’s critical that you use sensor data from real-world scenarios and virtual space to reduce false positives.

For your product to be scalable, the installation process must be simple, not technical. Many of today’s robots require an engineer for installation into new space since process is simply beyond the skillset of non-technical staff. This in-depth and technically complex launch can bottleneck this critical early stage; having an engineer sent on-site to every new Troops customer is not sustainable or scalable.

One way to counteract this challenge is to have your customers identify workforce who might be capable of taking on installation as a new project. Another is to do some preventative maintenance regarding design. You want your robot to be familiar to products your customers have used before. Make sure the user interface is lean and intuitive.

“We wanted to keep it as simple as possible… As you can see in this screenshot right here, there are only two choices for the user: Choose a route or teach a route. The user can use the machine the way they always have and while they are doing that, it creates a map of the space and records the routes, or they set to play.”

“Trusting” Robotic System to Make Quality Parts Opens Door to Build Usable Parts When and Where You are Working
Consider sustainment and how a maintainer can print a replacement part at sea, or a mechanic print a replacement part for a truck deep in the desert. This takes 3-D printing to the next, big step of deployment.

We are exploring how machine learning and artificial intelligence can make complex 3D printing more reliable and save hours of tedious post-production inspections.

In modern factories, 3D printing parts requires persistent monitoring by specialists to ensure intricate parts are produced without impurities and imperfections that can compromise the integrity of the part overall. To improve this labor intensive process, we are developing multi-axis robots that use lasers to deposit material and oversee the printing of parts.

Initial work will focus on developing computer models that can predict the microstructures and mechanical properties of 3D printed materials to generate simulation data to train with, looking at variables such as, the spot size of the laser beam, the rate of feed of the titanium wire and the total amount energy density input into the material while it is being manufactured.

This information helps the team predict the microstructure, or organisational structure of a material on a very small scale, that influences the physical properties of the additive manufactured part.

Information will be plugged into a model that predicts the mechanical properties of the printed component. By taking temperature and spot size measurements, the team can ensure they accurately controlling energy density, the power of both the laser and the hot wire that goes into the process.

All of that is happening before you actually try to do any kind of machine learning or artificial networks with the robot itself. That’s just to try to train the models to the point where we have confidence in the models.

One key problem could come in cleaning up the data and removing excess noise from the measurements. Thermal measurements are pretty easy and not data intensive, but when you start looking at optical measurements you can collect just an enormous amount of data that is difficult to manage.

We want to learn how shrink the size of that dataset without sacrificing key parameters, compressing and manipulating this data to extract the key information needed to train the algorithms.

Robots will begin producing 3D titanium parts and learn how to reliably construct geometrically and structurally sound parts. This portion of the program will confront challenges from the additive manufacturing and AI components of the project.

On the additive manufacturing side, the team will work with new manufacturing process, trying to understand exactly what the primary, secondary and tertiary interactions are between all those different process parameters.

As you are building the part depending on the geometric complexity, now those interactions change based on the path the robot has to take to manufacture that part. One of the biggest challenges is going to be to understand exactly which of those parameters are the primary, which are the tertiary and to what level of control we need to be able to manipulate or control those process parameters in order to generate the confidence in the parts that we want.

At the same time, AI machine learning challenges need to be tackled Like with other AI programs, it’s crucial the communication interface is learning the right information, the right way. The models will give the communication interface a good starting point, but this will be an iterative process that depends on the communication interface ability to self-correct.

At some point, there are some inaccuracies that could come into that model so now, the system itself has to understand it may be getting into a regime that is not going to produce the mechanical properties or microstructures that you want, and be able to self-correct to make certain that instead of going into that regime it goes into a regime that produces the geometric part that you want.

With a complete communication interface that can be trusted to produce structurally sound 3D printed parts, time-consuming post-production inspections will become a thing of the past.

Instead of nondestructive inspections and evaluations, if you have enough control on the process, enough in situ measurements, enough models to show that that process and the robot performed exactly as you thought it would, and produced a part that you know what its capabilities are going to be, you can immediately deploy that part.

That’s the end game, that’s what we’re trying to get to, is to build the quality into the part instead of inspecting it in afterwards.

Confidence in 3D printed parts could have dramatic consequences for soldiers are across the services. As opposed to waiting for replacement parts, service members could readily search a database of components, find the part they need and have a replacement they can trust in hours rather than days or weeks.

Advanced Weapons System Sustainment Team Strategies Utilise New Workflow Tools to Improve Tracking of Product Requirements

While sustainment strategies do not guarantee successful outcomes, they serve as a tool to guide operations as well as support planning and implementation of activities through the life-cycle of the aircraft. Specifically, at a high-level the strategy is aimed at integrating requirements, product support elements, funding, and risk management to provide oversight of the aircraft.

For example, these sustainment strategies can be documented in a life-cycle sustainment plan, postproduction support plan, or an in-service support plan, among other types of documented strategies.

Additionally, program officials stated aircraft sustainment strategies are an important management tool for the sustainment of the aircraft by documenting requirements that are known by all stakeholders, including good practices identified in sustaining each aircraft.

Agents are integrating time-saving applications into their own workflow with proven concepts across commands and services. These time-saving tools enable simultaneous access to data such as requirements, program notes and contractor documents.

You can move from a system where you’re waiting for status updates with comments to come back from an engineer or contractor to logging into a system where you can see what they’re working on, in real time, so team can work on a project without any lag time.

Acquisition rules require contractors, program managers, engineers and other stakeholders to work together but it usually involves dozens of technical documents to move from contractor to DoD on any given day, creating a nightmare for acquisition officials charged with ensuring each document is reviewed and meets expectations

Programs find themselves overwhelmed by the sheer volume of data, putting the program at risk, if data spends too long in the review state, missing the contractual response deadline.

Without a vector check from DoD, the contractor assumes a conditional acceptance and moves forward, assuming they are on the right track. This could lead to a host of problems, including a possible schedule slip.”

The team is now spreading the proven results of tools and showing individual operational units can use the tools to suit various purposes. Versatility is inherent in the review and content-sharing applications, so new users usually recognize ways to streamline the bulk of documents that shuttle through an acquisition enterprise.

The objective is simple: information dominance through the creation, review, approval and dissemination of data. If you have unrefined, makeshift processes, this won’t work for you, at least not yet. Business processes work more efficient with automation so there are now options if your workflow is functional, but uses outdated tools or requires intense use of workforce capital.

“The advantage of the utilities lies in the ability to collaborate on existing work, review past work and evolve the system architecture to meet changing needs. These tools provide value now, and value in the future by giving agents access to program work performed. Every agent in the “Blockchain” can see, through comment tracking, how data changed to capture the correct info and meet requirements.

This fidelity, coupled with the capacity to short-circuit future process complications, makes work order content-sharing applications ideal for long-running, complex weapons systems sustainment program offices looking to limit volume of work spent on acquisition schedule.

The parallel bucket-brigade content for sustainment execution, operational manoeuvre etc. communication interface is evaluated in the implementation of space-time parallel applications for massively parallel machines. It is shown that the simplified version of in-time contract content work order rules is valid in time-dependent problems, and that it can be implemented as the form of the bucket-brigade communication with simple computations.

Must evaluate performance of several configurations by the use of interface for the bucket-brigade communications. On the basis of robot/agent for sustainment business operation performance measurements or any other practical applications, effective strategies for the further tuning toward large scale computing should be discussed

There is way to parallelise operations dynamically on a “Blockchain” while maintaining both decentralisation and security. Looking on the internet, no one spelled out this design in particular. It might actually help with scaling.

In contrast to just breaking up the whole system into smaller parts like you do in a bucket brigade, this design provides a much higher adaptability to the current utilisation of the communication network, however, does not improve the storage consumption on each node like breaking up into small parts does.

There’s no reason not to implement this strategy on top of current practise for increasing flexibility and weapons system sustainment work order transaction throughput. Another factor to consider is that many of the top “Blockchain” projects are still utilising proof of work. The process of changing agent consensus rules can be rather difficult for any project and requires some time.

We show that collaboration is achieved only when robots are rewarded based on a non-discounted global reward averaged over time, concluding with work in agent modeling under communication.

We have used communication protocol for agents to subcontract subtasks to other agents. In this approach, each agent tries to decompose tasks into simpler subtasks and broadcasts announcements about the subtasks to all other agents in order to find “contractors” who can solve them more easily.

When agents are capable of learning about other agents’ task-solving abilities, communication is reduced from broadcasting to everyone to communicating exact messages to only those agents that have high probabilities to win the bids for those tasks.

A related approach is presented where learning is used to incrementally update models of other agents to reduce communication load by anticipating their future actions based on their previous ones.

Case-based learning has also been used to develop successful joint plans based on one’s historical expectations of other agents’ action. Multi-agent learning is a new field and its open research issues are still very much in development. Here, we single out several issues we observed recurring while surveying past implementation efforts.

We believe these specific areas have proven themselves important open questions to tackle in order to make multi-agent learning more broadly successful as a technique.

These issues arise from multi-agent learning, and may eventually require new learning methods special to multiple agents, as opposed to the more conventional single-agent learning methods of case-based learning, reinforcement learning, traditional developments computation now common in the field.

Shaping, layered learning, and fitness switching, are not multi-agent learning techniques, but they have often been applied in such a context.

Less work has been done on formal methods for decomposing tasks and behaviours into appropriate for multi-agent solutions, how agents’ sub-behaviours interact, and how /when the learning of these sub-behaviours may be constructed in parallel.

Consider robot soccer as an example: while it is true that agents must learn to acquire a ball and to kick it before they can learn to pass the ball, their counterparts must also have learned to receive the ball, and to ramp up difficulty, opponents may simultaneously/co-adaptively learn to intercept the ball.

Not much attention has been paid to examine how to form these “decomposition dependency graphs”, much less have the learning system develop them automatically.

Yet to construct the learning process in parallel to simplify the search space, and reduce more robust multi-agent behaviours, understanding these interactions is important. One notable exception occurs that in many domains the actions of some agents may be independent.

Predator-Prey pursuit is one of the most common work spaces in multi-agent learning research, and it is easy to implement. Pursuit games consist of a number of predator agents cooperatively chasing a prey. Individual predator agents are usually not faster than the prey, and often agents can sense the prey only if it is closeby. Therefore, the agents need to actively cooperate in order to successfully capture the prey.

The goal of our leveled consensus “sustainment contracting” mechanism is to allow some fexibility as in the case with no commitment while guaranteeing agents some level of security as in the case of full commitment.

Full commitment contracts can be viewed as one end of a spectrum where commitment-free contracts are at the other end. Leveled commitment contracts span this entire spectrum based on how the decommiting penalties are chosen. Leveled commitment is desirable because it speeds up the negotiation process by increasing parallelism.

An agent can make mutually exclusive low-commitment offers to multiple agents. In the case more than one accepts, the agent can backtrack from all but one so agent can address the other parties in parallel instead of addressing them one at a time and blocking to wait for an answer before addressing the next.

For example, if an agent wants one particular contract, it can offer that contract to several parties with meaningful commitment instead of no commitment at all that would be strategically meaningless. Load balancing is crucial for parallel applications since it is representative of good use of the capacity of the parallel processing units.

Here we look at applications putting high demand on the parallel interconnect in terms of throughput. Examples of such applications are compression applications which both process important amounts of data and require a lot of computations undesirable for most parallel architectures. The problem is exacerbated when working with heterogenous parallel hardware. This is the case when using a heterogenous cluster to execute parallel application

Reinforcement consists of redistributing bids made between subsequently chosen rules. The bid of each winner at each time-step is placed in a "bucket". A record is kept of the winners on the previous time step and they each receive an equal share of the contents of the current bucket; fitness is shared amongst activated rules. If a reward is received from the work space communications then this is paid to the winning rule which produced the last output. Each rule is much like the middleman in Bucket Brigade Blockchain.

1. Work forward and continue to pick units for your job on the forward line

2. When you exchange work with your successor, then work backward

3. If you are the last worker when you reach the end of the forward line transfer your job to the backward line and work backward;

4. If you catch up with your successor, who is crossing the aisle, then wait.

5. Work backward and continue to pick units for your job on the backward line

6. When you exchange work with your predecessor, then work forward

7. If you are the first worker to complete your job at the end of the backward line initiate a new job and work forward

8. If you catch up with your predecessor, who is crossing the aisle, then wait.

9. If you are on the forward line, remain idle until your successor finishes crossing the aisle, then work forward

10. If you are on the backward line, remain idle until your predecessor finishes crossing the aisle, then work backward.
　

0 Comments

Top 10 "Digital Twin" Framework Objects Close Engineering/Simulation Gap Introduce Structure Configuration

11/22/2018

0 Comments

Common means of describing complex systems using object orientation represents specification narratives, but there are no details given on how/when features are linked together during progress made in systems creation. The main concerns about using object-oriented narratives for real-time embedded systems is about speed/size characteristics of builder to be utilised.

Some points in support of object-oriented narratives for embedded systems include requirements for objects to be efficient so Site Visit Executive can write about larger systems with fewer defects. Obtaining results in less time is realised using Digital Twin simulation techniques instead of structured methods, and advances can be implemented in assembly narratives, in addition to others.

An integrated engineering network, spanning across the entire value chain, is operated to intelligently connect various service divisions, and to generate a work space for products/services. The conditions for the Digital Twin are determined in which the digital space can be fed into the real, and the real world back into the digital to deal such intelligent products with rising variations.

Digital twins allow you to access large amounts of data in real time. But you don’t have to keep all that data to yourself. In fact, you’d be wise to share it. Creating a digital twin network makes it easy to share data with internal workforce, external supply chain partners, and even customers. With access to the same insight, you, your partners, and your customers can collaborate to improve products and processes.

Supply chain partners benefit from a network of Digital Twins with enhanced visibility. If an asset malfunctions, your maintenance provider knows it needs to mobilise a team to fix the equipment. If your company manufactures a product ahead of schedule, your logistics provider knows it can pick up the goods and deliver them early.

Digital Twin networks help you get invaluable insight from your customers. By monitoring how customers interact with your goods, you can remove underused features from future product iterations or develop new products that highlight popular features. Enabling an open, collaborative workspace through a network of digital twins offers you the chance to transform engineering, operations, and everything else in between.

With a Digital Twin network you share with your customer, you can monitor the condition of your asset around the clock and accurately track how much your customer consumes. This reliable and transparent method ensures you’re always standing by to repair the asset. Thinking outside the box and exploring innovative as-a-service business models is a surefire way to remain profitable in today’s ever-evolving digital world.

The term Digital Twin can be described as a digital copy of a real factory, machine, worker etc., that is created and can be independently expanded, automatically updated as well as being widely available in real time. Every real product and production site is permanently accompanied by a Digital Twin. First prototypes of Digital Twins already exist in Logistics Learning programmes built on a multidimensional data and information model.

A standardised language of the robot control systems via agents and positioning systems has to be integrated. The aspect of the continuity of the real workshop in the digital factory as an efficient means of ensuring continuous actuality of digital models can function as the basis for change

For localisation sensor combinations that in addition to the hardware already contain the application required for the sensor data fusion should be used. Processing systems, scenario-live-simulations and digital shop floor management results in a mandatory procedural combination. Essential to the Digital Twin is the ability to consistently provide all subsystems with the latest state of all required information/methods.

A Digital Twin is intended to be a digital replica of physical assets, processes, or systems, in other words, a model. It is most often referenced as an outcome of networks where the expanding world of devices with sensors provides an equally fast expanding body of data about those devices that can be broken down and assessed for efficiency,, design, maintenance, and many other factors.

Since the data continues to flow, the Digital Twin model can be continuously updated and ‘learn’ in near real time any change that may occur. Digital twins can produce value without machine learning and AI if the system is simple. If for example there are limited variables and linear relations are easily discovered between inputs and outputs then no data science may be required.

However, the vast majority of target systems have multiple variables and multiple streams of data and do require science discipline to make sense of what’s going on. Even while many experts tend to equate all this with AI, great majority of the benefit of modeling can be achieved with traditional machine learning tools to discover patterns in sensor readings.

For example, video feeds of components during manufacture can already be used to detect defective items and reject them. Similarly audio inputs of large generators can carry signals of impending malfunctions like vibration even before traditional sensors can detect the problem.

At first, an asset may be operating as expected. Inside the machine, however, it’s another story. A glitch in the system is causing your asset to gradually slow down. Later. it’ll fail completely. Without the right technology, you’d never know that. But Digital Twins help you anticipate issues and prevent problems before they even occur. They enable you to detect anomalies and automate repair processes at the first sign of weakness. And by coming to your asset’s rescue sooner rather than later, you can avoid serious service interruption or prolonged downtime.

If you already operate with advanced networks, especially those connected to industrial machines and processes you are probably in the clean up spot for Digital Twins. But any predictive model is potentially subject to drift over time and needs to be maintained. For example, some sensors are notoriously noisy and as you start to isolate signal from noise your sensors will undoubtedly need to be updated.

Although the definition of Digital Twins often includes specific reference to ‘processes’, examples of processes modeled with Digital Twins other than mechanical factory processes are difficult to find. Since many don’t have complex or capital intensive machinery and industrial processes, what is the role of Digital Twins in ordinary business processes.

There are applications available today that can automatically detect the beginning and end points of each step in the transaction from network logs thus providing the same sort of data stream for service origination as sensors might for aircraft.

The more that human activity is included in the data of what is being modeled, the less accurate the model will be. So for those of you who have modeled machine-based or factory-process based data where very little human intervention occurs can regularly achieve accuracy

But if we are modeling a business process such as customer-views-to-order in service centers then the complexity of operator action will mean the best models may be of limited complexity.

Digital Twins drive innovation and performance to give operators predictive tools and they give companies the ability to improve customer experience. Digital Twins allow for better understanding of customer needs, enhanced existing products, streamlined operations, and improved service-after-sales; all while creating headway for new products and services.

But even with industrial applications the error rate still exists. Models have error rates. So for example, when you use Digital Twin models to predict preventive maintenance or equipment failure, in some percentage of cases the maintenance is performed too early and in some there is the inability to forsee an unexpected failure. The model can continually be improved as new data and techniques are available but it will always be a model, not a one-to-one identity with reality.

Digital Twins can be used as a representation of current reality and new machines, processes, or components are designed and built up from scratch using those assumptions about operating reality. So you have to be sure to understand how the error rate in the underlying model might mislead designers into serious errors about how the newly designed machine or process might perform in the current reality.

The great majority of your interaction with digital systems is still request driven so once a condition is observed you instruct or request the system to take action. This is being rapidly supplanted by event driven processing. The modeling of machines, systems, and processes is a precondition for the optimisation work that determines when specifications and decisions are needed. As the Digital Twin movement expands, more streaming applications will be enabled with automated event driven decision making.

Object-Oriented Programming Simplifies Digital Twins

The digital twin model offers a breakthrough approach to structuring state tsream processing applications. This model organises key network information about each data source in application components that tracks data source changing state and to interpret the state and generating real-time feedback.

Using digital twins offers three key benefits over more traditional, pipelined stream-processing techniques: automatic event correlation by data source, deeper dives with enhanced state information, and parallel assessments to discover aggregate trends for all data sources in real time. It represents a big step forward for building stream-processing applications.

When using the digital twin model, each data source in a physical system has a corresponding object in the stream-processing platform that encapsulates both state information and code. State information includes a time-ordered list of the device’s incoming event messages along with key state information about the dynamic state of the data source. This information could include parameters, service history, known issues, and much more.

Application code handles the management of event list and the real-time analysis of incoming events for performing device commands. This code benefits from the rich context provided by dynamic state information, enabling deeper introspection than analyzing the event stream alone.

The secret to keeping event assessments low when handling events from many data sources is to host these Digital Twin objects in memory data grid with an integrated compute engine minimise network bottlenecks by assessing events within the grid.

Object-oriented storage precisely fits the requirements for Digital Twin objects, making it straightforward to deploy and host these objects with both scalable performance and high availability. The grid transparently distributes the Digital Twin objects across a cluster of networks for scalable processing.

Let’s take a look at how object-oriented techniques can simplify the design of digital twins. Because a digital twin encapsulates state information and associated code, it can be represented as a user defined type/class within an object-oriented language.

The use of an object class to represent the controller conveniently encapsulates the data and code as a single unit and allows for creation of many instances of this type to manage different devices. For example, consider the Digital Twin for a basic controller with class properties status /event collection describing the controller status and class methods for assessing events and performing device commands.

You can also can make use of the class definition to construct various special purpose digital twins as subclasses, taking advantage of the object-oriented technique called inheritance. For example, we can define the Digital Twin for a hot water valve as a subclass of a basic controller that adds new properties, such as temperature and flow rate, with associated methods for managing them.

This subclass inherits all of the properties of a basic controller while adding new capabilities to manage specialised controller types. Using this object-oriented approach maximises code reuse and saves development time.

You can build a group of Digital Twins that represent successively higher levels of control for complex systems to leverage object-oriented techniques. Consider the following set of interconnected Digital Twin instances used in managing a pump room:

In this example, the pump room has Digital Twin partners connected directly to devices, one for a hot water valve and another for a circuit breaker. These twins are both implemented as subclasses of a basic controller and add properties and methods specific to their devices. They feed telemetry to a higher-level Digital Twin instance which manages overall operations for the pump room.

This Digital Twin also can be implemented as a subclass of a basic controller even though it is not connected directly to a device. What’s important to observe about this example is how object inheritance and group rank play separate roles in defining the Digital Twin objects which work together to assess event streams. The behaviour of Digital Twin models to customise actions and build systems of interconnected Digital Twins customised to process events at successively higher levels of virtual expression.

Digital Twin models for state stream-processing have developed from concepts largely unrelated to object-oriented programming, in particular, product life cycle management and industrial networks device twins. Object-oriented techniques developers powerful tools for applying Digital Twins to break down state stream-processing and streaming processes.

Understand Digital Twins object models and spatial intelligence graph

Digital Twins services powers comprehensive virtual representations of physical work space and associated devices, sensors, and work force. It improves development by organising domain-specific concepts into useful models. The models are then situated within a spatial intelligence graph to model the relationships and interactions between workforce spaces and devices.

Digital Twins object models describe domain-specific concepts, categories, and properties. Models are predefined by users who want to match the solution to operational requirements. Together, these predefined Digital Twins object models describe/customise field-level regions/zones of interactions. With Digital Twins object models in place, you can populate a spatial graph.

Spatial graphs are virtual representations of the many relationships between spaces, devices relevant in network solutions, bringing together spaces, devices, sensors, and users. Each is linked together in a way that models the real world. For example, for workstations with many different areas users are associated with their workstations and given access to portions of the graph.

Spatial intelligence graph

Spatial graph is group graph of spaces, devices, and workforce defined in the Digital Twins object model. The spatial graph supports inheritance, filtering, traversing, scalability, and extensibility so you can manage and interact with your spatial graph.

If you deploy a Digital Twins service in your subscription, you become administrator of the root node. You're then automatically granted full access to the entire structure. You can provision spaces in the graph by using sensors. Open source tools also are available to provision the graph in bulk.

Graph inheritance applies to the permissions and properties that descend from a parent node to all nodes beneath it. For example, when a role is assigned to a user on a given node, the user has that role permissions to the given node and every node below it. Each property key and extended type defined for a given node is inherited by all the nodes beneath that node.

Graph filtering is used to narrow down request results. You can filter by identifiers, name, types, subtypes, parent space, and associated spaces. You also can filter by sensor data types, property keys and values

Graph traversing means you can move to new locations in the spatial graph through its depth and breadth. For depth, traverse the graph top-down or bottom-up by using the parameters You can traverse the graph to get sibling nodes directly attached to a parent space or one of its descendants for breadth. When you query an object, you can get all related objects that have relationships to that object.

Digital Twins guarantee Graph scalability so it can handle your real-world workloads. Digital Twins can be used to represent large portfolios of infrastructure, devices, sensors, telemetry, and more.

Finally, Digital Twins can be customised by utilising Graph extensibility to customise the underlying Digital Twins object models with new types/groups. Your Digital Twins data also can be enriched with extensible properties and values. The following Digital Twins Models Support Object Categories

1. Spaces are virtual or physical locations

2. Devices are virtual or physical pieces of equipment

3. Sensors are objects that detect events

4. Resources are attached to a space represent resources to be used by objects in the spatial graph.

5. Property keys/values are custom characteristics of spaces, devices, sensors, used along with built-in characteristics,

6. Roles are sets of permissions assigned to users and devices in the spatial graph

7. Role assignments are the association between a role and an object in the spatial graph. For example, a user or a service principal can be granted permission to manage a space in the spatial graph.

8. User-defined functions allow customised sensor processing within the spatial graph to: Set a sensor value, Perform custom logic based on sensor readings, Set the output to a space, Send notifications when predefined conditions are met.

9. Matchers are objects that determine which user-defined functions are executed for a given telemetry message.

10. Endpoints are the locations where Digital Twins events can be routed, for example, Event Hub, Service Bus, and Event Grid.
　

0 Comments

<<Previous

Top 10 Product, Production and Performance "Digital Twin" Sets Consist of Multiple Virtual Models

Top 10 Stages in Internal Design Team Build of Sketch Models/Prototypes Create Run Samples

Top 10 Generative Design Benefits Using Digital Simulation Tool Creates Optimal Outputs

Top 10 Considerations Review Component "Digital Twin" Designs In Product Engineering Develop Phases

Top 50 Instructions to Rebuild Readiness Provide Regular Input on Status of Recovery Issues

Top 10 Virtual Reality Training Simulation Application to Field Activity Scenario Prototypes

Top 50 "Digital Twin" Virtual Prototype Questions Combine Connected Product/Network Platform

Top 10 Structure Assessment Tool Requirements Provide Evaluation of Status Update Framework

Top 10 Blockchain Rules Consider Work Order Consensus Agent Cooperation Bucket Brigade Routes

Top 10 "Digital Twin" Framework Objects Close Engineering/Simulation Gap Introduce Structure Configuration

Site Visit Executive

Archives

Categories