|
How VLM Works VEXTEC predicts the life of manufactured parts by simulating the behavior of the materials they’re made from. We also know that the materials we use to build complex components and systems are not homogenous. If they were, one could take seemingly identical components off the same assembly line and test them under the exact same conditions, and they would fail in exactly the same way at exactly the same moment. But the likelihood of that is exactly zero. Those supposedly identical components are not identical at all. Material scientists have known this for decades, but there’s never been a way to account for that variability as it relates to durability forecasting – until now. For the first time, VEXTEC’s Virtual Life Management technology provides a computational framework that accurately accounts for the variability in the material combined with all the various damage mechanisms reacting to the stress imposed on a product as an ongoing process throughout the products lifetime. The result is a Virtual Twin – a 3D, time and condition-based product simulation that accurately reflects the real world physics of how, when and why damage occurs.
Manufacturing production creates a variety of material microstructure complexities within each product coming off the assembly line. Products are flown, driven, pushed, pulled, heated, cooled, or exercised by the customer in any variety or combination of ways. Usage in this manner imposes force or stress into the product which is absorbed throughout its material makeup. Computational software like Finite Element Analysis (FEA) is used to predict how this energy is distributed in unequal patterns of high and low stress. It’s well known that not all product failures occur at the high stress areas; however, up to now industry has had little computational means to be able to quantify exactly where failures were going to occur. It’s also well known that failures do not originate at the global component level where FEA is applied. Failure is a localized event that occurs within the process material makeup – within the microstructure itself. A processed component is really assembly of millions of individual grains of minute size that have been formed together.
VEXTEC creates 3D representations of the material complexity at the grain level because that is where damage originates.
VLM simulates this microstructural complexity of the material response to stress by creating 3D representations of the granular arrangements in the component. Global stress from FEA is translated from the global component level to the simulation of an individual grain through the virtually created microstructural arrangement. Degradation is simulated on every simulated grain in the computer language of 1s and 0s. Component durability is the aggregation of all the millions of bits of degradation simulations. This is a level of computational processing that has been here-to-before nonexistent within Product Lifecycle Management (PLM) computational offerings. VLM simulates every grain within the arrangement, as well as the voids, inclusions, defects, various grain boundaries, etc., in short, all the various damage mechanisms present in the microstructure to determine and define how they will react to the stress energy imparted at each location. A single fleet durability prediction can consist of hundreds of billions of individual simulations which have been processed together. To make forecasts of component fleets being operated in complex mission scenarios or usage patterns, our VLM technology conducts these hundreds of billions of fleet simulations – dozens of times – in processing times of hours rather than months or years as required with physical testing today.
Our standard process for creating a Virtual Twin product simulation starts with garnering the knowledge our client already has about the product. Sometimes this is as little as a box of broken parts – while for other products this might include laboratory work, physical testing and computational modeling. The first step in the process is gathering the material statistical counts needed so that the material complexity can be simulated in 3-D. While it’s not uncommon for material science groups within large companies have a plethora of data on the material, they hardly ever have the statistical information we need because they have historically had no use for those characteristic details. After we perform this “CSI: Crime Scene” level investigation of the material, we conduct a simulation of the material behavior which we validate through a standard material calibration process. Once the material simulation is calibrated, we move to the next phase in which we simulate the product. The simulator is set up to accurately reflect the product details, including geometry, its manufacturing process, and in-service conditions. Uniquely the variability about all these parameters is entered into the simulation setup for assessment in combination with the processed material variability. Upon VEXTEC’s initial exercising of the simulator, it is prepared for activation via the web for direct client use. But that’s just the beginning. As our client begins exercising simulator perturbations of usage scenarios, manufacturing tolerance adjustments, and material processing changes –their new insights will impact their thinking about development refinements. We modify the simulator to reflect this new thinking so they can see the virtually derived impacts before significant amounts of engineering or physical test resources are expended. After all, VLM can conduct 1 million virtual testing in less time than it takes to physically test 10 product renderings. What follows is a list of the kinds of information we’re interested in, however, VLM is capable of performing accurate simulations with much less:
|
