Hemanth Kolera, MSC Software
The value of a simulation tool is tied to the level of engineering insight it provides. While the validity of the modeling approach, the model inputs, and simulation processes are critical considerations, engineering insight fundamentally depends on the quality and volume of simulation data produced. With the advent of high-performance computing and data analytics, engineering simulations can now be deployed at-scale to explore, understand the sensitivities of each parameter and optimize engineering design under uncertain conditions. An application that lends itself well to this capability is the virtual exploration of the vehicle dynamics design space. The inherently parametric nature of Adams lends itself well to design exploration.
As a case in point, suspension system design is a challenging multi-objective optimization task, due to the presence of several competing design objectives and the essential need for tradeoffs between them. As an example, ride quality and handling have different optimization norms; firmer the suspension means better control of body movements but lower ride quality. So how does one reconcile these competing objectives using an engineering simulation?
There are two avenues for users to optimize and quantify uncertainties in the responses from their mechanical systems with Adams. While staying within the Adams analysis environment one can use Adams Insight, or one can couple with specialized analytics software from outside the MSC portfolio using tailored interfaces. Here are a few highlights from recent advances in the area of design exploration using these avenues.
Support for vehicle dynamics optimization has been extended in Adams Insight in Release 2019. Below are a few highlights:
Optimization is only part of the design exploration process. Analytical methods like Uncertainty Propagation, Sensitivity Analysis, and Calibration can allow users to gauge the influence of uncertainty on the robustness of the design, determine its viability and risk or understand the difference between simulation and test results. For example, if manufacturing tolerances for a component lead to a wide variation in a vehicle dynamics study, Uncertainty Propagation can be used to gauge the magnitude of that variation and Sensitivity Analysis can be applied to identify the tolerances with the largest impact. These analytics can inform an engineer on how to tighten manufacturing tolerances in the most efficient manner or where to select a more robust design.
MSC Software has partnered with SmartUQ to bring its industry-leading uncertainty quantification and engineering analytics technology to Adams. Adams users can leverage the integrated solution to account for design, process and operational uncertainties and gauge how these would impact vehicle performance. At a high level, the integrated workflow would involve the creation of a design of experiments (DOE) in SmartUQ to sample the design space, execution of the DOE by the Adams solver and finally the construction of a statistical emulator to mimic the responses from the simulated dataset in SmartUQ. The emulator can then be used for rapid exploration of the design space and quickly perform Sensitivity Analysis and Uncertainty Propagation. To learn more about applying predictive analytics and uncertainty quantification techniques to Adams simulations, please join MSC and SmartUQ for a webinar (registration link) on the 2nd of May.
