High-fidelity models, such as those based on Finite Element Analysis (FEA), Computational Fluid Dynamics (CFD), and Computer-Aided Engineering (CAE), often demand extensive simulation times, making them impractical for all stages of development. For instance, while a finite element analysis model is invaluable for intricate component design, it proves too slow for system-level simulations necessary for control system verification or extensive system analyses requiring numerous simulations runs. Reduced-order modeling (ROM) offers a compelling solution to reduce the simulation time while maintaining sufficient level of accuracy. ROM comprises a suite of computational techniques that facilitate the transformation of high-fidelity models into swift, lower-fidelity approximations, preserving essential accuracy while significantly enhancing efficiency.

In this session, you will discover how to develop AI-based reduced-order models to replace complex high-fidelity models. Utilizing the Simulink add-on for Reduced Order Modeling, you will learn to conduct comprehensive design of experiments and leverage the resulting input-output data to train AI models using pre-configured templates of LSTMs, neural ODEs, and nonlinear ARX. Gain insights into integrating these AI models into your Simulink simulations for control design, Hardware-in-the-Loop (HIL) testing, or deployment to embedded systems for virtual sensor applications.