Systems Engineering Enhanced by AI-Driven Multiphysics Simulation: Multiphysics Modeling and Simulation with Artificial Intelligence / Multiphysics Modeling and Simulation for Technology Transfer Using Artificial Intelligence

Abstract

Recent advancements in artificial intelligence (AI) allow for more sophisticated modeling and simulation of complex engineering systems. This research explores the application of AI techniques like neural networks and genetic algorithms for multiphysics modeling, using an aerospace case example for educational purposes. A literature review examines existing physics-based and empirical modeling approaches in this domain. Subsequently, a multiphysics control model incorporating thermal, structural, and fluid dynamics interactions is developed. Neural networks can be trained on simulation data to learn these multiphysics relationships, with the potential to augment robotic assembly. Additionally, genetic algorithms optimize system designs by evolving populations of models based on performance objectives. This enables rapid virtual testing and discovery of optimal configurations. The integrated AI modeling framework builds on a systematic literature review, providing a reference architecture for multiphysics modeling and simulation. Literature findings facilitate developing an optimal methodology for a model example. The research demonstrates advancing complex engineering models via a sample pseudocode algorithm for electronic systems control. Integrative systems engineering research can enhance simulation-driven design. This pseudocode contributes knowledge on AI-driven multiphysics modeling within a scientific framework. The proposed technique has applications in innovating systems-level designs with prudent limitations.