Abstract

Industries such as automotive, aerospace, energy, industrials, high-tech, and healthcare are being transformed by Intelligent Systems that will be silicon-designed, software-enabled, and AI-driven. This talk will describe how AI/Machine Learning is being applied to the field of Engineering Simulation of Intelligent Systems. First, we will describe how AI/ML can be used to speed up engineering simulation by developing surrogate models by training neural networks using actual multi-physics simulations over various CAD models, and different boundary conditions. This requires the customers to train the AI models using an AI platform. Second, we will discuss how to develop foundational models for simulation by training AI models over a wide range of CAD models and boundary conditions. This approach does not require the customer to train any AI models, but instead they can use pretrained models on any given CAD geometry. Third, we will discuss how AI/ML models are used to make simulation tools easier to use by automatically setting the parameters of the simulation tools in order to get the best performance and accuracy. Fourth, we will discuss how generative models can be used to explore new designs and optimize product designs. Finally, we will conclude the talk by showing how Agentic AI techniques can be used in engineering simulation automate various tasks and workflows using the concept of “Agent Engineers”.

Bio

Prith Banerjee is Senior Vice President of Innovation at Synopsys, and a member of the Executive Leadership team. Prior to that, he was CTO of Ansys, CTO of Schneider Electric, CTO of ABB, Managing Director at Accenture, and Director of HP Labs. Previously, he spent 20 years in academia as Professor, Chairman and Dean at the University of Illinois and Northwestern University. In addition, Prith has founded two EDA software companies, Accelchip and Binachip. He has served on the Board of Directors of Cray, CUBIC, and Turntide. He is a Fellow of the AAAS, ACM, IEEE, and National Academy of Inventors. He received a B.Tech. in electronics engineering from the Indian Institute of Technology, Kharagpur, and an M.S. and Ph.D. in electrical engineering from the University of Illinois, Urbana.

Abstract

Bio

Abstract

In this talk we discuss a unified framework that combines explainable deep learning, deep reinforcement learning (DRL) and foundation models to advance both understanding and control of turbulence, with direct implications for accelerated design and discovery. First, we will show how explainable deep learning techniques can be used to identify the flow features that are truly responsible for key turbulent processes in wall-bounded flows. By systematically interrogating trained neural networks, we uncover the most influential coherent structures driving momentum transport and drag. Our results reveal that classically studied structures (while important) provide only a partial and sometimes misleading perspective, motivating a more data-driven and physics-aware view of turbulence organization. Building on these insights, we will demonstrate how deep reinforcement learning can be used to actively control turbulent flows by targeting the dynamically relevant structures identified through explainability. This approach achieves over 30% drag reduction in canonical wall-bounded turbulence and extends naturally to more complex configurations, including turbulent wings, highlighting the scalability of learning-based control strategies. Finally, we will introduce a foundation-model-based framework for accelerated design, optimization and scientific discovery. By learning compact, interpretable latent representations of high-dimensional flow physics, these models (combined with agentic-AI systems) enable rapid exploration of design spaces, causal reasoning and closed-loop optimization, bridging the gap between expensive simulations, control and engineering decision making. Together, these results illustrate how explainable and agentic AI are becoming essential for turbulence physics, flow control and next-generation engineering design.

Bio

Dr. Ricardo Vinuesa is the Associate Chair for Research and an Associate Professor at the Department of Aerospace Engineering, University of Michigan. He studied Mechanical Engineering at the Polytechnic University of Valencia (Spain), and he received his PhD in Mechanical and Aerospace Engineering from the Illinois Institute of Technology in Chicago. His research combines numerical simulations and data-driven methods to understand, control and predict complex wall-bounded turbulent flows, such as the boundary layers developing around wings and the flow in urban environments. Dr. Vinuesa has received, among others, an ERC Consolidator Grant, the Harleman Lecture Award, the TSFP Kasagi Award, the MST Emerging Leaders Award, the Goran Gustafsson Award for Young Researchers, the IIT Outstanding Young Alumnus Award and the SARES Young Researcher Award. He received the Outstanding Reviewer Prize of the Journal of Fluid Mechanics and he is also a member of the Young Academy of Science of Spain.

Abstract

The massive applications of additive manufacturing are hindered by the insufficient understanding of process-structure-property relationships and lack of intelligent monitoring/diagnostics/control tools. To this end, we have been developing intelligent additive manufacturing systems, where digitalization is a key enabler. We start from multi-scale multi-physics modeling to understand the fundamental mechanisms, spanning from powder and melt pool dynamics, to microstructure and mechanical properties, i.e., process-structure-property relationships. To enhance the computational efficiency, we have developed physics-informed AI models using physics-based simulation results as the training data and physical laws as the constraints. To achieve rapid diagnostics to identify defects, given that most sensor data only contains limited and indirect information, we leverage physics-based simulation and AI models to extract more in-depth and instructive information. Furthermore, we showcase a closed-loop control algorithm with the diagnostics as the feedback, to adjust the scan path in real time for minimization of thermal stresses.

Bio

Dr. Wentao Yan is an associate professor with tenure in the Department of Mechanical Engineering, National University of Singapore (NUS). Before joining NUS in 2018, Dr. Yan was a postdoctoral fellow at Northwestern University. He received his Ph.D. degree jointly at Tsinghua University (Beijing) and Northwestern University (US) in 2017, and Bachelor degree from Tsinghua University in 2012. He has established and integrated a series of multi-scale multi-physics models to reveal the process-microstructure-property relationships in additive manufacturing, and systematically validated against in-situ/ex-situ experiments. He has spearheaded global efforts to enhance the predictive accuracy of simulations in additive manufacturing, and received consistent top-tier recognition in NIST AM benchmark simulation challenges. He currently serves as the Senior Editor for Additive Manufacturing Journal and an editorial board member for International Journal of Machine Tools and Manufacture, and a few others.