Physics-based AI-assisted Design and Control in Flexible Manufacturing

Explore advanced flexible manufacturing processes using hybrid physics-based and data-driven approaches in this 56-minute talk by Professor Jian Cao. Delve into challenges faced in manufacturing and examine two flexible processes: metal powder-based additive manufacturing and rapid dieless forming for producing three-dimensional parts without geometry-specific tooling. Learn how integrating fundamental process mechanics, process control, and machine learning techniques achieves effective predictions of material behavior during manufacturing processes. Discover the application of machine learning for active sensing to enable effective in-situ local process control, addressing challenges such as long history-dependent properties, complex geometric features, and high-dimensional design spaces. Gain insights into innovative manufacturing processes, systems, and research directions from an expert in deformation-based and laser additive manufacturing processes.

Introduction
Lab Goals
Differentiable Simulation
Process Modeling
Multilayer Simulation
Process Control
Closed Loop Control
Data Fusion
Future
Doublesided Incremental
Hybrid Autonomous Manufacturing
Future Directions
Thank You
Questions
Simulation Experiments
Future Work
Control Variables