1-D System-Level Simulation leveraged for Engineering Controls’ Strategy Optimization

R​emi Duquette, Maya HTT Ltd.

Abstract

Machine Learning (ML), Deep Learning (DL), and Artificial Intelligence (AI) have been the talk of the town worldwide for a few years now. Many new DL algorithms provide new ways of making digital twins, and simulation models in general, more relevant to the manufacturing process or end-product operational conditions they are meant to improve. Powerful new AI agents can be created by combining reinforced learning algorithms and 1-D system-level simulation models to design and optimize engineering control strategies. Getting an AI agent to learn from “playing” with a 1-D system-level engineering simualtion model in a self-play mode provides new and amazing ways to optimize your complex operations and learn complex control plants built in for system-level enhanced controls.