Development of Nonlinear Controllers using Deep Neural Networks and Physically Inspired Neural networks
Many unit operations in chemical plants exhibit strongly nonlinear dynamic behavior. Developing control schemes for such systems using first principle models can prove to be a time-consuming and expensive task. There is growing interest in using deep neural networks and physically inspired neural networks for developing nonlinear dynamic models for such systems. This project explores the possibility of developing nonlinear control schemes under non-linear internal model control or nonlinear model predictive control schemes using deep and physically inspired neural networks.