Nonlinear model predictive control of an intensified continuous reactor using neural networks

In this work a neural network based nonlinear model predictive control algorithm is developed and applied for an intensified continuous reactor. At first, a neural network model of the process is trained and tested using available data sets generated from the first-principal model. Next, a local linearization of neural network model at every sample time is developed to guarantee an efficient online optimization. Simulations are implemented for set point tracking and model mismatch scenarios.