Neural Networks Approach to Automatic Startup and Control of An Exothermic Batch Reactor

This paper describes a neural network approach to the automatic startup and control of a batch reactor where exothermic reactions take place. A backpropagation neural network model is used for online determination of the startup switching time and a model predictive control strategy based on a back propagation neural network is to be designed for a regulatory control after the desired reactor temperature is reached during startup. Simulation results are presented to illustrate the applicability of the proposed neural network approach to the rapid startup and control of exothermic batch reactors.