CONTROL OF CONTINUOUS STIRRED TANK BIOREACTOR USING NEURAL NETWORK BASED MODEL REFERENCE CONTROLLER

In this article, the application of a neural network based Model Reference control strategy is presented to control a continuous stirred tank bioreactor. The Model Reference is used to design ANN controller. The idea is to have a control system that will be able to give optimal performance, reject load disturbance, and track set point change. With the aim of study the performance of the ANN controller, Proportional- Integral-Derivative controller (PID) strategy is used as benchmark. The neural network based model reference and PID strategies are used for controlling biomass concentration (X) and bioreactor temperature (T). As the results show, the neural network control based Model Reference control can control the concentration and temperature of continuous stirred tank bioreactor effectively in all states, and shown a good improvement in responses than PID mode in feedback control