Artificial Neural Network-Based Controllers for A Continuous Stirred Tank Heater Process

摘要

Two linear Artificial Neural Network (ANN)-based controllers for the control of a nonlinear system are presented. The optimization and pole placement control design strategies are used. For both strategies, linear ANNs are used to model a Continuous Stirred Tank Heater (CSTH) process. Third-order discrete-time models or Adaline ANNs are used whose parameters are updated at every sampling time to provide adaptability to the controllers. Simulation tests show that the linear ANN-based controllers can overcome strong nonlinear coupling effects, reject step disturbances, and provide adequate damping to the subsystem interactions. Both ANN-based control design strategies are shown to be interchangeable with each other for the control of the CSTH process.