Adaptive control of a CSTR with a neural network model

An adaptive control algorithm with a neural network model, previously proposed in the literature for the control of mechanical manipulators, is applied to a CSTR (Continuous Stirred Tank Reactor). The neural network model uses either radial Gaussian or ``Mexican hatʹʹ wavelets as basis functions. This work shows that the addition of linear functions to the networks signi®cantly improves the error convergence when the CSTR is operated for long periods of time in a neighborhood of one operating point, a common scenario in chemical process control. Then, a quantitative comparative study based on output errors and control e€orts is conducted where adaptive controllers using wavelets or Gaussian basis functions and PID controllers (IMC tuning with ®xed parameters and self tuning PID) are compared. From this comparative study, the practicality and advantages of the adaptive con- trollers over ®xed or adaptive PID control is assessed.