PID controller tuning by using extremum seeking algorithm based on annealing recurrent neural network

Abstract

This paper proposes a discrete-time extremum seeking algorithm based on annealing recurrent neural network (ESA-ARNN) for auto-tuning of PID controller parameters. Firstly, the process of tuning PID controller parameters is transformed into an extremum seeking problem by introducing a cost function, such as the integral squared error (ISE). Then, in order to solve this extremum seeking problem, a discrete-time ESA-ARNN is proposed, which can realize auto-tuning for PID controller parameters. Lastly, the novel auto-tuning method is applied to tuning PID controller parameters of the process system with second-order plus dead time (SOPDT). Simulation results indicate that PID controller parameters tuned by ESA-ARNN have better performance than those tuned by the eight prevalent PID tuning schemes.