A new online tuning approach for pid control of multivariable systems using diagonal recurrent neural Network

In this paper a new intelligent control based scheme has been proposed for tuning PID controller parameters (propositional, derivative, and integral gains) of multi-variable systems using diagonal recurrent neural Networks (DRNN). It utilizes the basics of back propagation algorithm with momentum constant. A simulation study of nonlinear, interactive, two input-two output (TITO) system is included to demonstrate the effectiveness of the scheme. The simulation result supports the contention that, without slowing down the rise time and overshoot the it is possible with proposed scheme to track the set point changes and also able to reject the disturbances that may enter.