Design of on-line auto-tuning PID controller for power plant process control

In the power plant process control, classical PID controllers are widely used. In order to get the desired control performances, the correct tuning of PID controllers is very important. In this paper, a new auto-tuning method of PID controller without using test signal generator has been developed. This method uses only the measured process output and controller output signals to identify the unknown control process. From the estimated process model, a controller design rule is derived such that it minimizes the average mean squared error between desired closed-loop step response and that of the control system. Because it is possible to estimate the process without using any special test signal generator, this method can be applied to the safety-related processes such as nuclear power plant control system.