Evolutionary computation applictaion in power level control of nuclear power station

The problem of controlling the output power of the nuclear power station (NPS) is considered. Automatic power stabilization control is the desired goal for any nuclear reactor operation policy, especially, NPS. In this paper the conventional controllers like Proportional-derivative (PD) were designed for the power control system of the NPS. Tuning techniques such as the Ziegler-Nichols and Cohen-Coon tuning formulae, but is difficult to do that for nonlinear and time varying systems. A method to improve the performance of proposed controller is introduced. The objective is to tune proportional gain and differential gain of proposed controller. Combinations of GA and PSO are carried out. The optimal parameters of desired controller are achieved. The PWR is selected as NPS, and it was simulated using MATLAB/SIMULINK package. Simulation results indicate that the designed controller greatly improves the performance of NPS.