Research on pressurizer water level control of nuclear reactor based on RBF neural network and PID controller

Abstract

In modern reactors a conventional PID controller is adopted in water level of pressurizer. Pressurizer is a controlled object which has the characteristics of large-delay, nonlinear and multi-disturbance. It is difficult to have precise mathematical model. So it cannot be satisfied with control requirements. Prompted by the feedback regulation mechanism of neural network, a composite control strategy based on RBF tuning PID control is presented in this paper. RBF neural network has the characteristics of strong robustness and better self-adaptive ability, which can be adaptive to the change in the parameters of the controlled plant and be easily accomplished on-line. A simulation study shows that it has strong robustness and adaptive abilities and higher control accuracy