Research on Pressurizer Water Level Control of Nuclear Reactor Based on CMAC and PID Controller

Abstract

A traditional PID regulator is usually used to control water level of pressurizer in modern reactors. For such a non-linear and complex time-varying control system, PID regulator is often shown by the large amount of overshoot and long setting time which are not the desired results. In this paper, a compound controller based on Cerebellar Model Articulation Controller (CMAC) Neural network and PID is presented, which adopts the dynamic errors and the desired values of the controlled plants for the input vectors of CMAC. This controller absorbs simple structure, fast - convergence and part learning advantages of CMAC to enhance traditional controller. The computer simulation experiment demonstrates that the controller performs very well. It is of strong robustness and adaptive abilities and higher control accuracy can be easily accomplished on-line.