Adaptive Neuro-Fuzzy Inference System PID controller for SG water level of nuclear power plant

In a nuclear power plant, the water level in the steam generator (SG) is one of main causes that shutdown the reactor, this problem has been of great concern for many years as the SG is a highly nonlinear system showing inverse response dynamics. For controlling the SG water level at a certain range, adaptive neuro-fuzzy inference system (ANFIS) PID controller is designed. It is based on the adaptive neuro-fuzzy inference system, using error and error change rate as inputs of fuzzy PID. Neural network has ability to self-study, so fuzzy rules of ANFIS PID can adjust adaptively on line. Applying ANFIS PID controller on the steam generator, comparing with cascade PI controller and fuzzy PID controller, result shows that ANFIS PID can control the level of SG more accurately and safely.