Study On Fuzzy Neural Network Control Of Hydraulic Turbine Governing System

Hydraulic turbine governing system is a time-varying and non-minimum-phase system, also it is a nonlinear system with the parameters changing along with the operating-point. The conventional hydraulic turbine governing system can't automatically modulate PID parameters according to the dynamic process of the system, the generator speed is unstable and the mains frequency fluctuation appears. To solve the above problem, on the basis of collecting the domestic and international data about hydraulic turbine governing system, the fuzzy neural network control of the hydraulic turbine governing system is proposed. Also, the study includes mathematical models building, simulation and parameters analysis.The mathematical models of hydraulic turbine governing system under small fluctuation condition are built on the basis of the analysis of hydraulic turbine principle. The Simulink toolbox of MATLAB is used to simulate in several typical operating conditions, and the impact of system dynamic performance is analyzed when the parameters change. The fuzzy neural network PID controller is applied to overcome the defects of simple fuzzy control and simple neural network control. Also, the system architecture, algorithm and simulation of the controller are introduced. Finally, the fuzzy neural network control is applied in hydraulic turbine governing system. The control scheme, controller architecture, controller parameters setting especial the design of the fuzzy rules are introduced in detail. The simulation results show that the dynamic process of the system can be reflected more precisely and the on-line adaptive control is implemented by the fuzzy neural network control. The rate of convergence, control performance and robustness is obviously superior to conventional PID control. This project has popularizing value and significance of theory.This research project is sponsored by"Study on Adaptive PID Algorithm of Hydraulic Turbine Governing System Based on Fuzzy Neural Network (Innovation Project of Guangxi Graduate Education, Project Number: 2006105950802M09)".