Study On Control Of Water Level In Marine Boiler Based On RBF Fuzzy Neural Network

In this paper, the control system of marine boiler is studied. Because of the nonlinear nature of boilers with instable parameter and difficulties of building accurate math model, combined fuzzy control theory and neural networks theory, and proposed a controller based on RBF neural network to control marine boiler. Simulation is performed to the boiler control method of using RBF fuzzy neural network by Matlab software. The results indicate that the control arithmetic presented in this paper is effective, the control effect and the performance is better than that of the PID controller, the fuzzy PID controller and the neural network controller.The main research of the paper is as follows:Firstly, the accurate models of boiler drum water system are established in accord with the theory of marine boiler.Second, relative theories of FNN are summarized. On the basis of fuzzy control and neural network theory, the paper presents a fuzzy neural networks controller which combines advantages of the fuzzy control and neural networks control through the analysis on the shortage of those. It is constructed by RBF fuzzy neural networks.Third, it is designed for the fuzzy PID controller, RBF neural network controller and RBF fuzzy neural network controller based on the same model of boiler water level.Fourth, simulation is performed to the PID controller, fuzzy PID controller, the neural network controller and RBF fuzzy neural network controller by Matlab software. The result indicates that RBF fuzzy neural network control system has a better adaptability and stronger anti-jamming ability, and it is practicable to apply in boiler water level control.