Study On Control System For Boiler Drum Water Level Based On Fuzzy Neural Network

Boiler is a necessary power equipment in the process of industry. In order to ensure the production and security, the auto- control of boiler is very important. The drum water level is an important variable to be controlled. Due to the adjustment process of the boiler water level being nonlinearity, instability, time delay and so on, the traditional control system adopting widely used PID mode is unsatisfactory.In the paper, a survey of basic principles of the fuzzy control,network control and fuzzy neural network is given.A fuzzy-neural network controller with Mamdani model is applied to the drum water level control to replace the PID controller.The controller can resolve the problem of the large time delay and parameter instability existing in the process of water level to achieve better energy-saving effects and satisfied with the quality index.At first through certain assumptions and simplification a mathematical model of the boiler drum water level is derived according to the working principle of boiler and the method of theoretical modeling. Described the mathematical model of drum water level under different disturbance conditions.By simulation comparison between cascade three-impulse control and fuzzy neural network control, the results show that the fuzzy neural network control system has good control effect in the adaptability, the advantages of anti-interference ability, indicating the feasibility and superiority of the control system.