The Study Of Fuzzy Control Based On Genetic Algorithm In The Power Plant Boiler Combustion System

Thermal power plant boiler combustion control system is an important partof the process of power plant thermal, it is often subject to the internal andexternal interference when its actual operation. It is a multi-input andmulti-output controlled object, and there is the serious coupling between thevariables, and it has multi-parameter, nonlinear, uncertain time-delay and thetime-varying characteristics. The traditional PID control is difficult to achievethe desired control effect.This paper first has studied the control tasks of the power plant boilercombustion and the characteristics of the controlled object. Based on thecharacteristics of the power plant boiler combustion control system, it has beensimplified to three subsystems; they are the fuel control system, air supplycontrol system and blower control system, according to the characteristics ofeach subsystem, using different control strategies: Fuel control system uses thefuzzy-PI cascade control, air supply control system using fuzzy self-optimizingcontrol, blower control system feedforward compensation control strategy. Then,the basic principles of fuzzy control and genetic algorithm was in depthresearched, design the fuel, air and blower subsystems fuzzy controllerrespectively. Genetic algorithm is a global probabilistic search algorithm, it isable to search the complex solution space effectively, and especially for complexproblems optimization effect is obvious. To avoid the limitation of conventionalfuzzy controller over-reliance experts or operator experience, introduced theidea of uneven division of the fuzzy domain, using the improved geneticalgorithm for global optimization the fuzzy controller, including the optimization of membership function and fuzzy control rules. Then improvedgenetic algorithm, proposed a cross-selection method to expand the search space,while using the non-uniform variation to improve the late local search ability ofthe algorithm. Finally, the fuzzy controller which is optimized by the geneticalgorithm is used in the power plant boiler blower control system, simulationresults show the effectiveness of the method used in this paper.