The Composite Control Of The Boiler Continuous Sewage Based Multi-parameter Fusion About Water

The boiler continuous blowdown system plays an important role in the safe andeconomic operation in the power plant units. With the increase of the unit parameterand the improvement of the standard of energy conservation and emission reduction,the traditional continuous blowdown control method can not satisfy the requirements.The research on the new intelligent continuous blowdown system should be adirection of stable operation technology in the power plant.According to the nonlinear, time varying, large delay and multivariable of boilercontinuous blowdown process, as well as hard to establish the accurate mathematicalmodel, the new type of boiler continuous blowdown compound control system isdesigned in this thesis. Compound control system is adopted data fusion technologyto complete the design of feedforward controller and multivariable integration ofcontinuous blowdown process, which is on the basis of feedforward and feedbackcontrol. The fuzzy self-tuning PID control method is chosen to replace the traditionalPID control to improve the robustness and accuracy in the feedback control.It can be mainly discusses the following two parts in this thesis.1The establishment of the system model. According to the feature of the boileroperation, the fusion architecture of the two levels data with local parallel and totalserial is designed for the original data collected by the instruments to calculate onelevel fusion by arithmetic average recursive estimate, which can increase theeffectiveness of the data. The single output network is designed a relationship whichis the fusion data chosen as the input of BP neural network and the output chosen asthe continuous blowdown flow. The fuzzy self-tuning PID controller is designed bythe characteristics of practical engineering application.2Model vilification. The feedforward model is trained and verified by the datafrom a600MW unit. Results have shown that the designed feedforward model can fulfill the task of every level fusion and meet the requirements of stability andaccuracy. Then the simulated experimental platform can be set up by themathematical description of the controlled object. The design fuzzy self-tuning PIDcontroller can realize the feedback control for the controlled object, which canimprove the dynamic and static performance compared with the traditional PIDcontrol.The model of the whole system is simulated by Simulink, which is shown thatthe compound control system can not be effective and available, but also hasadvantages of high control accuracy, short setting time, strong robustness and gooddynamic performance compared with the traditional control system.