The Optimization Of Back Pressure Control System Direct Air Cooled Consenser Based On NNPC

The application of air-cooled technology in the fire power plants is one of the effective methods in order to save water resource in the region where coal resource is very rich but water resource is relatively scarce. The steam Turbine back pressure is an important control parameter of the air-cooled unit. The quality of pressure system is directly related to the entire unit to safe and economic operation of the unit.The back pressure control system control the back pressure by adjust the fan speed to change the air flow through the downstream cooling unit and the counter-current cooling unit, in order to ensure the safety of the turbine, the economic operation. As current Monitoring conditions are limited, the study of the relationship between back pressure and air-cooled island environment meteorological parameters is not deep enough, as a result, control method of the back pressure system is for simple PID control, which failed to achieve the desired control effect.This subject first analyzes the characteristics of the direct air-cooled back-pressure system and its control performance requirements, and do some research on the optimal operation of the direct air cooling unit. Secondly, the paper makes the use of Changzhi thermal power plant simulator which is similar as the unit of Shanxi Pingshuo Gangue-fired Generation to do the experiments. Use the step response of the back pressure to build a the mathematical model of300MW direct air cooling, then choose neural network predictive control system of intelligent prediction instead of the traditional PID control system according to the characteristics of the back pressure of air-cooled unit. MATLAB simulation experiment show that compared with traditional PID control system, the neural network predictive control system has a better quality of regulation, as well as good resistance, therefore, the NNPC has a high practical value. In addition, the paper analyzes the main influencing factors on the direct air-cooled back-pressure system, establishes a direct air cooling of the neural network nonlinear model depending on the data collected at the scene. Lastly, obtains the relationship between these factors and the backpressure, and elaborates the operation conditions in extreme environments of Pingshuo Gangue-fired Generation and the measures to antifreeze and cross summer.