Research On Fuzzy Control Algorithm And Its Application In Main Steam Temperature Control Of Thermal Power Plants

In the operation of thermal power plant unit units,the main steam temperature is a very important parameter,which has a great impact on the safety and economy of thermal power plant operation.However,due to the large delay,large inertia and multi-disturbance characteristics of the superheated steam object,a large overshoot and a long adjustment time may occur,which makes the conventional PID control become more difficult,so the main steam temperature control is always a fire.The hot spot of power plant operation control research is also difficult.Current control methods such as predictive control,adaptive control,and state variable control,etc.,based on modern control theory,are rarely used in the field due to flaws in their algorithms.At the same time,intelligent control methods such as fuzzy control and neural network control are developing rapidly.This paper studies this situation.The main research work is as follows:(1)Introducing the long-term and short-term memory(LSTM)neural network into the prediction of the main steam temperature.Because the algorithm itself has the characteristics of time series,the main steam temperature is also used as one of the input variables to participate in the prediction,and the actual data of the thermal power plant is used for one day.The main steam temperature is predicted,and the feasibility and accuracy of the algorithm in the prediction of main steam temperature are proved by prediction results.(2)Improve the value of the weighting factor used to balance the deviation and the rate of change of the deviation in the fuzzy control.According to the fuzzy control idea,the weighting factor is based on the magnitude of the deviation and the rate of change of the deviation,and the Lagrangian interpolation method is used to perform algebraic interpolation on the weighting factor so that it can be continuously changed,so that the fuzzy control has adaptive ability..(3)A fuzzy control method based on long-short-term memory neural network algorithm prediction is proposed.The long-term and short-term memory neural network is used to predict the main steam temperature at the next moment,and the value of the main steam temperature at the next moment and the main steam at the current moment are obtained.The temperature is used as the input of the controller calibration link,and the controller calibration link is also fuzzy control.The deviation of the main steam temperature change and the rate of change of the deviation are input,and the difference between the main steam temperature and the set value difference is different at this time.The value of the calibration link rules,and finally the output of the calibration link and the original fuzzy control take the joint action on the water spray desuperheating valve tocontrol the main steam temperature.The effectiveness and safety of the method in the main steam temperature control are verified by simulation experiments.