Analysis Of Influencing Factors And Prediction Optimization Of Circulating Fluidized Bed Boiler Bed Temperature Based On Massive Data

In response to our country’s policy demands on"carbon emission reduction"and"carbon peaking",it is necessary to improve energy utilization efficiency and promote the development of clean technologies.Circulating fluidized bed boiler(CFB)has been an important development direction of clean energy utilization technology in recent years due to its unique advantages of fast load regulation,high desulfurization efficiency,and low pollution emissions.Bed temperature is an important parameter of the complex combustion system of the circulating fluidized bed boiler,which will affect the pollutant emission efficiency,combustion efficiency and even the overall operation safety of the boiler.Predictive control has always been a difficult problem.Accurately controlling the change of bed temperature so as to adjust the relevant influencing parameters in time is the key task of power plant operation.Therefore,it is of great significance to explore the influencing factors of bed temperature and establish an accurate dynamic prediction model for the further development of circulating fluidized bed boiler technology.This paper conducts a comprehensive study on the CFB boiler bed temperature problem.On the basis of the traditional heat transfer method of boiler bed temperature calculation,the massive data method is used to systematically analyze the factors affecting the bed temperature,and a corresponding prediction model is established.The research contents include:1.Research on factors affecting the uniform distribution of boiler bed temperature.Mainly from the perspective of the separation system of return material,the coal feeding and returning material characteristics of the target boiler are analyzed,and the separation characteristics of the system are calculated and analyzed,and the operation optimization and coal feeding suggestions are given.2.Research on the factors affecting the high bed temperature.The traditional calculation model is used to analyze the qualitative effects of primary air,fuel quantity and quality,fuel gas entering the furnace and other factors on the change of bed temperature level.At the same time,other reasons outside the traditional calculation model are also explored based on the actual DCS operation data.3.Establishment of accurate bed temperature prediction model and related controllers and bed temperature optimization.On the basis of the previous analysis of influencing factors,the feature selection method is used to screen the model features,and the particle swarm optimization algorithm is added while the support vector regression model is built,and then the optimal combination of order and parameter is obtained.Through the analysis and research of the massive data information collected in the above actual operation combined with the traditional calculation method,the comprehensiveness and accuracy of the optimization of the bed temperature operation proposal of the power plant are improved,and it has great application value.In this paper,a support vector regression time series model with system order h=3 and input order s=2 is established.The model result R~2 is 0.9939,and the absolute percentage error is 0.0013,indicating that the model can predict the change of bed temperature well.The accuracy of the model is verified under different air distribution and load conditions.The verification results show that the model is suitable for the daily load of the target power plant and the operating conditions of the air distribution scenario,and is an accurate bed temperature level prediction model.Then,the model predictive controller is designed according to the modeling results,and the control performance and anti-interference experiments are carried out.The results show that the controller can have excellent tracking and stability capabilities.Finally,combined with the genetic algorithm to optimize the bed temperature,the optimal parameter combination under each load operating condition was obtained,and the average optimization efficiency of the bed temperature reached 7.90%.