The Research Of Heating Boiler's Operation Optimization Based On Data Driven

Heating enterprises belong to high energy consumption industries,due to uneven extensive management,low level of automation and operator quality,in specific operation management and often because of the lack of effective "starting point",it is difficult to carry out targeted guidance.Therefore,it is of practical significance to study the man machine intelligence mutual and help take the right operation decision to realize the high efficiency and energy saving of boiler operation.Boiler heating is a typical complex industrial process,which has multi-input andoutput,strong non-linear and coupling,no self-balance and large lag and so on.It's difficult to use the traditional mechanism analysis to get accurate mathematical model in the optimization process.A large number of boiler operation data can be stored with the gradual improvement of the degree of automation of heating boilers and the application of information technology.But a lot of hidden information has not been effectively excavated due to the lack of effective data analysis.This thesis studies the optimization method of the data-driven heating boiler operating parameters based on the operation mechanism of the boiler.The main contents of this thesis include the following points.Firstly,this thesis untangle the boiler process flow and combustion heat balance equation.The optimization framework of boiler operation mode based on data driven was introduced,then the boiler exhaust gas temperature,the furnace negative pressure and boiler oxygen content were determined as the boiler state parameters.The operation mode framework of wind,blast,grate and coal were established as operating parameters and the model library was set up to optimize the boiler thermal efficiency.Secondly,the time series analysis method is used to deal with a large amount of historical data accumulated in boiler operation,and the Largest Lyapunov Exponent Method and C-C Method are chose to analyze the time series parameters of boiler,the BP neural network was used to study the function relationship between the boiler thermal efficiency,and the process parameters in the optimization model library was used to establish the prediction model of the boiler index chaotic time series,which is the objective function of the optimization of the operating mode.Finally,the optimization of boiler operation mode based on chaos algorithm is proposed.The relationship between boiler thermal efficiency and process parameter function established by BP neural network is taken as the objective function.The ergodicity of chaotic variables,and the traversing range of chaotic motion is extended to the range of the operating parameters to be optimized.Finally,a set of optimized operation parameters of the boiler are obtained on line from the chaotic variables of iterative search.Combined with a heating company in Dalian on the 4th 72MW DZL72-1.6/150/90-A II high temperature hot water coal-fired chain boiler operating data validation that the method can provide valuable guidance for the operator to improve the efficiency of the boiler,and has a positive effect on the stable and safe operation of the boiler.