Modeling And Combustion Optimization Of Heating Boiler Based On Intelligent Algorithm

The essence of the optimal operation of the boiler combustion system is to adjust the operation parameters online,without changing the parameters of the boiler equipment,so that all the operating parameters of the boiler combustion are in the best working condition,while reducing pollutant emissions and improving the boiler efficiency.The combustion process of Coal-fired boiler is complex and parameters have the nonlinear relationship.It is difficult to make the accurate mechanism model and conventional control methods effect is not ideal.With reasonable modeling method and intelligent optimization algorithm,the controllable parameters of the boiler combustion system are optimized in real time.Based on the common boiler modeling method and optimization method deeply,this thesis determines the coal-fired boiler combustion optimization and intelligent modeling algorithm based on the topic,the main research contents are as follows:First of all,the research of combustion control system and combustion efficiency equation of coal-fired boiler flue gas oxygen content are studied,determined the impact on emissions of pollutants,the modeling and optimization control goal is the thermal efficiency and flue gas oxygen content,designed collection establishing control target model by historical data.Optimization algorithm was used to obtain the optimal control vector,then release control vector by the WEB client,optimizing combustion control system of field adjustment of DCS system.Secondly,selected the input variables.Used the stability judgment and Pauta Criterion to remove the abnormal data.Used the Least Squares Support Vector Machine(LSSVM)to establish the thermal efficiency and flue gas oxygen content model.using simulated annealing particle Swarm Algorithm to determine the inertia weight(SAPSO)was established to optimize the parameters of the model.Although improved the model prediction ability and convergence speed,the prediction accuracy and precision has differents.The problem is To avoid parameter optimization is the empirical risk minimization,used the model of Cross Validation(CV)results in the construction of objective function and realized ructural risk minimization with sacrificing a part of the fitting precision.To improve the prediction ability of model further,established an accurate model based on SAPSO-CV LSSVM hybrid modeling algorithm.Finally,make the thermal efficiency as the main target,the oxygen content of the flue gas as a restrictive condition in Drosophila Optimization Algorithm(FOA)based on the introduction of the negative change amount can jump off the parameters,introduces the variable step to avoid local optimal improvement,to increase the precision of optimization in three-dimensional space,using Drosophila algorithm(VS-MFOA)based on the optimization model,optimization,obtained under the same load conditions,the control variable optimal combustion state.With the operation data of a high temperature hot water boiler which named 72MW DZL72-1.6/150/90-A ? of a heating company in Dalian,this method can provide the workers valuable decision-making guidance,it has a positive effect on the boiler safe and efficient operation,Expected to achieve thermal efficiency of 5%?15%boiler operation about ascension.