Modeling And Designing Of Power Boiler Combustion Controlled System Based On Neural Network And Mixed Intelligent Optimization Algorithms

The electric energy in china is mainly supplied by coal-fired power generation,while coal as a primary energy source has a utilization rate of only about 33 percent.Secondly,the thermal power generation consumes not only a large amount of primary energy,but also generates some pollutants such as dust,SO2,and nitrogen oxides,which may cause environmental pollution seriously.Thus,with the increase of the demand for the electricity,environmental protection and energy crises are intensifying,the research on optimization control technology for gas boiler combustion is of great significance for power production enterprises to reduce costs and increase efficiency,save energy and environmental protection as well as improve the social and the economic benefits.Aiming at the large energy consumption and high emission problems of domestic thermal power boilers,this paper establishes a neural network predictive control model based on NO_x emissions and boiler efficiency which takes Lianyuan Iron and Steel Group’s 23 MW generator set gas boiler as the research object.Then the paper adopts a variety of optimization algorithms to optimize the model,and on this basis,a set of gas boiler intelligent combustion optimization system is designed.According to the demands of the energy-saving and emission reduction technical transformation project of the gas boiler of the generator set,this paper analyzes the process principle of the gas boiler,the factors affecting the boiler efficiency and NO_x emission firstly,and further proposes the optimization requirements and the overall design plan of the gas boiler.And the BP network and RBF network are used to fit and predict the nonlinear function.After comparison and analysis of the simulation,select the RBF network with better prediction effect to model.Then the artificial fish school,particle swarm,artificial fish swarm and particle swarm hybrid algorithm are utilized to optimize the model,respectively.After comparing the advantages and disadvantages of three optimization algorithms,the artificial fish swarm particle swarm hybrid algorithm is selected for optimization.Finally,based on the previous theoretical analysis and the algorithm simulation,this paper designs a gas boiler intelligent combustion optimization control system and puts it into the production operation.The operation result shows that the intelligent combustion optimization system can increase the boiler efficiency by 4.32%,while reducing NO_x emissions by7.47 mg/m~3,meeting the requirements of the NO_x emission below 50mg/m~3,so as to save the energy and the reduce consumption,and to meet the needs of energy saving and emission reduction for gas fired boilers.