基于GA-BP的锅炉燃烧系统的仿真与研究

Simulation and Research of Boiler Combustion System Based on Genetic Algorithm Optimization Neural Network

在当今社会中，我国的发电仍然以火力发电为主．锅炉能否安全经济的燃烧是火力发电厂最为关心的问题．锅炉安全经济的燃烧不仅可以使得发电成本降低还能有效的减少有害物质的排放．依据某热电厂的现场数据选用BP神经网络来对锅炉燃烧系统进行建模分析．利用遗传算法对BP神经网络的权值、阈值进行优化，建立更为理想的GA—BP网络模型．实验结果表明，在采用遗传算法优化的神经网络之后，模型的收敛速度、训练精度得到了有效的提高．为了进一步的研究该系统，建立模糊PID优化控制器并仿真，仿真结果证明了模糊PID的优越性．以上研究说明，通过以上手段可以有效地优化热电厂的锅炉燃烧系统．

In today＇s society, China＇s power generation is still mainly based on thermal power generation. The safety and economic combustion of boiler has been the most concerned problem of thermal power plant. The combustion of boiler safety and economy can not only reduce the cost of power generationl but also reduce the emission of harmful substances. Based on the field data of a thermal power plant, the BP neural network is used to model the boiler combustion system. The weights and thresholds of BP neural network are optimized by genetic algorithm, and a more ideal GA-BP network model is established. Experimental results show that the convergence speed and the training accuracy of the model are improved after the optimization of the neural network using genetic algorithm. In order to studying the system further, a fuzzy PID optimization controller is established and simulated. The simulation results show the superiority of fuzzy PID. The above research shows that the boiler combustion system of the power plant can be optimized effectively by these methods.