基于神经网络和遗传算法的锅炉燃烧优化方法

Boiler combustion optimization based on neural network and genetic algorithm

针对锅炉燃烧控制系统送风调节系统存在的弊端,遵照火电厂锅炉燃烧既要提高效率又要降低污染物排放的要求,对神经网络和遗传算法在火电厂锅炉燃烧优化中的应用进行了研究。首先借助燃烧特性试验数据,建立了火电厂锅炉燃烧特性的神经网络模型,然后应用遗传算法寻找送风调节系统最佳氧量设定值,进而调节送风量,实现锅炉燃烧的整体优化。仿真结果表明:应用该方法指导锅炉燃烧,不仅能使锅炉节能,还能降低排放的烟气中氮氧化物的含量,减少对环境的污染。

To overcome the defects in forced draft control system, and to meet the requirement of high efficiency and low emission, a boiler combustion optimization method combining neural network and genetic algorithm was researched. First, a neural network model of boiler combustion characteristic was constructed based on the data of com- bustion experiment. Then genetic algorithm was used to seek the best setting point of Oxygen content in forced draft control system for the further control of supply air rate. The simulation shows that this combined method can not only make boiler run with the least energy but also reduce Nitrogen Oxide emission and diminish pollution to environment.