基于改进的神经网络寻优算法在煤粉锅炉优化中的应用

Application of an Enhanced Neural network Optimization Algorithm in Pulverized Coal Boiler Optimization

为进一步挖掘煤粉锅炉的燃烧效率,提出一种改进的神经网络算法搭建锅炉燃烧模型,并采用自适应遗传算法与动态优化边界相结合,分别对最佳氧量、一次风量、各二次风门以及燃尽风门等参数进行寻优计算,以使锅炉效率和污染物排放量所构建的多目标函数达到最优燃烧范围。在保证环保指标合格的前提下,显著提升了锅炉燃烧效率。

In a bid to delve deeper into the combustion efficiency of pulverized coal boilers,an improved neural network algorithm is proposed to build a boiler combustion model,and the adaptive genetic algorithm is combined with dynamic optimization boundary to optimize and calculate the optimal oxygen amount,primary air volume,secondary damper and burnout damper respectively.The goal is to achieve an optimal combustion range for the multi-objective function derived from boiler efficiency and pollutant emissions.While upholding compliance with environmental protection standards"for improved flow,the combustion efficiency of the boiler is significantly improved.