基于智能算法的燃煤锅炉低NO_x燃烧优化  

Low NO_x Emission Combustion Optimization for Coal-fired Utility Boilers Based on the Intelligent Algorithm

作　　者：余廷芳[1] 耿平[1] 

机构地区：[1]南昌大学机电工程学院,江西南昌330031

出　　处：《环境科学与技术》2016年第12期139-143,共5页Environmental Science & Technology

基　　金：国家自然科学基金项目(61262048)

摘　　要：基于MATLAB智能工具箱对某300 MW电站锅炉进行了燃烧优化建模,首先利用BP(back propagation)神经网络建立了锅炉热效率和NO_x排放模型,用以预测锅炉热效率和NO_x排放特性,锅炉热效率预测的平均相对误差为0.14%,NO_x排放量的平均相对误差为1.79%,表明模型具有良好的准确性和泛化能力。基于该燃烧特性预测模型,借助于改进的遗传算法(genetic algorithm,GA)优化模型,在锅炉热效率可接受的某一范围内寻求NO_x排放的最优解,实现锅炉低NO_x排放燃烧优化,对实际的电站锅炉燃烧具有一定的指导意义。MATLAB artificial intelligence toolbox was used to establish model to optimized combustion of a 300 MW utility boiler,firstly,BP（Back Propagation） neural network was used to establish boiler thermal efficiency and NO_x emission model to predict boiler thermal efficiency and NO_x emission,the average relative error of boiler thermal efficiency is 0.14%,and the average relative error of NO_x emissions is 1.79%,indicating that the model has good accuracy and generalization ability. With the aid of the improved genetic algorithm（GA） optimization model,to seek the optimal solution of NO_x emissions in a certain acceptable range of the boiler thermal efficiency,achieve low NO_x emission combustion optimization.The data has a certain guiding significance for the actual utility boiler combustion.

关 键 词：燃煤电站锅炉 NOX排放 燃烧优化 

分 类 号：X701.7[环境科学与工程—环境工程] K383[历史地理—历史学]