摘要
为了提高锅炉热效率和降低污染物排放,对锅炉燃烧过程进行多目标优化.采用BP神经网络与改进非支配排序遗传算法(INSGA-Ⅱ)建立锅炉燃烧系统多目标优化模型.基于BP神经网络分别建立NO_(x)排放和锅炉热效率模型;以降低NO_(x)排放质量浓度和提高锅炉热效率为目标,基于BP-INSGA-Ⅱ算法对锅炉燃烧系统进行多目标寻优;基于BP-INSGA-Ⅱ算法、BP-NSGA-Ⅱ算法、GRNN-INSGA-Ⅱ算法和GRNN-NSGA-Ⅱ算法分别建立锅炉燃烧优化模型,比较各优化模型的性能,验证锅炉燃烧优化模型的有效性.结果表明:NO_(x)排放质量浓度预测模型和锅炉热效率预测模型最大误差均不超过3%;基于BP-INSGA-Ⅱ算法建立锅炉燃烧优化模型使NO_(x)排放质量浓度平均降低15.42%,锅炉热效率平均提高0.1058%.结合BP神经网络与改进的多目标优化方法建立的锅炉燃烧优化模型能够同时提高锅炉热效率和降低NO_(x)排放.
To improve boiler efficiency and reduce pollutant emissions,the multi-objective combustion optimization for boiler was performed.BP neural network and an improved non-dominated sorting genetic algorithm(INSGA-Ⅱ)algorithm were adopted to establish a multi-objective optimization model for a boiler combustion system.The prediction models of NO_(x) emission concentration and boiler efficiency were established based on BP neural network,respectively.Aiming at reducing NO_(x) emission mass concentration and improving boiler efficiency,the multi-objective optimization for the boiler combustion system was carried out based on BP-INSGA-Ⅱalgorithm.The boiler combustion optimization model was established based on BP-INSGA-Ⅱ,BP-NSGA-Ⅱ,GRNN-INSGA-Ⅱ,and GRNN-NSGA-Ⅱalgorithms,respectively,and the performance of each optimization model was compared to verify the effectiveness of the boiler combustion optimization model.The results show that the maximum errors of the prediction models of the NO_(x) emission concentration and the boiler efficiency are less than 3%.The boiler combustion optimization model based on BP-INSGA-Ⅱalgorithm can reduce NO_(x) emission by 15.42%and increase the boiler efficiency by 0.1058%on average.The boiler combustion optimization model established by combining BP neural network and improved multi-objective optimization method can simultaneously improve the boiler efficiency and reduce the NO_(x) emission.
作者
徐文韬
黄亚继
曹歌瀚
陈波
金保昇
Xu Wentao;Huang Yaji;Cao Gehan;Chen Bo;Jin Baosheng(Key Laboratory of Energy Thermal Conversion and Process Measurement and Control of Ministry of Education,Southeast University,Nanjing 210096,China;Jiangsu Frontier Power Technology Co.,Ltd.,Nanjing 211102,China)
出处
《东南大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2022年第5期943-952,共10页
Journal of Southeast University：Natural Science Edition
基金
江苏方天电力技术有限公司科技资助项目(KJ201927)。
