基于改进NSGA-Ⅱ算法的W火焰锅炉燃烧系统多目标优化  

Multi-objective Optimization of W-flame Boiler Combustion System Based on Improved NSGA-II Algorithm

作　　者：庞梦媛 赵文杰[1] PANG Mengyuan;ZHAO Wenjie(Department of Automation,North China Electric Power University,Baoding 071003,China)

机构地区：[1]华北电力大学自动化系,河北保定071003

出　　处：《电力科学与工程》2024年第9期71-78,共8页Electric Power Science and Engineering

基　　金：河北省自然科学基金资助项目(F2014502059)。

摘　　要：以W火焰锅炉为例,首先探讨了可调变量与NOx排放量和锅炉燃烧效率之间的关系,将所得的先验信息采用单调约束的形式与数据驱动建模融合,建立了灰狼优化融合先验知识的支持向量机燃烧特性预测模型。结果表明,该模型提高了预测精度。在此基础上,针对NSGA-Ⅱ算法易陷入极值的问题,加入带有惩罚机制的锦标赛选择策略,对不同工况下的W火焰锅炉进行多目标燃烧优化实验,并提出将归一化环保和经济指标得到的综合效益因子作为评判锅炉燃烧优化结果的标准。实验表明,出力为290MW工况下,综合效益因子提高了7.11%。Taking the W-flame boiler as an example,the relationship among adjustable variables,NOx emissions,and boiler combustion efficiency is first explored.The obtained prior information is fused innovatively with data-driven modeling in the form of monotonic constraints,and a prediction model for combustion characteristics of support vector machine based on grey wolf optimization and prior knowledge fusion is established.The results show that the model improved the prediction accuracy.On this basis,in response to the problem of NSGA-II algorithm easily falling into extreme values,a tournament selection strategy with a penalty mechanism is added to conduct multi-objective combustion optimization experiments on W-flame boilers under different operating conditions.The comprehensive benefit factor obtained from normalized environmental and economic indicators is proposed as the standard for evaluating the combustion optimization results of the boiler.The experiment shows that under the operating condition of 290 MV output,the comprehensive benefit factor increased by 7.11%.

关 键 词：燃烧优化 NOX排放量 锅炉燃烧效率 支持向量机 NSGA-Ⅱ 

分 类 号：TK019[动力工程及工程热物理]