基于PSO优化锅炉氮氧化物的排放  

APPLICATION OF PSO IN NO_x EMISSION FROM BOILER

作　　者：凡荣荣[1] 姚翠兰[2] 杨巨生[1] 张永发[3] 谢克昌[3] 

机构地区：[1]太原理工大学电气与动力学院,山西太原030024 [2]山西电力职业技术学院,山西太原030021 [3]太原理工大学煤科学与技术教育部和山西省重点实验室,山西太原030024

出　　处：《热力发电》2012年第8期28-31,共4页Thermal Power Generation

基　　金：国家重点基础研究发展计划(2005CB221202);山西省攻关项目(20110313026-2)

摘　　要：以某超临界600MW机组锅炉为例,采用BP-adaboost算法建立氮氧化物预测模型,通过粒子群优化算法(PSO)优化锅炉运行参数进而在减少氮氧化物排放的同时保持较高的锅炉效率。结果表明,所建模型预测氮氧化物排放量的误差可控制在3%以内,且具有较强的泛化能力;在机组满负荷运行工况下,优化后锅炉实际燃烧效率为92.04%,NOx排放量为520.93mg/m3,相比优化前锅炉效率略有降低,但NOx排放量降低明显。Aimed at keeping high boiler efficiency while reducing the NOx emission,a prediction model was derived based on the BP-adaboost algorithm,taking a supercritical 600 MW unit boiler as the object.NOx and boiler efficiency were optimized by adjusting the boiler parameters with particle swarm optimization(PSO) algorithm.Results show that,the model has good generalization ability,and its error of predicted NOx emission can be controlled within 3%.Under full load operation condition,the actual combustion efficiency of the optimized boiler is 92.4%,a little lower than that before the improvement.However,the NOx emission is obviously decreased,which reaches 520.93 mg/m3.

关 键 词：600 MW机组 锅炉 NOX排放 锅炉效率 BP-adaboost算法 粒子群算法 优化 

分 类 号：TK227.1[动力工程及工程热物理—动力机械及工程] X511[环境科学与工程—环境工程]