基于改进人工蜂群算法的锅炉NO_x排放预测优化  

Optimization of the Prediction of the NOx Emissions of a Boiler Based on an Improved Swarm Algorithm

作　　者：牛培峰[1] 刘永超[1] 张先臣[1] 张向业[1] 

机构地区：[1]燕山大学工业计算机控制工程河北省重点实验室,河北秦皇岛066004

出　　处：《热能动力工程》2014年第4期427-433,462,共7页Journal of Engineering for Thermal Energy and Power

基　　金：国家自然科学基金资助项目(60774028);河北省自然科学基金资助项目(F2010001318)

摘　　要：针对电厂循环流化床锅炉NOx排放问题进行了研究,并对人工蜂群算法进行了改进,结合最小二乘支持向量机建立了锅炉燃烧NOx排放模型,对锅炉可调参量进行了优化,降低了NOx排放浓度。将改进的人工蜂群算法与基本的人工蜂群算法和粒子群算法进行比较,说明基于改进人工蜂群算法所建立的模型能够很好的预测NOx的排放浓度,具有很强的辨识能力和泛化能力,同时也表明了改进人工蜂群算法计算速度快的优点及优化数据上的优势,通过仿真试验,优化后NOx排放浓度明显降低,体现了其工程实用价值。Studied were the problems relating to the NOxemissions of circulating fluidized bed boilers in power plants and improved was the artificial swarm algorithm. In combination with the least square supporting vector machine,the authors established a model for the NOxemissions of boilers and optimized the adjustable parameters of the boiler and reduced the NOxemissions concentration. A comparison of the improved artificial swarm algorithm with the basic artificial swarm algorithm and the particle colony algorithm indicates that the model based on the improved artificial swarm algorithm can predict very well the NOxemissions concentration and boasts a very strong identification and generalization ability,and at the same time,it also indicates that the improved artificial swarm algorithm is quick in calculation and has an edge in optimizing data. Through a simulation test,the optimized NOxemissions concentration can obviously decrease,displaying its practical value in engineering applications.

关 键 词：锅炉燃烧优化 最小二乘支持向量机 NOx排放浓度 人工蜂群算法 

分 类 号：TK16[动力工程及工程热物理—热能工程] TP273[自动化与计算机技术—检测技术与自动化装置]