燃料成分分析及基于LS-SVM对锅炉效率的预测  

Component analysis of fuel and prediction of boiler efficiency based on Least Square-support Vector Machine

作　　者：李艳[1] 奚溪[1] 徐菲菲[1] 梅宁[1] 

机构地区：[1]中国海洋大学工程学院,山东青岛266100

出　　处：《热科学与技术》2014年第1期56-60,共5页Journal of Thermal Science and Technology

基　　金：国家自然科学基金资助项目(51276174);青岛市科技发展计划资助项目(11-2-3-69-nsh)

摘　　要：利用回归分析的方法,根据实验数据,拟合出灰分,挥发分,全硫与发热量之间的线性系数,确定成分相关性。基于最小二乘支持向量机(least square-spport vector machine,LS-SVM)建立了电站锅炉能源消耗及排放模型,实现了对排烟温度、飞灰含碳质量分数等模型参数的软测量以及对锅炉效率的预测。Both theoretical and experimental investigations were made to obtain the linear coefficient between the ash content,volatiles,total sulfur and calorific value.The regression analysis method was introduced,and relationships between ingredients were acquired.Besides,an energy consumption and emissions model for boiler was proposed based on the least square-support vector machine （LS-SVM）.Parameters such as the exhaust flue gas temperature,carbon content of fly ash,as well as the efficiency of boiler were predicted successfully.

关 键 词：回归分析 最小二乘支持向量机 锅炉效率 

分 类 号：TK229.6[动力工程及工程热物理—动力机械及工程]