摘要
为了降低大型电厂锅炉NOx的排放,并对燃烧进行优化和控制,应用支持向量机算法建立了大型四角切圆燃烧电站锅炉NOx排放特性模型,利用NOx排放的热态实炉试验数据对模型进行了训练和校验,并对支持向量机算法模型中的参数g和C的选择进行了较深入的探讨,定性地分析了模型参数g和C的变化对模型预测能力的影响,获得了最佳的模型参数.利用该模型对不同实验工况下NOx的排放作出了预测,结果说明采用支持向量机算法建模达到了比较准确的预测效果,与其他建模方法相比具有泛化能力好、计算速度快的优点.
In order to reduce the NOx emission and optimize the burning process of high capacity power station boiler, a support vector machine model predicting the NOx emission of a high capacity boiler was developed and verified with experimental data of NOx emission characteristics of that boiler. How to select the model's parameter g and C was discussed, and the effects of these two parameters on model's performance were analyzed. Good predicting performance was achieved with the proper learning parameters. The modeling results showed that support vector machine is a good tool for building combustion models and has better generalization ability and higher calculation speed compared with other modeling approaches.
出处
《浙江大学学报(工学版)》
EI
CAS
CSCD
北大核心
2006年第10期1787-1791,共5页
Journal of Zhejiang University:Engineering Science
基金
国家自然科学基金资助项目(60534030
50576081)
关键词
锅炉
NOx
支持向量机
预测
boiler
NOx, support vector machine
prediction