摘要
针对燃煤电厂SCR脱硝反应器入口NO_x质量浓度测量的不准确问题,提出了基于KPCA(核主元分析)和LSSVM(最小二乘支持向量机)的脱硝反应器入口NO_x质量浓度的软测量方法。首先对KPCA的输入样本进行了多函数样本优化,减小了样本冗余,然后经KPCA对样本数据进行了降维处理,再利用LSSVM对提取出的主元进行训练,建立了系统的软测量模型。由仿真结果可以看出,最后得到的测量对象预测值与实测值的总体误差较小,该方法有助于解决脱硝反应器入口NO_x质量浓度的准确测量问题。
Aimed at the low accuracy and reliability of NO_x mass concentration in the SCR reactor inlet, a method for soft sensor of NO_x content based on KPCA and LSSVM was presented. Firstly, used the method of similarity function optimization to simplify the input samples and reduce the redundant information. In order to improve the accuracy of NO_x contents, auxiliary parameters that connected to NO_x were analyzed. The least square support vector machine（LSSVM） with the kernel principal component analysis was used to analyze the related parameters and established the soft measurement model. The simulation results showed that the overall error was small between the predictive value and the actural measured value, the method could help to remove the measurement accurate of denitrification reactor inlet NO_x concentration.
出处
《内蒙古电力技术》
2016年第1期73-77,89,共6页
Inner Mongolia Electric Power
基金
河北省自然科学基金资助项目(F2014502059)
关键词
脱硝反应器
NO_x质量浓度
样本优化
核主元分析
最小二乘支持向量机
软测量
denitrification reactor
NO x concentration
sample optimization
kernel principal component analysis（KPCA）
least square support vector machine（LSSVM）
soft sensor
