基于KPCA和LSSVM的过热器异常诊断  

Diagnosis of Superheater Anomaly Based on KPCA and LSSVM

作　　者：陈国超 张悦[1,2] 卢聪 李帅 姜礼洁[1,2] CHEN Guochao;ZHANG Yue;LU Cong;LI Shuai;JIANG Lijie(North China Electric Power University,Baoding 071003,China;Hebei Technology Innovation Center of Simulation&Optimized Control for Power Generation,North China Electric Power University,Baoding 071003,China)

机构地区：[1]华北电力大学自动化系,河北保定071003 [2]华北电力大学河北省发电过程仿真与优化控制技术创新中心,河北保定071003

出　　处：《山东电力技术》2021年第2期63-66,共4页Shandong Electric Power

摘　　要：为提高过热器系统异常识别准确率,对核主元分析以及最小二乘支持向量机在过热器异常识别分类中的应用进行研究,提出了一种改进KPCA-LSSVM的过热器异常工况识别策略。采用核主元分析算法对获取到的过程数据提取主成分,并选择贡献度最大的主成分输入LSSVM中进行建模,建立KPCA-PSO-LSSVM分类模型对主汽温故障进行识别。结果表明,该模型准确率有所提高。In order to improve the accuracy of superheater system anomaly recognition,the application of kernel principal compo⁃nent analysis(KPCA)and least squares support vector machine(LSSVM)in superheater anomaly recognition and classification was studied,and an improved KPCA⁃least squares support vector machine(LSSVM)superheater anomaly recognition strategy was proposed.The KPCA algorithm was used to extract principal components from the process data.The principal component with the largest contribution was selected to input into LSSVM for modeling,and the KPCA-PSO-LSSVM classification model was estab⁃lished to identify the main air temperature fault.The results show that the accuracy of the model is improved.

关 键 词：核主元分析 最小二乘支持向量机 过热器异常 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]