摘要
为了对火电厂磨煤机作出早期故障预测并有效判别其故障类型,提出了基于核主元分析(KPCA)和最小二乘支持向量机(LSSVM)的磨煤机故障诊断新方法,并采用该方法对某电厂的HP碗式中速磨煤机的故障特征数据进行了仿真试验.结果表明:该方法可提取变量的特征信息,以有效地捕捉变量间的非线性关系,从而能有效地处理故障征兆与故障类型之间的不确定性,具有很好的分辨力,而且故障诊断的正确率很高.
In order to obtain prediction of mill faults, as well as the faults type, a new fault diagnosis method based on kernel principal component analysis （KPCA） and least square support vector machines （LSSVM） was given. A simulation by this method was carried out based on fault feature data of the mill. Results show that variables characteristic can be extracted effectively to describe the nonlinear relationship of original datasets. Indeterminacy between fault symptom and fault type can be disposed availably, and the method has good discrimination capability and right fault diagnosis.
出处
《动力工程》
CAS
CSCD
北大核心
2009年第2期155-158,共4页
Power Engineering
关键词
中速磨煤机
故障诊断
最小二乘支持向量机
核主元分析
medium speed mill
fault diagnosis
least square support vector machines
kernel principal component analysis
