摘要
支持向量机应用于故障诊断是近年来研究的热点,在支持向量机算法的基础上,以汽轮机故障为例,引入了K-L变换对故障特征进行提取。结果表明,经K-L变换后的支持向量机算法能够保证故障信息的完整性,有效识别临界故障状态,提高了故障的分类精度,扩展了支持向量机的应用范畴。
The application of support vector machine in fault diagnosis is the research hotspots in recent years. This paper first researches the algorithm of support vector machine, and introduces K - L transform method to extract the characteristic of the diagnosis of turbine. The result indicates that the algorithm of support vector machine based on K - L transform can ensure the integrality of diagnosis characteristic, and effectively recognize the critical diagnosis, which improves the precision of classification and extends thee application bound of support vector machine.
出处
《汽轮机技术》
北大核心
2007年第2期148-150,共3页
Turbine Technology
基金
华北电力大学博士教师学位科研基金(92104392)资助
关键词
支持向量机
K—L变换
特征提取
故障诊断
support vector machine
K - L transform
feature extraction
fault diagnosis
