摘要
针对在液压泵故障诊断中故障样本难以获得的问题,融合人工免疫系统中的实值否定选择算法和支持向量机算法提出了一种混合的故障诊断方法。在该混合方法中使用算法产生非己集合(故障样本),将其作为算法的输入进行训练,解决了难以获得故障样本的难题。应用小波分析完成液压泵端盖振动信号的消噪及特征提取。最后用柱塞泵脱靴故障样本进行诊断,正确率可达90%,验证了混合诊断方法的有效性。
A hybrid fault diagnosis approach was proposed, combining the RNS （ real -- valued negative selection） algorithm and the support vector machine,because it was very difficult to gain the fault samples in the fault diagnosis process of hydraulic pump. In this method,the RNS algorithm was used to generate the nonself set as the fault samples,which were used as input to SVM algorithm for training purpose. The problem, lacking the fault samples, was solved using this new method. It was accomplished to cancel the interference existing in the measured signals of hydraulic pump and pick up its features using the wavelet analysis method. At last,the hydraulic pump fault samples were tested using the hybrid approach. The classification right rate of this method is 90% ,so it is valid to the fault diagnosis of hydraulic pump.
出处
《中国机械工程》
EI
CAS
CSCD
北大核心
2008年第14期1736-1739,1743,共5页
China Mechanical Engineering
基金
国家自然科学基金资助项目(50775198)
河北省教育厅博士基金资助项目(B2004128)
关键词
免疫算法
故障诊断
支持向量机
液压泵
小波分析
immune algorithm
fault diagnosis
support vector machine （SVM）
hydraulic pump
wavelet analysis
