Mechanical Operation Reliability Assessment Based on Nonlinear Subspace Similarity Analysis Using Condition Monitoring Information
Operation reliability assessment is very meaningful for keeping the safety of mechanical equipment.An operation reliability assessment method based on nonlinear subspace identification and subspace similarity analysis is proposed in this paper.The aim of the proposed method is to assess the reliability of single mechanical equipment in time.The proposed approach has advantage over the conventional reliability assessment method in that it is with no need for historical failure samples and only the condition information is needed.Three main steps are included in the proposed approach.Firstly,condition monitoring information is utilized to form the kernel Hankel matrix.Then,the subspace identification method is performed to obtain nonlinear characteristic subspace of the kernel Hankel matrix.Finally,similarity of the subspaces from normal state and current states is calculated and the operation reliability index is defined by using the subspace similarity.The proposed method is validated by a case study and the positive results show the effectiveness of the method which offers a valid way for operation reliability assessment of the mechanical equipment.
operation reliability condition monitoring information kernel Hankel matrix nonlinear subspace subspace similarity reliability index
Chuang Sun Zhengjia He Zhousuo Zhang Yanyang Zi Bing Li Hongrui Cao Wenrong Xiao
State Key Laboratory for Manufacturing Systems Engineering,Xian Jiaotong University,Xian 710049
国际会议
南京
英文
1-4
2013-08-20（万方平台首次上网日期，不代表论文的发表时间）