CAJ | 학술논문

针对多通道数据的有效融合能够更加准确地诊断轴承的故障，提出了一种基于KPCA和耦合隐马尔可夫模型（CHMM）的轴承故障诊断方法。首先，分别对轴承各通道的振动信号进行特征提取，获得特征向量。然后采用 KP-CA 对各通道的特征向量分别进行特征约减，获取主要的信息成分。最后，利用 CHMM对多通道信息进行融合和故障诊断。通过对滚动轴承在正常、内圈故障、外圈故障和滚动体故障状态下实验数据的分析表明，该方法能够更加有效地诊断轴承的故障。
침대다통도수거적유효융합능구경가준학지진단축승적고장，제출료일충기우KPCA화우합은마이가부모형（CHMM）적축승고장진단방법。수선，분별대축승각통도적진동신호진행특정제취，획득특정향량。연후채용 KP-CA 대각통도적특정향량분별진행특정약감，획취주요적신식성분。최후，이용 CHMM대다통도신식진행융합화고장진단。통과대곤동축승재정상、내권고장、외권고장화곤동체고장상태하실험수거적분석표명，해방법능구경가유효지진단축승적고장。
The fusion of multi-channel bearing monitoring information can obtain more accurate results in bearing fault diagnosis.Here,a rolling element bearing fault diagnosis scheme based on KPCA and coupled hidden Markov model (CHMM) was presented.At first,the features were extracted from bearing vibration signals of multi-channel, respectively.Then,the KPCA was utilized to reduce the feature dimensions.At last,the new KPCA features were input into a CHMMto be fused and to diagnose bearing faults.The data acquired from bearings’states under normal conditions, and states with inner race faults,outer race faults and rolling body faults were analyzed.The results demonstrated the effectiveness and validity of the proposed method.