Abstract:Extraction of degradation state features is the key for identification and evaluation of rolling bearing degradation status.Nuisance attribute projection (NAP) can be used to overcome shortcomings of traditional methods, and accurately extract characteristics of rolling bearing degraded status, but its monotonicity and sensitivity are poor in the whole life duration.Ranking mutual information (RMI) can be used for NAP’s optimization to accurately evaluate bearing degradation status.Here, the optimized orthogonal match pursuing (OOMP) was used to denoise vibration signals.The feature vector PE value calculated using NAP was compared with the reference PE value to identify bearing degradation status.RMI was used to enhance PE value’s sensitivity to subtle changes in signals and its monotonicity in the whole life duration to accurately assess bearing degradation status.The tests showed that after using NAP and RMI, the recognition rate of rolling bearing performance degradation status is high; if using NAP and RMI, bearing performance degradation status can be evaluated with high precision and in stages.
夏均忠,郑建波,白云川,吕麒鹏,杨刚刚. 基于NAP和RMI的滚动轴承性能退化状态识别与评估[J]. 振动与冲击, 2019, 38(23): 33-37.
XIA Junzhong, ZHENG Jianbo, BAI Yunchuan, L Qipeng, YANG Ganggang. Performance degradation status identification and assessment for rolling bearing based on NAP and RMI. JOURNAL OF VIBRATION AND SHOCK, 2019, 38(23): 33-37.
[1] 夏均忠, 于明奇, 黄财, 等. 基于VMD和Infogram的滚动轴承故障特征提取[J]. 振动与冲击, 2017, 36(22):111-117.
Xia Jun-zhong, Yu Ming-qi, Huang Cai, et al. Fault feature extraction of rolling element bearing based on VMD and Infogram[J]. Journal of Vibration and Shock, 2017, 36(22):111-117.
[2] Georgoulas G , Mustafa M O , Tsoumas I P , et al. Principal Component Analysis of the start-up transient and Hidden Markov Modeling for broken rotor bar fault diagnosis in asynchronous machines[J]. Expert Systems with Applications, 2013, 40(17):7024-7033.
[3] Cerrada M , Sánchez, René-Vinicio, Li C , et al. A review on data-driven fault severity assessment in rolling bearings[J]. Mechanical Systems and Signal Processing, 2018, 99:169-196.
[4] Tomoaki N, Takayuki N , Daichi M , et al. Segmenting Continuous Motions with Hidden Semi-markov Models and Gaussian Processes[J]. Frontiers in Neurorobotics, 2017, 11:67-.
[5] Pavle Boskoski, Matej Gasperin, Dejan Petelin, et al. Bearing fault prognostics using Rényi entropy based features and G90510aussian process models[J]. Mechanical Systems and ssignal Processing, 2018, 52-53:327-337.
[6] Machlica L , Zbyněk Zajíc. Factor Analysis and Nuisance Attribute Projection Revisited[J]. Interspeech. 2012:1570-1573.
[7] Schaffernicht E , Christoph Möller, Debes K , et al. Forward Feature Selection Using Residual Mutual Information[C]. European Symposium on Esann. DBLP, 2012.
[8] Zhang X , Liu Z , Miao Q , et al. Bearing fault diagnosis using a whale optimization algorithm-optimized orthogonal matching pursuit with a combined time–frequency atom dictionary[J]. Mechanical Systems and Signal Processing, 2018, 107:29-42.
[9] Huang W , Cheng J , Yang Y . Rolling bearing fault diagnosis and performance degradation assessment under variable operation conditions based on nuisance attribute projection[J]. Mechanical Systems and Signal Processing, 2019, 114:165-188.
[10] 闫晓玲, 董世运, 徐滨士. 基于最优小波包Shannon熵的再制造电机转子缺陷诊断技术[J]. 机械工程学报, 2016, 52(4):7-12.
Yan Xiao-ling, Dong Shi-yun, Xu Bin-shi. Flaw diagnosis technology for remanufactured motor rotor based on optimal wavelet packet Shannon entropy[J]. Journal of Mechanical Engineering, 2016, 52(4):7-12.
[11] Kumar V, Bawa V S. No reference image quality assessment metric based on regional mutual information among images[J]. 2019.
[12] Center for Intelligent Maintenance Systems[EB/OL].