针对滚动轴承异常检测准确性差、精度低及数据维度灾难造成检测困难等问题,提出一种基于随机矩阵特征值之差指标的滚动轴承状态异常检测算法。运用平移时间窗对不同时刻的轴承信息锁定,并通过分段、随机化、扩增和维度重构等方法构造出高维随机特征矩阵;利用随机矩阵理论对高维数据良好的处理能力,给出了滚动轴承特征值之差指标的构造方法及动态检测阈值的数学公式,可降低噪声的干扰,提高检测指标的鲁棒性与检测结果的准确性。采用IMS滚动轴承全寿命数据进行应用研究,分析了不同误警率对检测结果的影响;从指标构建、阈值设定及异常检测等方面,将特征值之差算法与特征值之比算法进行比较。结果表明,最大最小特征值之差算法中检测指标构建及阈值设定更符合实际工况,对滚动轴承异常状态检测更准确,对早期异常状态的识别更敏感。
Abstract
A rolling bearing state anomaly detection algorithm based on the difference index of random matrix eigenvalues is proposed to solve problems of poor accuracy, low precision and difficulty of detection caused by data dimension disasters in rolling bearing anomaly detection. The bearing information at different times was locked by using moving time window, and a high-dimensional random feature matrix was constructed through methods such as segmentation, randomization, amplification and dimensional reconstruction. The use of random matrix theory has a good processing ability of high-dimensional data, and the construction method of the difference index of the rolling bearing eigenvalues and the mathematical formula of the dynamic detection threshold were provided, which can reduce the interference of noise, improve the robustness of the detection index and the accuracy of the detection result. Using IMS rolling bearing full-life data for application research, the impact of different false alarm rates on the detected results were analyzed; From the perspective of index construction, threshold setting and abnormal detection, the difference between the eigenvalue algorithm and the eigenvalue ratio algorithm were compared. The results show that the construction of the detection index and threshold setting in the algorithm of the difference between the maximum and minimum eigenvalues are more in line with the actual working state, more accurate detection of abnormal state of rolling bearings, and more sensitive to the identification of early abnormal state.
关键词
滚动轴承 /
随机矩阵理论 /
异常检测 /
检测阈值 /
特征值之差
{{custom_keyword}} /
Key words
rolling bearing /
random matrix theory /
anomaly detection /
detection threshold /
difference between eigenvalues
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 朱丹宸,张永祥,潘洋洋,等.基于多传感器信号和卷积神经网络的滚动轴承故障诊断[J].振动与冲击,2020,39(04):172−178.
ZHU Danchen, ZHANG Yongxiang, PAN Yangyang, et al. Fault diagnosis for rolling element bearings based on multi-sensor signals and CNN[J]. Journal of Vibration and Shock, 2020, 39(04):173−178.
[2] 陶洁,刘义伦,付卓,等.基于Teager能量算子和深度置信网络的滚动轴承故障诊断[J].中南大学学报(自然科学版),2017,48(01):61−68.
TAO Jie, LIU Yilun, FU Zhuo, et al. Fault damage degrees diagnosis for rolling bearing based on Teager energy operator and deep belief network[J]. Journal of Central South University (Science and Technology),2017, 48(01):61−68.
[3] Leite G D N P, Araújo, Alex Maurício, et al. Entropy measures for early detection of bearing faults[J]. Physica A Statistical Mechanics & Its Applications, 2019, 514:458−472.
[4] 刘志亮,刘仕林,李兴林,等.滚动轴承安全域建模方法及其在高速列车异常检测中的应用[J].机械工程学报,2017,53(10):116−124.
LIU Zhiliang, LIU Shilin, LI Xinglin, et al. Safety domain modelling of rolling bearings and its application to anomaly detection for high-speed rail vehicles[J].Journal of Mechanical Engineering,2017,53(10):116−124.
[5] 张西宁,张雯雯,周融通,等.采用单类随机森林的异常检测方法及应用[J].西安交通大学学报,2020,54(02):1−8+157.
ZHANG Xining, ZHANG Wenwen, ZHOU Rongtong, et al. Anomaly Detection Method Based on One-Class Random Forest with Applications[J]. Journal of Xi’an Jiaotong University, 2020,54(02):1−8+157.
[6] Mao W T, Tian S Y, Fan J J, et al. Online detection of bearing incipient fault with semi-supervised architecture and deep feature representation[J]. Journal of Manufacturing Systems, 2020, 55:179−198
[7] Zhang Q Y, Wan S H, Wang B , et al. Anomaly detection based on random matrix theory for industrial power systems[J]. Journal of Systems Architecture, 2019, 95:67-74.
[8] Ni G X, Chen J H, Wang H. Degradation assessment of rolling bearing towards safety based on random matrix single ring machine learning [J]. Safety Science, 2019, 118:403−408.
[9] Liu, H Y, Aue, A, Paul, D. On the Marcenko-Pastur Law for linear time series [J]. Annals of Statistics, 2015, 43(2): 675−712.
[10] Zeng Y , Liang Y C . Eigenvalue-based spectrum sensing algorithms for cognitive radio[J]. IEEE Transactions on Communications, 2009, 57(6):1784−1793.
[11] 刘威,张东霞,王新迎,等.基于随机矩阵理论的电力系统暂态稳定性分析[J].中国电机工程学报,2016,36(18):4854−4863+5109.
LIU Wei, ZHANG Dongxia, WANG Xinying, et al. Power system transient stability analysis based on random matrix theory[J]. Proceeding of the CSEE, 2016, 36(18):4854−4863+5109
[12] Dean Korošak, Rupnik M S . Random matrix analysis of Ca2+ signals in β-Cell collectives[J]. Frontiers in Physiology, 2019, 10:1194.
[13] Federico P, Roberto G, Maurizio A S. Cooperative spectrum sensing based on the limiting eigenvalue ratio distribution in wishart matrices [J]. IEEE Communications Letters, 2009, 13(7):507−509.
[14] Chiani M. On the probability that all eigenvalues of Gaussian and Wishart random matrices lie within an interval [J]. IEEE Transactions on Information Theory, 2017,63(7): 4521−4531.
[15] Tracy C A, Widom H. Level spacing distributions and the Airy kernel [J]. Communications in Mathematical Physics, 1994, 159(1): 151−174.
[16] 夏均忠,郑建波,白云川,等.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, et al. Performance degradation status identification and assessment for rolling bearing based on NAP and RMI [J]. Journal of Vibration And Shock,2019,38(23):33-37.
[17] 倪广县,陈金海,王恒.滚动轴承高维随机矩阵状态异常检测算法[J].西安交通大学学报,2019,53(10):65−71.
NI Guangxian, CHEN Jinhai, WANG Heng. A state anomaly detection algorithm for rolling bearing based on high dimensional random matrix [J]. Journal of Xi’an Jiaotong University, 2019, 53(10):65−71.
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}