Research on service life prediction on rolling bearings based on vibration signal analysis

doi:10.3969/j.issn.1674-1951.2021.05.006

Abstract

Abstract:

The prediction on residual life of rolling bearings is complex because of the multiple categories and multiple features.It is difficult to apply traditional prediction methods based on mechanics and probability statistics on engineering practices.In the following study,the service life of a rolling bearing was predicted based on vibration signal analysis.Firstly,feature extraction in time and frequency domains was made on the collected vibration signals and wavelet packet sample entropy.Then,the features with high correlation to the bearing's life span were selected by Pearson correlation analysis as the sample set of bearing life prediction.The selected characteristic parameters were taken as the inputs training the improved particle swarm optimization-General Regression Neural Network（PSO-GRNN） model to construct the bearing life prediction model.Comparing the results made by this bearing life prediction model,back propagation（BP） Neural Network model and PSO-GRNN model,the proposed model is verified to be more stable and accurate.

Key words: feature extraction, correlation analysis, GRNN, particle swarm optimization algorithm, BP Neural Network, rolling bearing, life prediction

CLC Number:

TH133.33：TK323

TAN Zhiling, CHEN Caiming, XU Shengchao, WU Zhihong, SONG Yin, WANG Pengfei. Research on service life prediction on rolling bearings based on vibration signal analysis[J]. Huadian Technology, 2021, 43(5): 36-44.

Figures/Tables 14

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Fig.8

Fig.9

Tab.1

Fig.10

Fig.11

Fig.12

Tab.2

References 16

[1]	沈长青. 旋转机械设备关键部件故障诊断与预测方法研究[D]. 合肥:中国科学技术大学, 2014.
[2]	张小丽, 陈雪峰, 李兵, 等. 机械重大装备寿命预测综述[J]. 机械工程学报, 2011,47(11):100-116.
	ZHANG Xiaoli, CHEN Xuefeng, LI Bing, et al. Review of life prediction for mechanical major equipments[J]. Journal of Mechanical Engineering, 2011,47(11):100-116.
[3]	PARIS P C, ERDOGAN F. A critical analysis of crack propagation laws[J]. Journal of Basic Engineering, 1963,85(4):528-534. doi: 10.1115/1.3656900
[4]	ANDREIKIV O E, LESIV R M, LEVYTS'KA N M. Crack growth in structural materials under the combined action of fatigue and creep (review)[J]. Materials Science, 2009,45(1):1-17. doi: 10.1007/s11003-009-9160-0
[5]	MAKKONEN M. Predicting the total fatigue life in metals[J]. International Journal of Fatigue, 2009,31(7):1163-1175. doi: 10.1016/j.ijfatigue.2008.12.008
[6]	TALLIAN T. Weibull distribution of rolling contact fatigue life and deviations therefrom[J]. ASLE Trans, 1962,5(5):183-196. doi: 10.1080/05698196208972465
[7]	ZHAO Y X, LIANG H. Modeling of the probabilistic fatigue S-N curves using the two parameter Weibull distribution[J]. Journal of Mechanical Engineering, 2015,51(20):208-212. doi: 10.3901/JME.2015.20.208
[8]	唐云冰, 高德平, 罗贵火. 航空发动机高速滚珠轴承力学特性分析与研究[J]. 航空动力学报, 2006(2):354-360.
	TANG Yunbing, GAO Deping, LUO Guihuo. Research of aero-engine high-speed ball bearing[J]. Journal of Aerospace Power, 2006(2):354-360.
[9]	LU C, CHEN J, HONG R, et al. Degradation trend estimation of slewing bearing based on LSSVM model[J]. Mechanical Systems & Signal Processing, 2016, 76-77:353-366.
[10]	CHOI S, PAZOUKI E, BAEK J, et al. Iterative condition monitoring and fault diagnosis scheme of electric motor for harsh industrial application[J]. IEEE Transactions on Industrial Electronics, 2015,62(3):1760-1769. doi: 10.1109/TIE.2014.2361112
[11]	邱晓梅, 隋文涛, 王峰, 等. 基于相关系数和BP神经网络的轴承剩余寿命预测[J]. 组合机床与自动化加工技术, 2019(4):63-65.
	QIU Xiaomei, SUI Wentao, WANG Feng, et al. Remaining life prediction of bearing based on correlation coefficient and BP neural network[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2019(4):63-65.
[12]	张永建, 孙燕芳, 邢龙超. 基于小波包分解的滚动轴承故障诊断方法的研究[J]. 煤矿机械, 2014,35(5):256-258.
	ZHANG Yongjian, SUN Yanfang, XING Longchao. Research of rolling bearing fault diagnosis based on wavelet-packet decomposition method[J]. Coal Mine Machinery, 2014,35(5):256-258.
[13]	何勇, 张祥金, 姚宗辰. 改进样本熵最优小波包阈值选择算法在信号降噪中的应用[J]. 兵器装备工程学报, 2019,40(3):149-154.
	HE Yong, ZHANG Xiangjin, YAO Zongchen. Improved optimal threshold selection algorithm applied to denoising[J]. Journal of Ordnance Equipment Engineering, 2019,40(3):149-154.
[14]	车远宏, 贾雍, 汤卓, 等. 皮尔逊相关系数在风电功率组合预测中的应用[J]. 广西电力, 2016,39(3):50-53.
	CHE Yuanhong, JIA Yong, TANG Zhuo, et al. Application of pearson correlation coefficient in wind power combination prediction[J]. Guangxi Electric Power, 2016,39(3):50-53.
[15]	张成龙, 郑凯, 刘杰. 基于小波包能量谱和改进FOA-GRNN的轴承寿命预测[J]. 组合机床与自动化加工技术, 2020(7):73-76,80.
	ZHANG Chenlong, ZHENG Kai, LIU Jie. Prediction method of bearing remaining useful life based on wavelet packet energy spectrum and improved FOA-GRNN[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2020(7):73-76,80.
[16]	金亚军, 张丹, 隋文涛, 等. 基于粒子群优化神经网络的滚动轴承剩余寿命预测[J]. 组合机床与自动化加工技术, 2020(8):64-66,70.
	JIN Yajun, ZHANG Dan, SUI Wentao, et al. Prediction of rolling bearing residual life based on particle swarm optimization and neural network[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2020(8):64-66,70.

工况	转速/(r·min^-1)	载荷/kN
1	2 100	12
2	2 250	11
3	2 400	10

工况	BP神经网络模型	PSO-GRNN模型	改进后的PSO-GRNN模型
1	6.08	4.77	2.31
2	6.75	4.61	2.19
3	5.84	5.08	2.01