Abstract：The advantages and disadvantages of the extracted features in the life prediction model based on deep learning need to be indirectly evaluated by the final prediction accuracy. Feature resolvability is poor. To this end, a new evaluation index of life prediction characteristics is proposed. It is different from the classic time correlation, robustness and monotony indicators. This indicator aims at the characteristics of the large dispersion of product life. It characterizes its advantages and disadvantages by evaluating the trend consistency of the characteristics between different samples. First, normalization and down-sampling methods are used to process the data. Secondly, the correlation formula is used to calculate the correlation value of the same feature between two bearings. Finally, the score is averaged to obtain the score of the indicator. A bearing data set is used as an example. The proposed evaluation index and classic evaluation index are used to evaluate and filter the signal features extracted by the deep learning model. And the selected features are used to predict the remaining service life of bearings. The experimental results based on the leave-one-out method show that: Among the 17 groups of experimental samples, there are 11 groups of samples that use the proposed index to screen the characteristics of the prediction results, which are better than the prediction results of no feature selection and the three classic evaluation indicators of time correlation, robustness and monotonicity. Among them, the obtained comprehensive average of the root mean square error was reduced by 21.0%, 27.6%, 25.8% and 19.5% respectively. Using this evaluation index is helpful to improve the interpretability of the features extracted by the deep learning model.
曾大懿，蒋雨良，邹益胜，张笑璐，李海浪. 一种新的轴承寿命预测特征评价指标的构建与验证[J]. 振动与冲击, 2021, 40(22): 18-27.
ZENG Dayi，JIANG Yuliang，ZOU Yisheng，ZHANG Xiaolu，LI Hailang. Construction and verification of a new evaluation index for bearing life prediction characteristics. JOURNAL OF VIBRATION AND SHOCK, 2021, 40(22): 18-27.
[1] Lei Y G, Li N P, Guo L, et al. Machinery health prognostics: A systematic review from data acquisition to RUL prediction[J]. Mechanical Systems and Signal Processing, 2018, 104: 799-834.
[2] Ali J B, Fnaiech N, Saidi L, et al. Application of empirical mode decomposition and artificial neural network for automatic bearing fault diagnosis based on vibration signals[J]. Applied Acoustics,2015,89(3):16-27.
[3] Pan D, Liu J B, Cao J. Remaining useful life estimation using an inverse Gaussian degradation model[J]. Neurocomputing, 2016, 185:64-72.
[4] Lei Y, Li N, Gontarz S. et al. A Model-Based Method for Remaining Useful Life Prediction of Machinery[J]. IEEE Transactions on Reliability, 2016, 65(3): 1314-1326.
[5] Li N, Lei Y, Lin J, et al. An Improved Exponential Model for Predicting Remaining Useful Life of Rolling Element Bearings[J]. IEEE Transactions on Industrial Electronics, 2015, 62(12): 7762-7773.
[6] 何群,李磊,江国乾,等.基于PCA和多变量极限学习机的轴承剩余寿命预测[J].中国机械工程,2014,25(07):984-989.
HE Qun, LI Lei, JIANG Guo-qian, et al. Residual life predictions for bearings based on PCA and MELM [J]. China Mechanical Engineering,2014,25(07):984-989.
[7] 肖婷,汤宝平,秦毅,等.基于流形学习和最小二乘支持向量机的滚动轴承退化趋势预测[J].振动与冲击,2015,34(09):149-153.
XIAO Ting, TANG Bao-ping, QIN Yi, et al. Degradation trend prediction of rolling bearing based on manifold learning and least squares support vector machine [J]. Vibration and Shock,2015,34(09):149-153.
[8] Liu Y B, He B, Liu F, et al. Remaining Useful Life Prediction of Rolling Bearings Using PSR, JADE, and Extreme Learning Machine[J]. Mathematical Problems in Engineering, 2016, 2016:1-13.
[9] Zhang B, Zhang L, Xu J. Degradation Feature Selection for Remaining Useful Life Prediction of Rolling Element Bearings[J]. Quality and Reliability Engineering International, 2016, 32(2): 547-554.
[10] Guo L, Li N, Jia F, et al. A recurrent neural network based health indicator for remaining useful life prediction of bearings[J]. Neurocomputing, 2017, 240: 98-109.
[11] Wang F, Liu X, Deng G, et al. Remaining Life Prediction Method for Rolling Bearing Based on the Long Short-Term Memory Network[J]. Neural Processing Letters, 2019.
[12] 康守强,叶立强,王玉静,等.基于MCEA-KPCA和组合SVR的滚动轴承剩余使用寿命预测[J].电子测量与仪器学报,2017,31(09):1365-1371.
KANG Shou-qiang, YE Li-qiang, WANG Yu-jing, et al. Remaining useful life prediction of rolling bearing based on MCEA-KPCA and combined SVR [J]. Journal of Electronic Measurement and Instrument,2017,31(09):1365-1371.
[13] 柏林,闫康,刘小峰.基于状态追踪特征相空间重构的轴承寿命预测方法[J].振动与冲击,2019,38(23):119-125.
BAI Lin, YAN Kang, LIU Xiao-feng. Bearing life prediction method based on phase space reconstruction of state tracking features [J]. Vibration and Shock,2019,38(23):119-125.
[14] 谷广宇,刘建敏,乔新勇,等.基于特征评价的发动机寿命预测方法研究[J].汽车工程,2020,42(01):108-113+120.
GU Guang-yu, LIU Jian-min, QIAO Xin-yong, et al. Research on Engine Life Prediction Method Based on Characteristics Evaluation [J]. Automotive Engineering,2020,42(01):108-113+120
[15] 刘胜兰,高凌寒,杜剑维,等.基于自适应顺序的滚动轴承最优特征选取与寿命预测[J].舰船科学技术,2019,41(21):71-76.
LIU Sheng-lan, GAO Ling-han, DU Jian-wei, et al. Optimal feature selection and life prediction of rolling bearings based on adaptive sequence[J]. Ship Science and Technology,2019,41(21):71-76.
[16] 曾大懿,杨基宏,邹益胜,等.基于并行多通道卷积长短时记忆网络PMCCNN-LSTM的轴承寿命预测方法[J].中国机械工程,2020, 31(20): 2454-2462.
ZENG Dayi, YANG Jihong, ZOU Yisheng, et al. Bearing life prediction method based on PMCCNN-LSTM [J].
China Mechanical Engineering ,2020, 31(20): 2454-2462.
[17] 张继冬,邹益胜,邓佳林,等.基于全卷积层神经网络的轴承剩余寿命预测[J].中国机械工程,2019,30(18):2231-2235.
ZHANG Jidong, ZOU Yisheng, DENG Jialin, et al Bearing remaining life prediction based on full convolutional layer neural networks [J]. China Mechanical Engineering, 2019, 30(18): 2231-2235.
[18] Nectoux P, Gouriveau R, Medjaher K, et al. PRONOSTIA: An experimental platform for bearings accelerated degradation tests[C]. IEEE International Conference on Prognostics and Health Management, 2012.
