基于多通道融合及贝叶斯理论的刀具剩余寿命预测方法

doi:10.3901/JME.2021.13.214

机械工程学报 ›› 2021, Vol. 57 ›› Issue (13): 214-224.doi: 10.3901/JME.2021.13.214

扫码分享

基于多通道融合及贝叶斯理论的刀具剩余寿命预测方法

王艺玮, 邓蕾, 郑联语, 王亚辉

北京航空航天大学机械工程及自动化学院北京 100191

收稿日期:2020-07-15 修回日期:2020-12-10 出版日期:2021-07-05 发布日期:2021-08-31
通讯作者: 郑联语(通信作者),男,1967年出生,博士,教授,博士研究生导师。主要研究方向为数字化与智能制造技术、柔性智能工艺装备、先进测量与质量控制、可穿戴和AR辅助智能装配技术、闭环自适应智能加工。E-mail:lyzheng@buaa.edu.cn
作者简介:王艺玮,女,1988年出生,博士,讲师,博士研究生导师。主要研究方向为工业智能,复杂装备智能运维系统,设备故障诊断,剩余寿命预测及健康管理。E-mail:wangyiwei@buaa.edu.cn
基金资助:
国家自然科学基金资助项目（51805262）

A Multi-channel Signal Fusion and Bayesian Theory Based Method for Tool Remaining Useful Life Prediction

WANG Yiwei, DENG Lei, ZHENG Lianyu, WANG Yahui

School of Mechanical Engineering and Automation, Beihang University, Beijing 100191

Received:2020-07-15 Revised:2020-12-10 Online:2021-07-05 Published:2021-08-31

摘要/Abstract

摘要： 刀具监测及可用剩余寿命（RUL）预测对降本增效及保证加工质量意义重大。针对单一传感器预测精度波动大、数据利用率低、可靠性低等问题，提出一种多通道信号融合及贝叶斯更新的刀具剩余寿命预测方法。通过计算多通道信号所提取特征的时间序列与对应时间矢量的斯皮尔曼等级相关系数对特征时序做单调性排序，取单调性得分高的特征用主成分分析进行融合并构建健康因子作为观测数据，基于贝叶斯理论及马尔科夫链蒙特卡洛采样估计退化模型参数，并随着时间推进及监测数据序贯可获，实时在线更新退化模型参数以逐渐逼近刀具磨损退化趋势，同时对每时刻剩余寿命进行迭代估计。所提方法可避免基于深度学习方法需要依赖大量全寿命数据离线训练预测模型且模型对新预测任务适应性有限的局限性。用PHM2010公开数据挑战赛中三槽球头硬质合金铣刀切削不锈钢过程磨损全寿命数据集验证了方法有效性。

关键词: 刀具RUL预测, 多通道信号融合, 贝叶斯理论, 马尔科夫链蒙特卡洛采样

Abstract: Tool monitoring and remaining useful life (RUL) prediction are significant for reducing costs and ensuring processing quality. Aiming at the problems of single-sensor prediction accuracy fluctuations, low data utilization, and low reliability, a multi-channel signal fusion and Bayesian update tool remaining life prediction method is proposed. By calculating the time series of the features extracted from the multi-channel signal and the Spearman rank correlation coefficient of the corresponding time vector, the feature time is monotonically ranked. Features with a high monotonicity score are fused using principal component analysis and the health factors are constructed as observation data based on Bayesian theory and Markov chain Monte Carlo sampling to estimate the degradation model parameters, and as time progresses and monitoring data is available sequentially, the degradation model parameters are updated online in real time to gradually approach the trend of tool wear degradation. The remaining life of each moment is estimated iteratively. The proposed method can avoid the limitation that deep learning-based methods need to rely on many full-life data to train the prediction model offline and the model has limited adaptability to new prediction tasks. Validate with the full life data set of the three-slot ball-end carbide milling cutter for cutting stainless steel in the open data challenge.

Key words: tool RUL prediction, multi-channel signal fusion, Bayesian theory, Markov chain Monte Carlo sampling

中图分类号:

TH17
TG506

王艺玮, 邓蕾, 郑联语, 王亚辉. 基于多通道融合及贝叶斯理论的刀具剩余寿命预测方法[J]. 机械工程学报, 2021, 57(13): 214-224.

WANG Yiwei, DENG Lei, ZHENG Lianyu, WANG Yahui. A Multi-channel Signal Fusion and Bayesian Theory Based Method for Tool Remaining Useful Life Prediction[J]. Journal of Mechanical Engineering, 2021, 57(13): 214-224.

参考文献

[1] 赵帅, 黄亦翔, 王浩任, 等. 基于随机森林与主成分分析的刀具磨损评估[J]. 机械工程学报, 2017, 53(21):181-189.ZHAO Shuai, HUANG Yixiang, WANG Haoren, et al. Random forest and principal components analysis based on health assessment methodology for tool wear[J]. Journal of Mechanical Engineering, 2017, 53(21):181-189.
[2] 雷亚国, 韩天宇, 王彪, 等. XJTU-SY滚动轴承加速寿命试验数据集解读[J]. 机械工程学报, 2019, 55(16):1-6.LEI Yaguo, HAN Tianyu, WANG Biao, et al. XJTU-SY rolling element bearing accelerated life test datasets:A tutorial[J]. Journal of Mechanical Engineering, 2019, 55(16):1-6.
[3] 张石平, 王智明, 杨建国. 机床刀具可靠性及寿命评估[J]. 计算机集成制造系统, 2015, 21(6):1579-1584.ZHANG Shiping, WANG Zhiming, YANG Jianguo. Tool reliability and life assessment for machine tools[J]. Computer Integrated Manufacturing Systems, 2015, 21(6):1579-1584.
[4] 裴洪, 胡昌华, 司小胜, 等. 基于机器学习的设备剩余寿命预测方法综述[J]. 机械工程学报, 2019, 55(8):1-13.PEI Hong, HU Changhua, SI Xiaosheng, et al. Review of machine learning based remaining useful life prediction methods for equipment[J]. Journal of Mechanical Engineering, 2019, 55(8):1-13.
[5] LI X, DING Q, SUN J Q. Remaining useful life estimation in prognostics using deep convolution neural network[J]. Reliability Engineering and System Safety, 2018, 72:1-11.
[6] LI X, ZHANG W, DING Q. Deep learning-based remaining useful life estimation of bearing using multi-scale feature extraction[J]. Reliability Engineering and System Safety. 2019, 182:208-218.
[7] YU W N, KIM II Y, MECHEFSKE C. Remaining useful life estimation using a bidirectional recurrent neural network based autoencoder scheme[J]. Mechanical Systems and Signal Processing, 2019, 129:764-780.
[8] WANG B, LEI Y G, LI N P et al. Deep separable convolutional network for remaining useful life prediction of machinery[J]. Mechanical Systems and Signal Processing, 2019, 134:1-18.
[9] 蔡伟立, 胡小锋, 刘梦湘. 基于迁移学习的刀具剩余寿命预测方法研究[J/OL]. (2019-11-29). 计算机集成制造系统. https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST&filename=JSJJ20191129007&v=HEpgAltAcr%25mmd2BdFXfCzJRiSCWfv2E7Lk5S06E%25mmd2FgjQF69oSffJNeq0ApFE2wAM1gR3L. CAI Weili, HU Xiaofeng, LIU Mengxiang. Research on prediction method of tool remaining useful life based on transfer learning[J]. (2019-11-29). Computer Integrated Manufacturing Systems. https://kns.cnki.net/kcms/detail/detail.aspx?dbcode=CAPJ&dbname=CAPJLAST&filename=JSJJ20191129007&v=HEpgAltAcr%25mmd2BdFXfCzJRiSCWfv2E7Lk5S06E%25mmd2FgjQF69oSffJNeq0ApFE2wAM1gR3L.
[10] WANG Y W, GOGU C, KIM N H, et al. Noise-dependent ranking of prognostics algorithms based on discrepancy without true damage information[J]. Reliability Engineering and System Safety, 2019, 184:86-100.
[11] HU Y, BARALDI P, ZIO E. A particle filtering and kernel smoothing-based approach for new design component prognostics[J]. Reliability Engineering and System Safety, 2015, 134:19-31.
[12] WANG Y W, GOGU C, BINAUD N, et al. A model-based prognostics method for fatigue crack growth in fuselage panels[J]. Chinese Journal of Aeronautics, 2019, 32(2):396-408.
[13] 王国峰, 董毅, 杨凯, 等. 基于深度学习与粒子滤波的刀具寿命预测[J]. 天津大学学报, 2019, 152(11):1109-1116.WANG Guofeng, DONG Yi, YANG Kai, et al. Tool life prediction based on deep learning and particle filtering[J]. Journal of Tianjin University, 2019, 152(11):1109-1116.
[14] 任子强, 司小胜, 胡昌华, 等. 融合多传感器数据的动机剩余寿命预测方法[J]. 航空学报, 2019, 40(12):1-12.REN Ziqiang, SI Xiaosheng, HU Changhua, et al. Remaining useful life prediction method for engine combining multi-sensors data[J]. Chinese Journal of Aeronautics, 2019, 40(12):1-12.
[15] Bayesian-MCMC-Prognostics[EB/OL].(2021-01-13). https://github.com/Andy-design-code/Bayesian-MCMC-Prognostics
[16] 胡昌华, 施权, 司小胜, 等. 数据驱动的寿命预测和健康管理技术研究进展[J]. 信息与控制, 2017, 46(1):72-82.HU Changhua, SHI Quan, SI Xiaosheng, et al. Data-driven life prediction and health management:state of the art[J]. Information and Control. 2017, 46(1):72-82.
[17] 赵申坤, 姜潮, 龙湘云. 一种基于数据驱动的贝叶斯理论的机械系统寿命寿命预测方法[J]. 机械工程学报, 2018, 54(12):115-124.ZHAO Shenkun, JIANG Chao, LONG Xiangyun. Remaining useful life estimation of mechanical systems based on the data-driven method and Bayesian theory[J]. Journal of Mechanical Engineering, 2018, 54(12):115-124.
[18] MOSALLAM A, MEDJAHER K, ZERHOUNI N. Data driven prognostic method based on Bayesian approaches for direct remaining useful life prediction[J]. Journal of Intelligent Manufacturing, 2016, 27(5):1037-1048.
[19] JOLLIFFE I T. Principal component analysis[M]. New York:Springer Press, 2002.
[20] FOSTER P. Exploring multivariate data using directions of high density[J]. Statistics and Computing, 1988, 8(4):347-355.
[21] LI N P, LEI Y G, 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.
[22] WANG Y H, DENG C, WU J, et al. Failure time prediction for mechanical device based on the degradation sequence[J]. Journal of Intelligent Manufacturing, 2015, 26:1181-1199.
[23] 朱新建, 陈学东, 吕宝林, 等. 基于多维高斯贝叶斯的机械设备失效/故障智能诊断及参数影响分析[J]. 机械工程学报, 2020, 56(4):35-41.ZHU Xinjian, CHEN Xuedong, LÜ Baolin, et al. Smart failure/fault diagnosis and influence analysis for mechanical equipment with multivariate Gaussian Bayesian method[J]. Journal of Mechanical Engineering. 2020, 56(4):35-41.
[24] KIM N H, AN D, CHOI J H. Prognostics and health management of engineering systems[M]. New York:Springer Press, 2017.
[25] PHM2010 conference data challenge[EB/OL]. (2010-05-18). http://www.phmsociety.org/competition/phm/10.
[26] LI X, LIM B S, ZHOU J H, et al. Fuzzy neural network modelling for tool wear estimation in dry milling operation[C]//Annual Conference of the prognostics and health management society, 2009.
[27] BENKEDJOUH T, MEDJAHER K, ZERHOUNI N, et al. Health assessment and life prediction of cutting tools based on support vector regression[J]. Journal of Intelligent manufacturing, 201526, 213-223.
[28] WU D, JENNINGS C, TERPENNY J, et al. Cloud-based parallel machine learning for tool wear prediction[J]. Journal of manufacturing science and engineering, 2018, 140(4):041005.
[29] SAXENA A, CELAYA J, SAHA B, et al. On applying the prognostic performance metrics[C]//Annual conference of the prognostic and health management society, 27 September-1 October 2009, San Diego, California, USA.

基于多通道融合及贝叶斯理论的刀具剩余寿命预测方法

A Multi-channel Signal Fusion and Bayesian Theory Based Method for Tool Remaining Useful Life Prediction

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价