考虑寿命预测误差的机械装备维护决策方法研究

doi:10.3901/JME.260304

机械工程学报 ›› 2026, Vol. 62 ›› Issue (13): 155-165.doi: 10.3901/JME.260304

扫码分享

考虑寿命预测误差的机械装备维护决策方法研究

雷亚国¹, 陈泽训¹, 杨彬¹, 武通海¹, 孔德同²

1. 西安交通大学现代设计及转子轴承系统教育部重点实验室西安 710049;
2. 华电电力科学研究院有限公司杭州 310030

收稿日期:2025-07-07 修回日期:2025-12-05 发布日期:2026-08-28
作者简介:雷亚国，男，1979年出生，教授，博士研究生导师。主要研究方向为智能故障诊断与寿命预测、机械装备运行维护决策优化。E-mail:yaguolei@mail.xjtu.edu.cn;杨彬(通信作者)，男，1992年出生，助理教授，硕士研究生导师。主要研究方向为高端装备大数据智能运维。E-mail：binyang@xjtu.edu.cn
基金资助:
教育部基础学科和交叉学科突破计划、国家自然科学基金(52435003，52305129)和中央高校基本科研业务费专项资金资助项目。

Research on Maintenance Decision-making Method of Machinery Considering Life Prediction Error

LEI Yaguo¹, CHEN Zexun¹, YANG Bin¹, WU Tonghai¹, KONG Detong²

1. Key Laboratory of Education Ministry for Modern Design and Rotor-Bearing System, Xi'an Jiaotong University, Xi'an 710049;
2. Huadian Electric Power Research Institute Co., Ltd., Hangzhou 310030

Received:2025-07-07 Revised:2025-12-05 Published:2026-08-28

摘要/Abstract

摘要： 预测性维护能够依据机械装备的健康状态退化趋势或剩余寿命等指标合理制定维护计划，为装备维护选择最佳时机，从而保障装备运行安全性并降低装备维护成本。然而，工程实际中预测性维护决策模型的性能依赖于寿命预测方法的精度，剩余寿命预测值偏大或者偏小都会使得装备运维错过最佳维护时机，严重制约维护决策模型的工程应用效果。因此，如何降低预测误差对维护策略的负面影响，成为机械装备智能运维领域中亟待解决的关键问题。考虑到真实剩余寿命显著影响寿命预测误差，本文提出了一种考虑寿命预测误差的非周期维护策略。首先对寿命预测误差进行建模，建立预测误差分布与真实剩余寿命之间的动态映射关系；然后将预测误差的影响分类为适时维护、意外维护和过早维护三种决策结果，并计算每种决策结果的维护成本；最后基于装备寿命分布构建非周期维护模型来生成非周期维护计划，并基于维护计划计算综合维护成本率，形成非周期维护决策模型。通过机床铣刀维护的案例对提出策略进行验证，结果表明提出方法能够在考虑预测误差的基础上合理安排检测计划，有效降低铣刀的维护成本。

关键词: 机械装备, 维护决策, 寿命预测误差, 非周期维护

Abstract: Predictive maintenance can be based on the health state degradation trends of the machinery or the remaining life of the machinery and other indicators to rationally develop the maintenance plan, to select optimal time for machinery maintenance, so as to ensure the reliability of the machinery and reduce the maintenance costs. However, in engineering practice, the performance of the predictive maintenance decision-making model heavily relies on the accuracy of the life prediction methods. Both overestimated and underestimated remaining life may cause the machinery operation and maintenance to miss the optimal time for maintenance, and severely restrict the engineering application effect of the maintenance decision-making model. Therefore, how to mitigate the negative impact of prediction errors on maintenance strategies has become a critical issue that urgently needs to be addressed in the field of intelligent maintenance for machinery. Considering the significant impact of real remaining useful life on life prediction errors, this paper proposes a non-periodic maintenance strategy considering life prediction errors. Firstly, a dynamic relationship between prediction error distribution and real remaining life is established through error modeling. Subsequently, the impacts of prediction errors are categorized into three decision outcomes - timely maintenance, unexpected maintenance and premature maintenance, with corresponding maintenance costs calculated for each scenario. Finally, a non-periodic maintenance model is developed based on machinery life distribution to generate maintenance schedules, with the integrated maintenance cost rate calculated accordingly, forming a non-periodic maintenance decision-making framework. The proposed strategy is validated through a case study on machine tool milling cutter maintenance. Results demonstrate that the methodology can effectively arrange inspection plans while considering prediction errors, achieving significant maintenance cost reduction.

Key words: machinery, maintenance decision, life prediction error, non-periodic maintenance

中图分类号:

TH17

雷亚国, 陈泽训, 杨彬, 武通海, 孔德同. 考虑寿命预测误差的机械装备维护决策方法研究[J]. 机械工程学报, 2026, 62(13): 155-165.

LEI Yaguo, CHEN Zexun, YANG Bin, WU Tonghai, KONG Detong. Research on Maintenance Decision-making Method of Machinery Considering Life Prediction Error[J]. Journal of Mechanical Engineering, 2026, 62(13): 155-165.

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: http://www.cjmenet.com.cn/CN/10.3901/JME.260304

http://www.cjmenet.com.cn/CN/Y2026/V62/I13/155

参考文献

[1] LEI Y，LI N，LI X. Big data-driven intelligent fault diagnosis and prognosis for mechanical systems[M]. Singapore：Springer，2023.
[2] 胡昌华，樊红东，王兆强. 设备剩余寿命预测与最优维修决策[M]. 北京：国防工业出版社，2018. HU Changhua，FAN Hongdong，WANG Zhaoqiang. Residual life prediction and optimal maintenance decision for a piece of equipment[M]. Beijing：National Defense Industry Press，2018.
[3] YU W，TU W，KIM I Y，et al. A nonlinear-drift-driven Wiener process model for remaining useful life estimation considering three sources of variability[J]. Reliability Engineering & System Safety，2021，212：107631.
[4] 谢志强，周伟，余泽睿. 动态调整设备维护开始时间的综合调度算法[J]. 机械工程学报，2021，57(4)：240-246. XIE Zhiqiang，ZHOU Wei，YU Zerui. Integrated scheduling algorithm for dynamic adjustment of equipment maintenance start time[J]. Journal of Mechanical Engineering，2021，57(4)：240-246.
[5] GRALL A，DIEULLE L，BéRENGUER C，et al. Continuous-time predictive-maintenance scheduling for a deteriorating system[J]. IEEE Transactions on Reliability，2002，51(2)：141-150.
[6] ALBERTI A R，CAVALCANTE C A V，SCARF P，et al. Modelling inspection and replacement quality for a protection system[J]. Reliability Engineering & System Safety，2018，176：145-153.
[7] ZHANG Y，OUYANG L，MENG X，et al. Condition-based maintenance considering imperfect inspection for a multi-state system subject to competing and hidden failures[J]. Computers & Industrial Engineering，2024，188：109856.
[8] DRIESSEN J P C，PENG H，VAN HOUTUM G J. Maintenance optimization under non-constant probabilities of imperfect inspections[J]. Reliability Engineering & System Safety，2017，165：115-123.
[9] JAIN A K，DHADA M，HERNANDEZ M P，et al. Influence of imperfect prognostics on maintenance decisions[C]//Proceedings of the 11th International Conference on Precision，Meso，Micro & Nano Engineering (COPEN 11)，December 12-14，2019，Indore，India. Indore：Indian Institute of Technology Indore，2019.
[10] 董青，郑建飞，胡昌华，等. 基于两阶段自适应Wiener过程的剩余寿命预测方法[J]. 自动化学报，2022，48(02)：539-553. DONG Qing，ZHENG Jianfei，HU Changhua，et al. Remaining useful life prognostic method based on two-stage adaptive Wiener process[J]. Acta Automatica Sinica，2022，48(02)：539-553.
[11] ZHANG Z，SI X，HU C，et al. Degradation data analysis and remaining useful life estimation：a review on Wiener-process-based methods[J]. European Journal of Operational Research，2018，271(3)：775-796.
[12] OKUMURA S，JARDINE A K S，YAMASHINA H. An inspection policy for a deteriorating single-unit system characterized by a delay-time model[J]. International Journal of Production Research，1996，34(9)：2441-2460.
[13] VAN ZYL J P，ENGELBRECHT A P. Set-based particle swarm optimisation：a review[J]. Mathematics，2023，11(13)：2980.
[14] SAYYAD S，KUMAR S，BONGALE A，et al. Data-driven remaining useful life estimation for milling process：sensors，algorithms，datasets，and future directions[J]. IEEE Access，2021，9：110255-86.
[15] WANG Z，WANG W，HU C，et al. A prognostic-information-based order-replacement policy for a non-repairable critical system in service[J]. IEEE Transactions on Reliability，2015，64(2)：721-728.
[16] YAN T，LEI Y，WANG B，et al. Joint maintenance and spare parts inventory optimization for multi-unit systems considering imperfect maintenance actions[J]. Reliability Engineering & System Safety，2020，202：106994.
[17] HAJIPOUR Y，TAGHIPOUR S. Non-periodic inspection optimization of multi-component and k-out-of-m systems[J]. Reliability Engineering & System Safety，2016，156：228-243.

考虑寿命预测误差的机械装备维护决策方法研究

Research on Maintenance Decision-making Method of Machinery Considering Life Prediction Error

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 13

编辑推荐

Metrics

本文评价

[1]	綦法群, 胡琴, 张桉铭, 王茜莎, 黄洪钟. 考虑客户满意度的带缓冲串行系统维护决策[J]. 机械工程学报, 2024, 60(13): 173-181.
[2]	李响, 付春霖, 雷亚国, 李乃鹏, 杨彬. 保证数据隐私的装备协同智能故障诊断联邦迁移学习方法[J]. 机械工程学报, 2023, 59(6): 1-9.
[3]	林京, 焦金阳. 可解释机械智能诊断技术的研究进展与挑战[J]. 机械工程学报, 2023, 59(20): 215-224.
[4]	郭永兴, 熊丽, 周兴林, 朱建阳, 侯宇, 朱攀. 用于机械装备的高性能光纤光栅倾斜传感器[J]. 机械工程学报, 2022, 58(8): 71-78.
[5]	王金龙, 高斯博, 杨宇星, 鲍永杰, 张元良. 机械装备可再制造性评价研究综述[J]. 机械工程学报, 2022, 58(3): 221-234.
[6]	雷亚国, 许学方, 蔡潇, 李乃鹏, 孔德同, 张勇铭. 面向机械装备健康监测的数据质量保障方法研究[J]. 机械工程学报, 2021, 57(4): 1-9.
[7]	李乃鹏, 蔡潇, 雷亚国, 徐鹏程, 王文廷, 王彪. 一种融合多传感器数据的数模联动机械剩余寿命预测方法[J]. 机械工程学报, 2021, 57(20): 29-37,46.
[8]	沈长青, 汤盛浩, 江星星, 石娟娟, 王俊, 朱忠奎. 独立自适应学习率优化深度信念网络在轴承故障诊断中的应用研究[J]. 机械工程学报, 2019, 55(7): 81-88.
[9]	雷亚国, 杨彬, 杜兆钧, 吕娜. 大数据下机械装备故障的深度迁移诊断方法[J]. 机械工程学报, 2019, 55(7): 1-8.
[10]	雷亚国, 贾峰, 孔德同, 林京, 邢赛博. 大数据下机械智能故障诊断的机遇与挑战[J]. 机械工程学报, 2018, 54(5): 94-104.
[11]	张元良;张洪潮;赵嘉旭;周志民;王金龙. 高端机械装备再制造无损检测综述[J]. , 2013, 49(7): 80-90.
[12]	杨兆建;王学文. 矿山机械装备云仿真CAE服务系统研究与应用[J]. , 2013, 49(19): 111-118.
[13]	钟群鹏;有移亮;张峥;吴素君;骆红云. 机械装备构件轻量化主要技术途径的探讨[J]. , 2012, 48(18): 2-6.