基于数据驱动的小子样复杂系统剩余使用寿命预测方法综述及展望  

Review and Prospect of Remaining Useful Life Prediction Methods for Small Sample Complex Systems Based on Data-Driven

作　　者：齐金平[1,2,3] 王康 QI Jin-ping;WANG Kang(Mechatronics T&R Institute,Lanzhou Jiaotong University,Lanzhou 730070,China;Gansu Engineering Technology Center for Informatization of Logistics&Transport Equipment,Lanzhou 730070,China;Gansu Industry Technology Center of Logistics&Transport Equipment,Lanzhou 730070,China)

机构地区：[1]兰州交通大学机电技术研究所,甘肃兰州730070 [2]甘肃省物流及运输装备信息化工程技术研究中心,甘肃兰州730070 [3]甘肃省物流与运输装备行业技术中心,甘肃兰州730070

出　　处：《测控技术》2023年第3期1-10,31,共11页Measurement & Control Technology

基　　金：国家自然科学基金(71861021);甘肃省高等学校科研项目(2018A-026,2018C-10);甘肃省重点研发计划项目(17YF1FA122);甘肃省教育厅优秀研究生“创新之星”项目(2021CXZX-575)。

摘　　要：针对小子样背景下复杂系统剩余使用寿命(RUL)预测的工程需求,结合复杂系统失效的时间数据、监测数据特点和RUL预测的不确定性问题,综述了小子样数据驱动的复杂系统RUL的预测方法。在小子样数据驱动的寿命预测技术中,数据的真实性、连续性和完整性等问题成为制约RUL预测准确度的重要因素。深入分析了基于失效时间数据、性能退化数据和多源数据融合的RUL预测技术的基本研究方法和发展动态,最后探讨了RUL预测领域未来可能的研究方向。For the engineering requirements of remaining useful life(RUL) prediction of complex systems under the background of small samples,combined with the characteristics of failure time data and monitoring data of complex systems and the uncertainty of RUL prediction,RUL prediction methods for small sample driven complex systems are reviewed.The authenticity,continuity and integrity of small sample data have become important factors restricting the accuracy in RUL prediction.The basic research ideas and development trends of RUL prediction technology based on failure time data,performance degradation data and multi-source data fusion are deeply analyzed.Finally,the possible future research directions in the field of RUL prediction are discussed.

关 键 词：小子样 RUL预测 数据驱动 现状 展望 

分 类 号：TP202[自动化与计算机技术—检测技术与自动化装置]