考虑动态环境效应的退化型产品剩余寿命预测方法
结题报告
批准号:
72001210
项目类别:
青年科学基金项目
资助金额:
24.0 万元
负责人:
刘天宇
学科分类:
工业工程与质量管理
结题年份:
2023
批准年份:
2020
项目状态:
已结题
项目参与者:
刘天宇
国基评审专家1V1指导 中标率高出同行96.8%
结合最新热点,提供专业选题建议
深度指导申报书撰写,确保创新可行
指导项目中标800+,快速提高中标率
客服二维码
微信扫码咨询
中文摘要
剩余寿命预测是预测与健康管理技术(PHM)的核心内容。不同于实验室场景,大多数产品实际工作时所处自然环境、工况是不断变化的。当前剩余寿命预测研究多基于理想试验数据建模,未全面考虑环境变动对预测结果的影响。因此,本项目以应力敏感型退化失效产品(如锂离子电池)为对象,针对现场应力剖面随时间不断变化的特点,研究动态环境下产品退化建模和剩余寿命预测方法。首先,针对环境变动对退化过程可能造成的加速、补偿、恢复三种影响,研究基于随机过程的退化建模方法,突破目前加速退化建模理论框架的局限性。然后,在考虑在线算法时效性的基础上,综合利用贝叶斯解析推导、滤波、数值仿真等方法给出动态环境下复杂退化模型更新和剩余寿命分布估计方法。最后,结合实际案例对本理论成果的可行性和有效性进行验证。本研究旨在打破当前剩余寿命预测理论方法研究与工程实践应用之间的壁垒,为产品在线健康管理与视情维修提供更为科学的决策依据。
英文摘要
Remaining/residual useful life (RUL) prediction is the core of prognostic and health management (PHM) technique. Unlike the laboratory environment, most products operate under field conditions, where some environmental/operating profiles vary with time. There is a great gap between practical application and the existing ideal laboratory data based RUL prediction methods, which do not consider the influences of condition changes comprehensively. This project aims at the features of degraded-failed and stress-sensitive products under field conditions (such as lithium ion battery), including the time-varying stress profile and its continuous influences on the degradation process, and proposes the degradation modeling and RUL prediction methods under dynamic conditions. Firstly, the stochastic process is used to model a product’s degradation process, with consideration of three potential complex environment effects, namely acceleration, compensation and recovery. The novel degradation model can break through the restriction of traditional degradation modeling framework only considering stress acceleration. And then, considering the computation complexity of online algorithms, the complex degradation model updating and RUL distribution evaluation methods under dynamic conditions are studied by synthetically using Bayesian analytical derivation, filtering and numerical simulation methods. Finally, some real-world case studies are used to demonstrate and valid our methods. This study can help break down the barriers between RUL theoretical study and engineering application, and provide scientific bases for products health management and condition-based maintenance.
期刊论文列表
专著列表
科研奖励列表
会议论文列表
专利列表
DOI:https://doi-org-s.libyc.nudt.edu.cn:443/10.1177/1748006X231183196
发表时间:2023
期刊:Proceedings of the Institution of Mechanical Engineers, Part O: Journal of Risk and Reliability
影响因子:--
作者:Liu Tianyu;Pan Zhengqiang;Song Guopeng
通讯作者:Song Guopeng
DOI:10.1016/j.apm.2021.12.041
发表时间:2022-01
期刊:Applied Mathematical Modelling
影响因子:5
作者:Lulu Zhang;Guanghao Jin;Tian-yu Liu;Ran Zhang
通讯作者:Lulu Zhang;Guanghao Jin;Tian-yu Liu;Ran Zhang
DOI:--
发表时间:2023
期刊:系统工程与电子技术
影响因子:--
作者:张然;刘天宇;金光
通讯作者:金光
DOI:10.12305/j.issn.1001-506x.2023.07.18
发表时间:2023
期刊:系统工程与电子技术
影响因子:--
作者:张路路;潘正强;刘天宇;金光
通讯作者:金光
DOI:10.3390/math10224216
发表时间:2022-11
期刊:Mathematics
影响因子:2.4
作者:Tianyu Liu;Lulu Zhang;G. Jin;Zhengqiang Pan
通讯作者:Tianyu Liu;Lulu Zhang;G. Jin;Zhengqiang Pan
国内基金
海外基金