摘要
分闸附件作为动作触发机构,其机械性能对于万能式断路器可靠运行至关重要。针对其寿命预测,提出了一种基于二元维纳性能退化分析的寿命预测方法。该方法结合分闸附件的工作机制,采取振电联合分析模式。首先,利用CEEMDAN-WD与短时能熵比双门限法,实现分闸过程触发阶段特定动作振动事件的精准标定;然后,研究振动事件所对应的机构动作时间以及驱动能量参数作为性能退化指标;在此基础之上构建二元维纳模型,以振电联合分析指标作为预测模型的输入,最终实现剩余寿命的定量预测。通过对3台断路器的测试表明,基于振电协同分析的寿命预测方法具有更高的预测精度,3台试品平均相对误差分别为4.3%、9.77%、8.46%,兼具有工程实用性。
As an action trigger mechanism,the mechanical performance of the opening accessory is very important for the reliable operation of conventional circuit breaker.Aiming at its life prediction,a life prediction method based on binary Wiener performance degradation analysis is proposed.Combining the working mechanism of the opening accessory,the method adopts the vibro-electrical joint analysis mode.First,CEEMDAN-WD and double-threshold method based on short-term energy entropy ratio are used to realize the accurate calibration of the vibration events for the specific actions in the triggering stage of the opening process.Then,the time and driving energy parameters of the mechanism action cor-responding to the vibration event are used as the performance degradation index.On this basis,a binary Wiener model is constructed with the vibro-electrical joint analysis index as the input of the prediction model,and the quantitative predic-tion of the remaining life is realized finally.The tests on three circuit breakers show that the life prediction method based on vibro-electrical synergistic analysis has higher prediction accuracy.The average relative errors of the three test prod-ucts are 4.3%,9.77%,and 8.46%,respectively,which has engineering practicality.
作者
孙曙光
邵旭
王景芹
魏硕
SUN Shuguang;SHAO Xu;WANG Jingqin;WEI Shuo(College of Artificial Intelligence and Date Science,Hebei University of Technology,Tianjin 300130,China;State Key Labora-tory of Reliability and Intelligence of Electrical Equipment,Hebei University of Technology,Tianjin 300130,China)
出处
《高电压技术》
EI
CAS
CSCD
北大核心
2024年第2期503-514,共12页
High Voltage Engineering
基金
河北省自然科学基金(E2021202136)。
关键词
万能式断路器
振电信号
性能退化
二元维纳过程
剩余寿命预测
conventional circuit breaker
vibro-electrical signals
performance degradation
binary Wiener process
re-maining useful life prediction