状态检修决策中的电气设备故障率推算  

Prediction of electrical equipment failure rate for condition-based maintenance decision-making

作　　者：潘乐真[1] 张焰[1] 俞国勤[2] 杜成刚[2] 

机构地区：[1]上海交通大学电气工程系,上海200240 [2]上海市电力公司,上海200122

出　　处：《电力自动化设备》2010年第2期91-94,共4页Electric Power Automation Equipment

摘　　要：针对统计分析所得的电气设备故障率存在可信度低、应用滞后性的问题,提出了电气设备故障率推算的新方法。以电气设备实际的运行状态为依据,以相关的评价导则为标准,采用健康指数量化设备状态的劣化程度,推算当前状态下的设备故障率,进而确定设备的实际役龄;采用Marquardt法推求设备的故障率函数,并结合威布尔分布对设备故障率曲线进行分段拟合;在上述基础上,通过引入役龄回退因子描述检修工作对设备的修复程度,进而确定检修后设备的等效役龄,最终实现对检修后的电气设备故障率的预测,为后续实施多台设备状态检修决策优化提供支持。通过实际算例分析,说明了该项研究的有效性。As the electrical equipment failure rate calculated by post-failure statistic analysis has low reliability,a way to predict electrical equipment failure rate is introduced,which is based on the actual operating state of equipment and the relevant evaluation rules. The health index is adopted to quantify the equipment deterioration ,which is used to extrapolate the current equipment failure rate and determine its actual service age. The Marquardt method is used to deduce the failure rate function and the Weibull distribution is combined to piecewise fit the curve. The age reduction factor is introduced to describe the repair extent,the equivalent service age of the equipment after maintenance is then determined and its failure rate is predicted,which supports the follow-up implementation of optimized condition-based maintenance for multiple devices. Case study shows the effectiveness of presented method.

关 键 词：健康指数 威布尔分布 役龄回退因子 等效役龄 故障率预测 状态检修 

分 类 号：TM732[电气工程—电力系统及自动化]