在运智能电能表批量剩余寿命预判方法

Prediction method for batch residual life of smart electricity meter in operation

该方法主要目的在于通过对批次性电能表历史故障数据的分析,判断故障分布浴盆曲线拐点和预测下一个时间段的失效率。该方法利用对电力公司用采系统和营销系统故障数据的二次挖掘,提取有技术价值的故障数据,通过分析研究电能表产品整体批次失效率随时间变化的特征,利用威布尔分布拟合的方法从电能表无差别数据和考虑外部应力条件后的故障数据两个层级对产品的批次可靠性状态进行判断和下阶段的寿命指标进行预测,并输出相应的批次电能表当前和下一个阶段失效率、累计失效率和分布类型的形状参数,进而为批次性电能表轮换周期节点的判断做出技术数据支撑,输出高故障率风险提示、警告。

The main purpose of this method is to predict the inflection point of fault distribution bathtub curve and the change of failure rate in the next period of time for batch of electricity meters,and to be compatible with the quality evaluation of the full-scale inspection data and the statistical analysis of failure.This method utilizes the secondary mining of fault data of mining system and marketing system used by power companies to continuously collect fault data.By analyzing and studying the characteristics of overall batch failure rate of electricity meter products,the residual life of products is predicted and judged from two levels of electricity meters,fault module and stress condition by using Weibull distribution fitting method,and the corresponding batch electricity meter is output.The shape parameters of current and the next stage failure rate and cumulative failure rate and distribution type of the electricity meter can be used to predict the rotation cycle nodes and output high failure rate risk warnings.