基于Logistic算法的电能计量装置异常数据识别方法

A Method for Identifying Abnormal Data of Electric Energy Metering Devices Based on Logistic Algorithm

为解决当前电能计量装置异常数据识别范围较小、识别速率低的问题,提出一种基于Logistic算法的电能计量装置异常数据识别方法。通过提取异常数据识别特征,采用多目标的方式,扩大数据识别范围。设计奇异值分解(Singular Value Decomposition,SVD)+多目标快速辨识矩阵,构建Logistic测算电能计量装置异常数据识别模型,采用多区间边界修正实现数据识别处理。测试结果表明,此次所设计的基于Logistic算法的电能计量装置异常数据识别方法的识别速率均可以达到6 Mb/s以上,即所设计方法的异常数据识别效果更佳,误差可控,具有较好的实际应用价值。

In order to solve the problems of small identification range and low identification rate of abnormal data of electric energy metering devices,a design and analysis of abnormal data identification method of electric energy metering devices based on Logistic algorithm is proposed.By extracting the characteristics of abnormal data identification,the multi-objective method is adopted to expand the scope of data identification.Singular Value Decomposition(SVD)+multi-objective rapid identification matrix is designed,and the abnormal data identification model of Logistic electric energy metering device is constructed,and the data identification processing is realized by multi-interval boundary correction.The test results show that the recognition rate of abnormal data of electric energy metering device based on Logistic algorithm can reach above 6 Mb/s,that is,the abnormal data recognition effect of the designed method is better,the error is controllable,and it has good practical application value.