基于距离的离群点算法和三相不平衡电流分析的用户窃电研究

Study on Anti-Electricity Stealing Based on Outlier Algorithm and Analysis of Three Phase Unbalanced Current

传统依托用电量和台区线损关联性来定位、现场排查窃电的模式存在效率低、准确性不高等问题,同时具有一定的滞后性。通过全面分析目前的窃电手段,绘制出实施窃电方式的帕累托图;针对窃电方式占比较高的失流窃电方式,结合96个点的日冻结电流数据,开展基于距离的离群点算法来辅助定位疑似窃电用户;为了剔除三相不平衡电流对失流窃电算法的影响,开展三相不平衡台区的理论线损值评估计算;最终根据这两个条件来筛选重点疑似窃电对象,为基层供电公司开展精准有效反窃电提供了一种新思路。

There are many ways to steal electricity.The traditional mode of locating and investigating on-site based on the correlation between power consumption and line loss is often inefficient,inaccurate and lagging.This paper comprehensively analyzes the current means of stealing electricity,draws the Pareto Diagram of the implementation of power stealing;for the high proportion of power stealing methods,combined with the daily freezing current data of 96 points,the outlier algorithm based on distance is developed to assist locating the suspected power stealing users;in order to eliminate the influence of three-phase unbalanced current on the loss of current stealing algorithm,the theory of three-phase unbalanced current is carried out.Finally,according to the above two conditions to screen the key suspected electricity stealing objects,which provides a new idea for the power supply companies to carry out accurate and effective anti-stealing power.