基于粒子群算法的异常用电检测方法

窃电行为严重危害了电网的正常运行,并且损害了国家和电力企业的利益。针对现有异常用电检测方法的不足,提出一种基于粒子群算法的异常用电检测新方法。以用户历史负荷数据为基础,采用粒子群算法提取用户相同行业的负荷模式曲线和用户历史数据的负荷模式曲线,并根据用户考察日负荷曲线与上述两种负荷模式匹配的不同特点,使用不同的负荷数据预处理方式以及模式匹配评价方法。实例分析表明,新方法能有效检测到异常用电的情况,验证了模型的有效性。

Study on Abnormal Electricity Utilization Detection Method Based on P SO Algorithm

Stealing power acts seriously endanger the regular operation of power grid and damage the interests of national and electrical enterprise. According to the shortages of the existing methods to abnormal electricity utilization detection, this paper proposes a new method based on PSO algorithm to solve this problem. On the basis of customers'historic load data, PSO algorithm is chosen to extract the load pattern curve of customers'which are in the same business and the load pattern curve of customers'historic load data. Accord?ing to different characteristics of the matching between customers'detection?day load curve and the above mentioned load pattern, dif?ferent pre?process methods of load data mode and pattern matching evaluation methods are used. this studies show that the new method can successfully detect the abnormal electricity utilization situation and verify the effectiveness of the mode.