基于CUSUM滑动窗的二维离散模糊数电力负荷检测方法

A two-dimensional discrete fuzzy number power load detection method based on CUSUM moving window

针对负荷事件检测和负荷辨识在非侵入式负荷监测上的应用,提出了一种基于CUSUM(cumulative sum)滑动窗的二维离散模糊数电力负荷检测方法。该方法首先根据基于改进的滑动窗的CUSUM负荷事件检测算法,摒弃了累加值阈值判断机制,采用累积所占比判断设备的负荷投切情况;然后从有功P和无功Q两种典型的负荷标签特征出发,采用二维离散模糊数算法,将用电设备的特征转换为离散模糊数矩阵,并结合矩阵质心和评判比例得出最终评价值,进而实现负荷的辨识。最后,通过多个用电负荷投切实验,发现通过文中方法得到的结果与实际负荷投切的结果一致,并与其他现有的方法进行了比较,验证了该方法的有效性。

Aiming at the application of load event detection and load identification in non-intrusive load monitoring, a two-dimensional discrete fuzzy number method for load detection based on the cumulative sum(CUSUM) moving window is proposed. At first, the load event detection method is improved by using the moving windows of CUSUM. Instead of the cumulative value threshold judgment mechanism, the cumulative proportion is adopted to judge the load switching of the equipment. Then, based on the two typical load label features of active power P and reactive power Q, a two-dimensional discrete fuzzy number algorithm is adopted to transform the characteristics of the electrical equipment into a discrete fuzzy number matrix, and the final evaluation value is obtained by combining the center of mass of the matrix and the evaluation ratio,so as to realize load identification. Finally, the experiments on the real environment demonstrate the performance of load event detection and load identification, and the results show that the proposed method can coincide with those obtained by the actual load switching experiments, and the recognition effectiveness of this method is verified by comparing with other methods.