Abnormal Electricity Customer Clustering Method Based on Electricity Big Data
Qingshen Gong,Yukun Jin,Yipin Han,Chunxu Sang,Rui Wang,Bowen Zhou
DOI: https://doi.org/10.1109/ispec48194.2019.8974928
2019-01-01
Abstract:With the continuous improvement of the information system of power systems and the rapid growth of the amount of data used for distribution, various devices and systems have generated a large amount of data, and the data is large in scale and contains a wide variety of event information. There are various risks in the production and operation of power companies, and the risk management has always been a key issue of concern. The risk management directly affects the cost and profits of the enterprises. This paper proposes a method of clustering the electricity customers according to the electricity consumption information. Firstly, the principal component analysis method is used to reduce the dimensionality of the electricity consumption information, and to extract the useful information. The superimposed information is saved, and the important information contained in the original variables is retained, which facilitates efficient clustering. Then, the k-means clustering algorithm based on particle swarm optimization is used to cluster the results of dimensionality reduction. The cluster-centered perturbation operator is added to the k-means clustering algorithm based on particle swarm optimization. The simulation results show that compared with the traditional k-means clustering algorithm, the proposed algorithm does not only get rid of the dependence on the initial value, but also not trap in the local optimum. It can also result in faster global convergence, which improves the efficiency of the algorithm and makes the results more accurate.