Research and Application of Power Consumers Behavior Analysis Based on K-means Algorithm

Abstract

K-means algorithm is used to analyze pretreated monthly load data of power consumers from a certain domestic area in China. It can get different clusters in three months, and those can be roughly divided power consumers into different types, such as residents, businesses and factories based on power customers' behavior analysis. A measure is introduced to evaluate the reliability of the algorithm. According to the above clustering results, it can be found out that the power consumption characteristics of all kinds of consumers in the corresponding months. In view of these characteristics, this paper can play a guiding role in formulating the corresponding power service strategy for different types of power consumers in power supply companies.

Klasterisasi Bibit Terbaik Menggunakan Algoritma K-Means dalam Meningkatkan Penjualan

“…Pengelompokan data tersebut digunakan untuk meningkatan manajemen stok dan strategi penjualan [10]. Penerapan algoritma K-Means juga digunakan untuk mengusulkan strategi Customer Relationship Management (CRM) dengan mengidentifikasi pola pembelian Produk [11]. Algoritma K-Means juga pernah digunakan untuk menganalisa data beban bulanan pra-konsumsi dari konsumen listrik dari daerah domestik tertentu di Cina [12].…”