Analysis Clustering of Electricity Usage Profile Using K-Means Algorithm

International Conference on Engineering and Technology for Sustainable Development 2015

工业技术;经济学

Amri, Yasirli^1 ; Fadhilah, Amanda Lailatul^1 ; Fatmawati^1 ; Setiani, Novi^1 ; Rani, Septia^1

Department of Informatics Engineering, Universitas Islam Indonesia, Yogyakarta, Indonesia^1

关键词: clustering;    Clustering techniques;    Economic growths;    Electricity production;    Electricity usage;    Electricity-consumption;    k-Means algorithm;    Production Planning;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/105/1/012020/pdf
DOI  :  10.1088/1757-899X/105/1/012020

学科分类：工业工程学

来源: IOP

【 摘 要 】

Electricity is one of the most important needs for human life in many sectors. Demand for electricity will increase in line with population and economic growth. Adjustment of the amount of electricity production in specified time is important because the cost of storing electricity is expensive. For handling this problem, we need knowledge about the electricity usage pattern of clients. This pattern can be obtained by using clustering techniques. In this paper, clustering is used to obtain the similarity of electricity usage patterns in a specified time. We use K-Means algorithm to employ clustering on the dataset of electricity consumption from 370 clients that collected in a year. Result of this study, we obtained an interesting pattern that there is a big group of clients consume the lowest electric load in spring season, but in another group, the lowest electricity consumption occurred in winter season. From this result, electricity provider can make production planning in specified season based on pattern of electricity usage profile.