Kusuma, Dine Tiara and Arvio, Yozika and Sangadji, Iriansyah B.M and Ahmad, Norashikin and Syed Ahmad, Sharifah Sakinah and Puspitasari, Fadhilah (2024) An analysis of electrical energy usage in social consumers using Fuzzy C Means. In: International Conference on Electrical Engineering, Computer Science and Informatics (EECSI).
![[img]](http://eprints.utem.edu.my/style/images/fileicons/text.png) |
Text
An Analysis of Electrical Energy Usage in Social Consumers Using Fuzzy C Means.pdf Download (485kB) |
Abstract
The objective of this study is to examine the patterns of electrical energy usage in social consumers by employing the Fuzzy C Means clustering method to assist utility companies in the development of business strategies, operations, planning, and energy policies that are customized to the specific needs of each group. The problem of unpredictable social activities, changes in work patterns, and community activities, particularly among social consumers, result in significant declines and irregular electricity consumption patterns. It is crucial for energy providers to comprehend these evolving patterns to effectively manage electricity supplies. This research was carried out in a 5-step process, namely Data Collection, Data Preprocessing, Clustering Process with Fuzzy C Means (FCM), find pattern and clustering validity. This study sources its data from the Automatic Meter Reading (AMR) of social consumers, specifically places of worship, educational institutions, and health services. We employ the Fuzzy C Means (FCM) Clustering method to categorize electrical energy consumption patterns into multiple groups, considering similarities in the data. The variables utilized in this analysis are kilowatt (kW) and kilovolt-ampere (kVA). The findings indicate that places of worship, educational institutions, and health services exhibit distinct patterns of electrical energy usage, although belonging to the same electricity tariff category. The quality of cluster results in this research using the Davies- Bouldin index (DBI) method is 0.51, which shows that the resulting clustering has a fairly good level of separation and compactness.
| Item Type: | Conference or Workshop Item (Paper) |
|---|---|
| Divisions: | Faculty of Information and Communication Technology |
| Depositing User: | NURHASHIRAH BORHAN |
| Date Deposited: | 31 Jul 2026 07:28 |
| Last Modified: | 31 Jul 2026 07:28 |
| URI: | http://eprints.utem.edu.my/id/eprint/30119 |
| Statistic Details: | View Download Statistic |
Actions (login required)
 |
View Item |