Electrical Load Profile Analysis Using Clustering Techniques. (March 2017)

• Record Type: Journal Article
• Authors: Damayanti, R; Abdullah, A G; Purnama, W; Nandiyanto, A B D
• Is Part Of: IOP conference series. Volume 180(2017)
• Publication Date: 2017-03
• Subjects: electric load profile -- k-means -- fuzzy c-means -- k-harmonic means -- davies-bouldin index -- loss factor -- load factor; Materials science -- Periodicals; 620.1105
• DOI: 10.1088/1757-899X/180/1/012081
• Languages: English
• ISSNs: 1757-8981

Abstract

Abstract: Data mining is one of the data processing techniques to collect information from a set of stored data. Every day the consumption of electricity load is recorded by Electrical Company, usually at intervals of 15 or 30 minutes. This paper uses a clustering technique, which is one of data mining techniques to analyse the electrical load profiles during 2014. The three methods of clustering techniques were compared, namely K-Means (KM), Fuzzy C-Means (FCM), and K-Means Harmonics (KHM). The result shows that KHM is the most appropriate method to classify the electrical load profile. The optimum number of clusters is determined using the Davies-Bouldin Index. By grouping the load profile, the demand of variation analysis and estimation of energy loss from the group of load profile with similar pattern can be done. From the group of electric load profile, it can be known cluster load factor and a range of cluster loss factor that can help to find the range of values of coefficients for the estimated loss of energy without performing load flow studies.