“KLASTERISASI DATA HISTORICAL PENGGUNAAN DAYA PELANGGAN AMR (AUTOMATIC METER READING) MENGGUNAKAN METODE K-MEANS PADA PT. PLN(PERSERO) DISTRIBUSI JAKARTA RAYAâ€
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Perusahaan Listrik Negara
Author
TUPAMAHU, MELDRIN
Fitriani, Yessy
Prathama, Muhammad Fadli
Fitriani, Yessy
Prathama, Muhammad Fadli
Subject
Teknik Informatika
Datestamp
2022-09-22 03:37:25
Abstract :
AMR System (Automatic Meter Reading) PT. PLN (Persero) Distribusi Jakarta Raya applied to be able to detect losses (loss of electric power).Non-technical losses is one type of losses that gives a big effect on power loss. Currently to detect the losses itself is still checked directly by reading and checking historical power usage data of every incoming customer. For that we need a system that can be used to make data analysis and evaluation easier. This research uses Clustering K-means method to classify data based on electric power usage and Davies-Bouldin Index method to determine which cluster set is most optimal for use in grouping. This application is designed using a flowchart to describe each process performed within the application. For functional testing in applications using black-box method. Test results on the grouping of AMR's historical power usage data show that K-Means method is able to classify the customer's electricity usage with the optimal number of cluster sets, 3 based on the optimal cluster set using Davies-Bouldin Index. With this application is expected result of power usage pattern obtained, can be used as a reference in analyzing every usage graph from AMR Customer historical power usage data and can detect non-technical losses customer error.
AMR System (Automatic Meter Reading) PT. PLN (Persero) Distribusi Jakarta Raya applied to be able to detect losses (loss of electric power).Non-technical losses is one type of losses that gives a big effect on power loss. Currently to detect the losses itself is still checked directly by reading and checking historical power usage data of every incoming customer. For that we need a system that can be used to make data analysis and evaluation easier. This research uses Clustering K-means method to classify data based on electric power usage and Davies-Bouldin Index method to determine which cluster set is most optimal for use in grouping. This application is designed using a flowchart to describe each process performed within the application. For functional testing in applications using black-box method. Test results on the grouping of AMR's historical power usage data show that K-Means method is able to classify the customer's electricity usage with the optimal number of cluster sets, 3 based on the optimal cluster set using Davies-Bouldin Index. With this application is expected result of power usage pattern obtained, can be used as a reference in analyzing every usage graph from AMR Customer historical power usage data and can detect non-technical losses customer error.
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
