Coal Characterization of South Sumatera Basin using the Unsupervised Machine Learning Method

• Published in: IOP conference series. Earth and environmental science Vol. 830; no. 1; pp. 12043 - 12050
• Main Authors: Ayustyana, E, Wibisono, S A, Sihombing, F M H
• Format: Journal Article
• Language: English
• Published: Bristol IOP Publishing 01.09.2021
• Subjects: Airbenakat Formation; Carbon; Clusters; Coal; Learning algorithms; Machine Learning; Moisture; Muaraenim Formation; Nitrogen; Sedimentary basins; South Sumatera Basin; Sulfur; Sulfur content; Unsupervised learning; Sumatra; Indonesia
• ISSN: 1755-1307; 1755-1315
• DOI: 10.1088/1755-1315/830/1/012043

Abstract South Sumatra Basin is one of the sedimentary basins in Indonesia which has coal bearing formation, coal-bearing formations are rock formations that have coal layers in its area. Numerous coal proximate data (free moisture, total moisture, moisture, inherent moisture, ash content, fixed carbon, sulfur content) and ultimate data (carbon (C), hydrogen (H), oxygen (O), nitrogen (N) have been collected in this basin. This study aims to determine the characterization of coal in the South Sumatra Basin based on the proximate data and ultimate data using the unsupervised machine learning methods. The machine learning method has several basic concepts, namely being able to predict data by studying several patterns and factors that have been trained in a short amount of time. The study able to cluster coal in the basin into two cluster of coals with striking difference. The distribution of the two coal clusters in the South Sumatra Basin possibly influenced by the age of the formation in the South Sumatra Basin. In the first cluster, it is distributed in the older Airbenakat Formation and Muaraenim Formation, while in the second cluster it is scattered in the younger Muaraenim Formation and the Kasai Formation. The formation ages of the youngest are the Kasai Formation, the Muaraenim Formation, and the Airbenakat Formation.