Predictive maintenance magnetic sensor using random forest method. (April 2020)

• Record Type: Journal Article
• Authors: Aji, A S; Sashiomarda, J A; Handoko, D
• Is Part Of: Journal of physics. Volume 1528(2020)
• Publication Date: 2020-04
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/1528/1/012030
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: BMKG has observations of earth's magnets which are scattered in Indonesia. The BMKG earth magnetic sensor produces real-time data output. This research focuses on predictive maintenance models on earth magnetic sensors based on data output. The resulting data output is in the form of a delimited format in the form of spaces so that it is easy to process. The magnetic component used is the total component data (F) from the earth's magnetic sensor. Data processing using python scripts with the algorithm used is the random forest method by comparing the resulting value difference to find out whether the data generated is still in tolerance or not. The results of the prediction model with the number of estimators 10 produce RF Score = 0.98 and MAE = 0.83.