Estimasi Sisa Masa Hidup Komponen Berputar (Ball Bearings) Pada Motor Coal Grinding Dengan Pendekatan Model Degradasi Berbasis Data Akselerasi

Zaqi, Muhammad Fairuz (2026) Estimasi Sisa Masa Hidup Komponen Berputar (Ball Bearings) Pada Motor Coal Grinding Dengan Pendekatan Model Degradasi Berbasis Data Akselerasi. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5003221068-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Rolling Element Bearing (REB) merupakan salah satu komponen kritis pada mesin industri yang rentan mengalami degradasi akibat operasi secara terus-menerus. Kegagalan komponen yang tidak terdeteksi dapat menyebabkan penghentian proses produksi secara tiba-tiba serta meningkatkan biaya pemeliharaan. Oleh karena itu, diperlukan metode prognosis untuk mengestimasi Remaining Useful Life (RUL) sehingga aktivitas pemeliharaan dapat direncanakan secara optimal. Penelitian ini bertujuan menganalisis karakteristik degradasi REB berdasarkan Health Indicator (HI), membandingkan performa model degradasi pada setiap siklus operasi, serta mengestimasi RUL menggunakan model degradasi yang paling representatif. Data yang digunakan merupakan data sekunder sensor getaran REB pada motor coal grinding PT. XYZ selama periode 1 Januari 2025 hingga 26 April 2026. HI dibentuk menggunakan metode Principal Component Analysis (PCA). Selanjutnya, data dibagi menjadi tujuh siklus operasi berdasarkan riwayat aktivitas pemeliharaan, kemudian setiap siklus dimodelkan menggunakan model degradasi linear dan eksponensial. Model terbaik pada masing-masing siklus dipilih berdasarkan nilai Root Mean Square Error (RMSE), kemudian model yang paling dominan digunakan untuk mengestimasi RUL pada siklus operasi berikutnya dengan memanfaatkan rata-rata degradation rate dari model dominan tersebut. Hasil penelitian menunjukkan bahwa PC1 mampu menjelaskan 96,93% keragaman data sehingga dipilih sebagai HI.. Berdasarkan hasil evaluasi, model degradasi eksponensial merupakan model yang paling representatif dengan menjadi model terbaik pada lima dari tujuh siklus operasi. Menggunakan model tersebut diperoleh estimasi Remaining Useful Life (RUL) sebesar 1.490,38 jam atau sekitar 62 hari operasi. Hasil penelitian menunjukkan bahwa pendekatan pemodelan degradasi berbasis siklus operasi mampu merepresentasikan karakteristik degradasi REB dan memberikan dasar kuantitatif dalam mendukung perencanaan predictive maintenance.
==========================================================================================================================================
Rolling Element Bearings (REBs) are among the most critical components in industrial machinery and are susceptible to gradual degradation due to continuous operation. Undetected bearing failures can lead to unexpected production downtime and increased maintenance costs. Therefore, prognostic approaches are required to estimate the Remaining Useful Life (RUL) so that maintenance activities can be planned more effectively. This study aims to analyze the degradation characteristics of REBs based on a Health Indicator (HI), compare the performance of degradation models for each operating cycle, and estimate the RUL using the most representative degradation model. The study utilized secondary vibration sensor data collected from the REB of a coal grinding motor at PT. XYZ during the period from January 1, 2025, to April 26, 2026. The HI was constructed using Principal Component Analysis (PCA). Subsequently, the data were divided into seven operating cycles based on maintenance history, and each cycle was modeled using both linear and exponential degradation models. The best model for each cycle was selected based on the Root Mean Square Error (RMSE), after which the most dominant degradation model was employed to estimate the RUL for the subsequent operating cycle using the average degradation rate of the dominant model. The results indicate that the first principal component (PC1) explains 96.93% of the total data variance and was therefore selected as the HI. Based on the evaluation results, the exponential degradation model was identified as the most representative model, providing the best performance in five out of seven operating cycles. Using this model, the estimated Remaining Useful Life (RUL) was 1,490.38 operating hours, equivalent to approximately 62 days of operation. These findings demonstrate that the cycle-based degradation modeling approach is capable of representing the degradation behavior of REBs and provides a quantitative basis for supporting predictive maintenance planning.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Data Akselerasi, Model Degradasi, Predictive Maintenance, Rolling Element Bearing,Acceleration Data, Degradation Model, Predictive Maintenance, Rolling Element Bearing
Subjects:	H Social Sciences > HA Statistics > HA31.3 Regression. Correlation. Logistic regression analysis. H Social Sciences > HA Statistics > HA31.7 Estimation Q Science > QA Mathematics > QA278.2 Regression Analysis. Logistic regression Q Science > QA Mathematics > QA278.5 Principal components analysis. Factor analysis. Correspondence analysis (Statistics) Q Science > QA Mathematics > QA278 Cluster Analysis. Multivariate analysis. Correspondence analysis (Statistics)
Divisions:	Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User:	Muhammad Fairuz Zaqi
Date Deposited:	04 Aug 2026 08:13
Last Modified:	04 Aug 2026 08:13
URI:	http://repository.its.ac.id/id/eprint/143694

Actions (login required)

View Item