Estimasi Waktu Kegagalan Generator Diesel Menggunakan Model Degradasi Untuk Menunjang Strategi Predictive Maintenance pada Kereta Api

Permatasari, Virliana Septi (2025) Estimasi Waktu Kegagalan Generator Diesel Menggunakan Model Degradasi Untuk Menunjang Strategi Predictive Maintenance pada Kereta Api. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6009222002-Master_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (4MB) | Request a copy

Abstract

Penelitian ini bertujuan untuk mengembangkan metode estimasi waktu kegagalan generator diesel pada kereta api dengan menggunakan model degradasi eksponensial sebagai bagian dari strategi predictive maintenance. Generator diesel merupakan komponen penting dalam menunjang operasi kereta api, namun rentan terhadap gangguan seperti air gap eccentricity yang dapat menyebabkan kegagalan fungsi. Dalam mengestimasi waktu kegagalan generator diesel, dilakukan pengumpulan data dari eksitasi arus generator selama 95 hari, kemudian dilakukan ekstraksi fitur menggunakan time domain analysis dari data tersebut. Hasil dari ekstraksi fitur, selanjutnya dilakukan analisis fitur menggunakan metode Principal Component Analysis untuk membangun Health Indicator (HI). Berdasarkan nilai HI, didapatkan model degradasi yang digunakan untuk mengestimasi waktu kegagalan generator diesel dengan mencari nilai sisa umur pakai (RUL). Estimasi waktu kegagalan menunjukkan bahwa generator diesel mengalami down pada hari pengamatan ke-90. Hasil analisis performansi dari model degradasi yang dibangun dapat mengestimasi waktu kegagalan generator diesel dengan rata-rata error sebesar 1,5%.
=====================================================================================================================================
This research aims to develop a method of estimating the failure time of diesel generators on trains using an exponential degradation model as part of a predictive maintenance strategy. Diesel generators are important components in supporting train operations, but are potentially vulnerable to disturbances such as air gap eccentricity that can cause malfunctions. In estimating the failure time of diesel generators, data was collected from the generator excitation current for 95 days, then feature extraction using time domain analysis was carried out from the data. The results of feature extraction, then feature analysis using the Principal Component Analysis method to build a Health Indicator (HI). Based on the HI value, a degradation model is obtained which is used to estimate the failure time of the diesel generator by finding the remaining useful life (RUL) value. The estimated time of failure shows that the diesel generator goes down on the 90th observation day. The performance analysis results of the degradation model built can estimate the failure time of diesel generators with an average error of 1.5%.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	air gap eccentricity, diesel generator, degradation model, health indicator, predictive maintenance, air gap eccentricity, generator diesel, health indicator, model degradasi, predictive maintenance.
Subjects:	T Technology > TF Railroad engineering and operation > TF193 Estimates, costs, etc.
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30101-(S2) Master Thesis
Depositing User:	Virliana Septi Permatasari
Date Deposited:	06 Feb 2025 07:53
Last Modified:	06 Feb 2025 07:53
URI:	http://repository.its.ac.id/id/eprint/118462

Actions (login required)

View Item