Pemodelan Degradasi Tappered Roller Bearing SKF 30304 dengan Data NDT Vibrasi Menggunakan Metode ARIMA

Irwanto, Rizqullah (2024) Pemodelan Degradasi Tappered Roller Bearing SKF 30304 dengan Data NDT Vibrasi Menggunakan Metode ARIMA. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5007201224-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (2MB) | Request a copy

Abstract

Bearing merupakan komponen utama yang mendukung poros agar tetap berputar saat dikenai beban aksial dan radial. Bearing sering mengalami kegagalan sebelum mencapai life rated time yang menyebabkan unpredicted downtime pada mesin. Kegagalan tersebut seringkali terjadi karena vibrasi yang diterima oleh bearing dalam jangka waktu yang sangat lama. Kondisi fisik bearing dapat terdegradasi karena adanya fenomena fatigue yang dapat diketahui tingkat degradasinya melalui condition monitoring seperti Non-Destructive Test (NDT) dan modelling data degradasi bearing menggunakan machine learning. Penelitian ini dilakukan untuk membuat model degradasi bearing menggunakan data waveform vibrasi dengan mengaplikasikan algoritma Autoregressive Integrated Moving Average (ARIMA). Algoritma tersebut dapat menginisiasi model berdasarkan variasi jumlah data training yang akan digunakan, sehingga diperoleh kondisi degradasi pada bearing dan mengetahui kapan bearing tersebut sudah tidak layak untuk digunakan. Pemodelan degradasi ini membutuhkan data fitur getaran yang dibagi menjadi set training dan testing. Training model degradasi dengan metode ARIMA dilakukan dengan jumlah persentase data training yang berbeda (50%, 60%, 70%, 80%). Hasil yang diperoleh dari penelitian adalah analisis error RMSE, MAE, dan MAPE. Model 70% data training menunjukkan nilai RMSE dan MAE yang paling rendah sebesar 0.09 dan 0.075 secara berurutan. Model 80% data training menunjukkan nilai MAPE yang paling rendah sebesar 0.21%. Hal ini menandakan bahwa metode ARIMA memiliki kemampuan prediksi degradasi pada bearing yang sangat baik, sehingga penelitian ini dapat memberikan kontribusi signifikan dalam pembuatan model degradasi bearing.
===================================================================================================================================
Bearings are key components that support shafts to keep them rotating under axial and radial loads. Bearings often fail before reaching their life rated time which causes unpredictable downtime on the machine. This failure often occurs due to vibrations received by the bearing over a very long period. The physical condition of bearings can be degraded due to fatigue phenomena, the level of degradation of which can be determined through condition monitoring such as Non-Destructive Tests (NDT) and modeling bearing degradation data using machine learning. This research was conducted to create a bearing degradation model using vibration waveform data by applying the Autoregressive Integrated Moving Average (ARIMA) algorithm. This algorithm can initiate a model based on variations in the amount of training data that will be used, to obtain degradation conditions in the bearing and know when the bearing is no longer suitable for use. Modeling this degradation requires vibration feature data which is divided into training and testing sets. Training of the degradation model using the ARIMA method was carried out with different percentages of training data (50%, 60%, 70%, 80%). The results obtained from the research are RMSE, MAE, and MAPE error analysis. The 70% training data model shows the lowest RMSE and MAE values of 0.09 and 0.075 respectively. The 80% training data model shows the lowest MAPE value of 0.21%. Values from the error indicates that the ARIMA method has very good bearing degradation prediction capabilities, so this research can make a significant contribution in creating bearing degradation models.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	ARIMA, Bearing, Downtime, Vibrasi,Vibration
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ174 Maintenance and repair of machinery
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Mechanical Engineering > 21201-(S1) Undergraduate Thesis
Depositing User:	Rizqullah Irwanto
Date Deposited:	12 Aug 2024 03:00
Last Modified:	28 Aug 2024 08:56
URI:	http://repository.its.ac.id/id/eprint/114244

Actions (login required)

View Item