Deteksi Kerusakan Bearing Spindle CNC Milling 3-Axis Menggunakan Neural Network (NN)

Ningrum, Vinanda Asti (2022) Deteksi Kerusakan Bearing Spindle CNC Milling 3-Axis Menggunakan Neural Network (NN). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 10311810000009-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (3MB) | Request a copy

Abstract

Mesin CNC merupakan mesin yang digunakan untuk kebutuhan memotong atau mengukir benda kerja. Kelebihan mesin ini adalah ketelitian potongan hingga 0,1 mm. Salah satu kegagalan pada mesin tersebut terjadi di motor spindle terutama pada bagian bearing. Terjadinya kerusakan pada bearing spindle dapat menyebabkan getaran yang berlebih yang dapat mempengaruhi ketelitian potongan atau ukiran produk, sehingga kualitas produk dapat berkurang. Dari dampak tersebut, dibutuhkan sistem untuk deteksi kerusakan pada bearing spindle Mesin CNC. Pada penelitian ini, sistem deteksi dilakukan berdasarkan kondisi getarannya. Menggunakan data getaran yang diperoleh dari sensor accelerometer, dilakukan ekstraksi fitur menggunakan metode statistik dengan menghitung nilai maksimal, minimal, rata-rata dan standar deviasi sampel. Kemudian, hasil ekstraksi fitur digunakan untuk training dan testing menggunakan Neural Network (NN) agar didapat hasil deteksi kondisi bearing spindle rusak atau normal. Berdasarkan hasil penelitian, diperoleh akurasi hingga 98% dan running time 5,7 detik. Serta, menggunakan Odoo sebagai user interface hasil deteksi, dilakukan pengujian UAT (User Acceptance Testing) dan responden sementara sebanyak 15 orang, dengan 6 pernyataan. Menunjukkan hasil tingkat penerimaan responden terhadap sistem deteksi kerusakan bearing spindle mengggunakan Odoo sebagai user interface sebesar 89%. Serta berdasarkan hasil validasi test sistem deteksi dapat menampilkan kondisi bearing spindle sesuai dengan kondisi sebenarnya.
====================================================================================================================================
A CNC machine is a machine used for cutting or engraving workpieces. The advantage of this machine is the accuracy of the cut up to 0.1 mm. One of the failures on the machine occurred in the spindle motor, especially in the bearings. The occurrence of damage to the spindle bearing can cause excessive vibration which can affect the accuracy of the cut or engraving of the product, so that the quality of the product can be reduced. From this impact, a system is needed to detect damage to the spindle bearing of CNC machines. In this study, the detection system was carried out based on the vibration conditions. Using vibration data obtained from the accelerometer sensor, feature extraction is carried out using statistical methods by calculating the maximum, minimum, average and standard deviation values of the sample. Then, the result of feature extraction is used for training and testing using Neural Network (NN) in order to get the results of detection of damaged or normal spindle bearing conditions. Based on the results of the study, the accuracy is up to 98% and the running time is 5.7 seconds. Also, using Odoo as a user interface for detection results, UAT (User Acceptance Testing) testing was carried out and 15 temporary respondents, with 6 statements. Shows the results of respondents' acceptance of the spindle bearing damage detection system using Odoo as a user interface of 89%. And based on the results of the validation test, the detection system can display the condition of the spindle bearing in accordance with the actual conditions.

Item Type:	Thesis (Other)
Additional Information:	RSEO 621.822 Nin d-1 2022
Uncontrolled Keywords:	CNC Milling 3-Axis, Getaran, Neural Network (NN), Odoo, Vibration, Neural Network
Subjects:	Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Mr. Marsudiyana -
Date Deposited:	23 Apr 2026 07:30
Last Modified:	23 Apr 2026 07:30
URI:	http://repository.its.ac.id/id/eprint/132894

Actions (login required)

View Item