Analisis Vibrasi Untuk Klasifikasi Kerusakan Mesin Conveyor Finish Good Menggunakan Neural Network

Prasetyo, Muhammad Dwi (2023) Analisis Vibrasi Untuk Klasifikasi Kerusakan Mesin Conveyor Finish Good Menggunakan Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 10311910000059-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (6MB) | Request a copy

Abstract

PT Suntory Garuda Beverage adalah perusahaan yang bergerak pada bidang manufaktur produk minuman, Dalam proses produksinya menggunakan mesin Conveyor Finsih Good (CFG) untuk memindahkan produk yang telah dikemas menuju ke gudang penyimpanan. Berdasarkan informasi dari tim teknis, saat melakukan perawatan belum bisa menetukan treatment yang akan dilakukan karena belum ada sistem monitoring. Mesin CFG adalah jenis rotating machinery, pada dasarnya semua rotating machinery menghasilkan getaran yang merupakan fungsi kelurusan (alignment) dan keseimbangan (balance) dari komponen yang berputar. Pengukuran intensitas getaran dapat memberikan informasi tentang ketepatan kelurusan poros dan keseimbangannya. Oleh karena itu, PT Suntory Garuda Beverage menerapkan sistem pemeliharaan prediktif pada mesin CFG dengan pemantauan vibrasi. Data pengukuran vibrasi dapat digunakan untuk mengenali kerusakan yang terjadi pada mesin CFG. Identifikasi kerusakan mesin CFG dilakukan dengan menganalisis spektrum vibrasi yang memiliki karakteristik berbeda pada tiap jenis kerusakan. Proses analisis spektrum vibrasi dapat dilakukan menggunakan Fast Fourier Transform (FFT). Analisis domain frekuensi memudahkan dalam mengidentifikasi amplitudo dan frekuensi spektrum vibrasi, sehingga diperoleh informasi yang lebih rinci untuk mengenali jenis kerusakan mesin CFG. Dalam penelitian ini juga menggunakan Neural Network (NN) untuk mengklasifikasikan jenis kerusakan mesin CFG. Pada penelitian ini dilakukan 18 kali percobaan untuk menentukan arsitektur NN yang terbaik. Berdasarkan pengujian, arsitektur terbaik menggunakan 10 neuron pada hidden layer, dan menggunakan iterasi sebanyak 50 kali, yang mendapatkan nilai MSE sebesar 0,045. Sistem yang dibuat dapat mengklasifikasikan jenis kerusakan mesin CFG dalam 4 kondisi, yaitu normal, unbalance, misalignment, dan looseness dengan nilai akurasi hingga 100%
==================================================================================================================================
PT Suntory Garuda Beverage is a company engaged in the manufacturing of beverage products. In the production process, it uses a Conveyor Finsih Good (CFG) machine to move packaged products to the storage warehouse. Based on information from the technical team, when carrying out the treatment, it was not possible to determine the treatment to be carried out because there was no monitoring system. CFG machine is a type of rotating machinery, basically all rotating machines produce vibration which is a function of the alignment and balance of the rotating components. Vibration intensity measurements can provide information about the accuracy of shaft alignment and balance. Therefore, PT Suntory Garuda Beverage implements a predictive maintenance system on CFG engines with vibration monitoring. Vibration measurement data can be used to identify damage that occurs in CFG engines. Identification of CFG engine damage is done by analyzing the vibration spectrum which has different characteristics for each type of damage. The process of vibrational spectrum analysis can be carried out using the Fast Fourier Transform (FFT). Frequency domain analysis makes it easier to identify the amplitude and frequency of the vibration spectrum, so that more detailed information is obtained to identify the type of damage to the CFG machine. This study also uses a Neural Network (NN) to classify the type of CFG engine damage. In this research, 18 experiments were carried out to determine the best NN architecture. Based on testing, the best architecture uses 10 neurons in the hidden layer, and uses 50 iterations, which gets an MSE value of 0.045. The system created can classify the types of CFG engine damage in 4 conditions, namely normal, unbalance, misalignment, and looseness with an accuracy value of up to 100%

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Fast Fourier Transform, Motor, Neural Network, Vibration analysis
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ1077 Lubrication and lubricants. T Technology > TJ Mechanical engineering and machinery > TJ1398 Conveyors T Technology > TJ Mechanical engineering and machinery > TJ174 Maintenance and repair of machinery T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2785 Electric motors, Induction. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK351 Electric measurements. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5103.2 Wireless communication systems. Two way wireless communication T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.888 Web sites--Design. Web site development. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7870.23 Reliability. Failures T Technology > TS Manufactures > TS195 Packaging
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Muhammad Dwi Prasetyo
Date Deposited:	08 Aug 2023 01:23
Last Modified:	08 Aug 2023 01:23
URI:	http://repository.its.ac.id/id/eprint/104025

Actions (login required)

View Item