Diagnosa Kegagalan Pada Rotating Machinery Menggunakan Jaringan Syaraf Tiruan

Mas'udi, Mohammad (2008) Diagnosa Kegagalan Pada Rotating Machinery Menggunakan Jaringan Syaraf Tiruan. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2103100079-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (47MB) | Request a copy

Abstract

Kegagalan pada rotating machinery dapat menurunkan kinerja mesin dan apabila tidak ditangani akan mengakibatkan kerusakan. Untuk mengembalikan kinerja mesin, diperlukan diagnosa kegagalan mesin sebelum menentukan perbaikan. Salah satu metode yang digunakan adalah Fast Fourier Transform (FFT), metode ini mengubah data vibrasi domain waktu menjadi domain frekuensi sehingga dapat diketahui ketidaknormalan kondisi mesin. Untuk mendapatkan data vibrasi, dilakukan eksperimen pengambilan sinyal vibrasi dalam berbagai kondisi kegagalan mesin. Mesin yang digunakan dalam penelitian ini terdiri dari motor, kopling, bantalan, dan roda gigi. Sinyal vibrasi yang diperoleh ditransfomasikan menjadi data vibrasi domain frekuensi menggunakan FFT. Untuk menentukan jenis kegagalan yang akurat, maka digunakan aplikasi Jaringan Syaraf Tiruan (JST). Fungsi JST ini adalah malakukan proses identifikasi sinyal vibrasi domain frekuensi dengan cara membandingkan sinyal vibrasi yang baru dengan sinyal hasil training. Hasil eksperimen penyusunan JST, diperoleh jaringan type backpropagation, komposisi 3 layer dengan (512 input, 70 hidden, 6 output) neuron. Struktur tersebut telah berhasil mengidentifikasi failure yang terdiri dari misalignment, unbalance, cacat bantalan, cacat roda gigi, dan kombinasinya dengan toleransi error sebesar 10
====================================================================================================================================
Failure in rotating machinery can reduce machine performance and if left untreated will result in damage. To restore machine performance, machine failure diagnosis is required before determining repairs. One method used is the Fast Fourier Transform (FFT), this method converts time domain vibration data into the frequency domain so that abnormal machine conditions can be identified. To obtain vibration data, experiments were conducted to capture vibration signals under various machine failure conditions. The machine used in this study consists of a motor, clutch, bearing, and gear. The obtained vibration signals were transformed into frequency domain vibration data using the FFT. To determine the type of failure accurately, an Artificial Neural Network (ANN) application was used. The function of this ANN is to carry out the process of identifying frequency domain vibration signals by comparing new vibration signals with the training results signal. The results of the ANN compilation experiment obtained a backpropagation type network, a composition of 3 layers with (512 input, 70 hidden, 6 output) neurons. The structure has successfully identified failures consisting of misalignment, unbalance, bearing defects, gear defects, and their combinations with an error tolerance of 10

Item Type:	Thesis (Other)
Additional Information:	RSM 621.802 856 32 Mas d-1 2008 (weeding)
Uncontrolled Keywords:	rotating machinery, vibrasi, FFT, JST; rotating machinery, vibration, FFF, NN
Subjects:	T Technology > TJ Mechanical engineering and machinery > TJ1058 Rotors T Technology > TJ Mechanical engineering and machinery > TJ217.6 Predictive Control
Divisions:	Faculty of Industrial Technology > Mechanical Engineering > 21201-(S1) Undergraduate Thesis
Depositing User:	EKO BUDI RAHARJO
Date Deposited:	17 Nov 2025 04:31
Last Modified:	17 Nov 2025 04:31
URI:	http://repository.its.ac.id/id/eprint/128796

Actions (login required)

View Item