Sistem Klasifikasi Kerusakan Misalignment Pada Motor Brushless Direct Current (BLDC) Berdasarakan Getaran Menggunakan Metode Artificial Neural Network

Ardiansyah, Rizky (2024) Sistem Klasifikasi Kerusakan Misalignment Pada Motor Brushless Direct Current (BLDC) Berdasarakan Getaran Menggunakan Metode Artificial Neural Network. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2040201058-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (6MB) | Request a copy

Abstract

Misalignment merupakan penyebab utama kerusakan pada rotating equipment, yang tidak hanya memperpendek umur mesin tetapi juga menurunkan kinerja motor. Pada pembangkit listrik, misalignment sering terjadi pada area boiler, turbin, dan sistem umum akibat permasalahan assembly setelah perbaikan. PLN Nusantara Power Services menggunakan motor BLDC untuk mensimulasikan kerusakan motor Boiler Feed Water Pump (BFWP) guna memahami dan mengatasi masalah misalignment. Untuk mengatasi masalah pada Proyek Akhir ini, akan dibuat sistem monitoring getaran untuk mendeteksi misalignment pada motor Brushless Direct Current (BLDC) menggunakan metode Artificial Neural Network (ANN). ANN dipilih karena kemampuannya untuk mempelajari pola-pola kompleks dari data dengan tingkat akurasi yang tinggi. Sistem ini dirancang untuk memonitor getaran motor BLDC secara kontinu dan mengidentifikasi jenis misalignment yang terjadi, termasuk misalignment parallel dan misalignment angular. Metode Fast Fourier Transform (FFT) diterapkan untuk menganalisis frekuensi sinyal getaran, mengubahnya dari domain waktu ke domain frekuensi, dan mengidentifikasi komponen frekuensi yang signifikan terkait dengan misalignment. Data velocity digunakan untuk melatih model ANN. Berdasarkan hasil pengujian model ANN terbaik menggunakan pembagian data training 80% dan data testing sebesar 20% dengan hidden layer sebanyak 20 neuron dan 100 epoch. Hasil penelitian menunjukkan bahwa dengan menggunakan arsitektur neural network, sistem ini dapat mengklasifikasi jenis kerusakan misalignment dengan menggunakan motor BLDC dengan tingkat akurasi mencapai 98.67% dan nilai loss sebesar 0,0713%. Dataset diklasifikasikan menjadi 3 jenis kondisi yaitu normal, misalignment angular, dan misalignment parallel.
==================================================================================================================================
Misalignment is a major cause of damage to rotating equipment, which not only shortens the life of the machine but also reduces motor performance. In power plants, misalignment often occurs in the boiler area, turbine, and general system due to assembly problems after repair. PLN Nusantara Power Services uses BLDC motors to check Boiler Feed Water Pump (BFWP) motor damage to understand and overcome misalignment problems. To overcome the problems in this Final Project, a vibration monitoring system will be created to detect misalignment in Brushless Direct Current (BLDC) motors using the Artificial Neural Network (ANN) method. ANN is chosen because of its ability to learn complex patterns from data with a high degree of accuracy. This system is designed to continuously monitor BLDC motor vibration and identify the types of misalignment that occur, including parallel misalignment and angle misalignment. The Fast Fourier Transform (FFT) method is applied to analyze the vibration of the frequency signal, convert from the time domain to the frequency domain, and identify significant frequency components related to misalignment. The data rate is used to train the ANN model. Based on the results of testing the best ANN model using 80% training data and 20% testing data with a hidden layer of 20 neurons and 100 epochs. The results show that by using a neural network architecture, this system can classify the type of misalignment damage using a BLDC motor with an accuracy level of 98.67% and a loss value of 0.0713%. The dataset is classified into 3 types of conditions, namely normal, angular misalignment, and parallel misalignment.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Fast Fourier Transform, Getaran, Misalignment, Motor BLDC, Neural Network, BLDC Motor, Misalignment, Vibration.
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL152.5 Motor vehicles Driving T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL214.T87 Automobiles--Motors--Turbochargers
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Rizky Ardiansyah
Date Deposited:	26 Aug 2024 04:12
Last Modified:	26 Aug 2024 04:12
URI:	http://repository.its.ac.id/id/eprint/115526

Actions (login required)

View Item