Pengembangan Sistem Predictive Maintenance Motor Induksi 3 Phase pada Penggerak Flywheel Generator Berbasis Artificial Neural Network sebagai Upaya Meningkatkan Keandalan Operasional

Kosesar, Lintang Herinda (2026) Pengembangan Sistem Predictive Maintenance Motor Induksi 3 Phase pada Penggerak Flywheel Generator Berbasis Artificial Neural Network sebagai Upaya Meningkatkan Keandalan Operasional. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2042221017-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (13MB) | Request a copy

Abstract

Motor induksi tiga phase digunakan sebagai penggerak utama flywheel generator di Laboratorium Robotika ITS. Namun, sistem pemantauan kondisi motor untuk mendukung predictive maintenance masih belum tersedia secara optimal. Pemeliharaan yang dilakukan bersifat manual dan kurang mampu mendeteksi gejala awal kerusakan, sehingga berpotensi menimbulkan gangguan operasional serta meningkatkan biaya perawatan. Penelitian ini mengembangkan sistem predictive maintenance berbasis neural network dengan metode Lead Time to Repair, menggunakan sensor terhubung ke PLC Siemens S7-1200 (tegangan 220 V, arus 0–3 A, RTD PT100, proximity, velocity 0–50 Hz, vibrasi 0–10 mm/s). Dataset terdiri dari 1000 data pelatihan dan 400 data pengujian (250 per kelas), menghasilkan akurasi 98,75% dan training loss 0.0384 sebesar, dengan database mampu menyimpan 200.000 data. Sistem menampilkan website monitoring real-time (refresh 1 s), menyimpan history data, dan menampilkan jadwal maintenance saat kerusakan. Rekayasa kerusakan meliputi Unbalance (lonjakan harmonik 1x) dan Mechanical Looseness (lonjakan harmonik 1x–3x acak), dengan RMS tertinggi 8 mm/s pada 30 Hz.
===================================================================================================================================
A three-phase induction motor is used as the main driver of the flywheel generator in the ITS Robotics Laboratory. However, the motor condition monitoring system for supporting predictive maintenance is still not optimally available. Maintenance is performed manually and is unable to detect early signs of failure, potentially causing operational disruptions and increasing maintenance costs. This study develops a neural network-based predictive maintenance system using the Lead Time to Repair method, with sensors connected to a Siemens S7-1200 PLC (voltage 220 V, current 0–3 A, RTD PT100, proximity, velocity 0–50 Hz, vibration 0–10 mm/s). The dataset consists of 1000 training data and 400 testing data (250 per class), achieving 98,75% accuracy and training loss 0.0384, with a database capable of storing 200,000 data points. The system provides a real-time monitoring website (1 s refresh), stores history data, and displays a maintenance schedule when faults occur. Fault simulation includes Unbalance (spike at 1× harmonic) and Mechanical Looseness (random spikes at 1×–3× harmonics), with a maximum RMS of 8 mm/s at 30 Hz.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Artificial Neural Network, Internet of Things, Predictive Maintenance
Subjects:	T Technology > T Technology (General) > T11 Technical writing. Scientific Writing T Technology > TA Engineering (General). Civil engineering (General) > TA355 Vibration. T Technology > TA Engineering (General). Civil engineering (General) > TA593.35 Instruments, cameras, etc. T Technology > TJ Mechanical engineering and machinery > TJ212 Control engineering systems. Automatic machinery (General) T Technology > TJ Mechanical engineering and machinery > TJ213 Automatic control. T Technology > TJ Mechanical engineering and machinery > TJ223.P76 Programmable controllers T Technology > TJ Mechanical engineering and machinery > TJ230 Machine design T Technology > TJ Mechanical engineering and machinery > TJ541 Flywheels. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2797 Motor-generator sets. Cascade T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK4055 Electric motor T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.585 TCP/IP (Computer network protocol) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7888.3 Digital computers
Divisions:	Faculty of Vocational > Instrumentation Engineering
Depositing User:	Lintang Herinda Kosesar
Date Deposited:	31 Jul 2026 01:05
Last Modified:	31 Jul 2026 01:07
URI:	http://repository.its.ac.id/id/eprint/140105

Actions (login required)

View Item