Hamida, Nazilah Salwah (2024) Implementasi Artificial Neural Network untuk Diagnosa Jenis Kerusakan Motor Pompa Sea Water Intake sebagai Sistem Peringatan Dini. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2040201090_Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2040201090_Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (8MB) | Request a copy |
Abstract
Penelitian ini bertujuan untuk mengembangkan sistem perawatan motor pompa Sea Water Intake di industri kimia PT. Polychem Indonesia Tbk. agar dapat mengurangi masalah penyebab kerusakan motor yang umum terjadi, yaitu ketidakseimbangan konstruksi akibat korosi pada baseplate motor dan keterlambatan respons terhadap kerusakan akibat jauhnya letak motor dari plant utama. Solusi yang diberikan untuk permasalahan ini adalah penerapan Sistem Peringatan Dini dengan menggunakan sensor akselerometer ADXL345 untuk membaca dan memantau nilai kecepatan getaran setiap waktu, serta menggunakan transformasi Fast Fourier Transform untuk mendapatkan frekuensi dari data kecepatan getaran. Nilai frekuensi ini akan menjadi masukan untuk model Artificial Neural Network, yang akan menghasilkan diagnosa jenis kerusakan motor tiga kondisi, yaitu normal, poros bengkok, dan kelonggaran mekanis. Penelitian ini menunjukkan bahwa penggunaan sensor akselerometer ADXL345 menghasilkan rata-rata kesalahan sebesar 1,132% untuk motor dengan kecepatan 1500 RPM dan 1,588% untuk motor dengan kecepatan 3000 RPM. Transformasi Fast Fourier Transform berhasil mendapatkan frekuensi dari nilai kecepatan getaran dengan beberapa prosedur, seperti teori Nyquist, penskalaan magnitudo spektrum, zero frequency offset, dan penentuan jumlah data yang perlu ditransformasikan. Model Artificial Neural Network yang dihasilkan mampu mendiagnosa jenis kerusakan motor dengan benar. Dengan demikian, sistem ini diharapkan dapat menjadi langkah awal dalam meningkatkan efektivitas pemeliharaan motor pompa Sea Water Intake, serta membantu meningkatkan kinerja dan mempermudah analisis kerusakan motor dalam industri tersebut.
=====================================================================================================================================
This research aims to develop a maintenance system for Sea Water Intake pump motors in the chemical industry at PT. Polychem Indonesia Tbk. to mitigate common motor damage issues, such as structural imbalance due to corrosion on the motor baseplate and delayed response to damage due to the motor's remote location from the main plant. The proposed solution involves implementing an Early Warning System using the ADXL345 accelerometer sensor to continuously monitor and read the vibration velocity values. Additionally, the Fast Fourier Transform is used to extract frequency values from the vibration velocity data. These frequency values serve as input data for the Artificial Neural Network model, which will diagnose the motor's condition for three states: normal, bent shaft, and mechanical looseness. The research findings indicate that using the ADXL345 accelerometer sensor results in an average error of 1.132% for motors operating at 1500 revolutions per minute and 1.588% for motors at 3000 revolutions per minute. The Fast Fourier Transform successfully retrieves the vibration frequency spectrum values from the vibration velocity data, following procedures such as the Nyquist theorem, spectrum magnitude scaling, zero frequency offset, and determining the amount of data to be transformed. The resulting Artificial Neural Network model can accurately diagnose the type of motor damage. Therefore, this system is expected to be a preliminary step in enhancing the effectiveness of Sea Water Intake pump motor maintenance, as well as improving performance and facilitating motor damage analysis in the industry.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Analisa Spektrum Getaran, Artificial Neural Network, Fast Fourier Transform, Sistem Peringatan Dini Artificial Neural Network, Early Warning System, Fast Fourier Transform, Vibration Spectrum Analysis |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TL Motor vehicles. Aeronautics. Astronautics > TL448 Electric motorcycles |
| Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
| Depositing User: | Nazilah Salwah Hamida |
| Date Deposited: | 28 Aug 2024 00:49 |
| Last Modified: | 28 Aug 2024 00:49 |
| URI: | http://repository.its.ac.id/id/eprint/115542 |
Actions (login required)
 |
View Item |