Utama, Harris Fikri Satria Utama (2025) Sistem Identifikasi Kerusakan Bearing Berbasis Frekuensi Karakteristik Dan Klasifikasi Kondisi Gearset Menggunakan Metode K-Nearest Neighbors (KNN) Pada Roll Konveyor Robo Arm Palletizer. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 2040211009-Undergraduate _Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
2040211009-Undergraduate _Thesis.pdf Restricted to Repository staff only Download (7MB) | Request a copy |
Abstract
Bearing pada sistem konveyor memiliki peran penting dalam menjaga kelancaran operasional, terutama pada Robo Palletizer Fuji yang digunakan di Departemen Pergudangan dan Pengantongan PT. Petrokimia Gresik. Penelitian ini bertujuan untuk merancang dan mengembangkan sistem klasifikasi kerusakan bearing dan gearset menggunakan metode K-Nearest Neighbors (KNN). Parameter utama yang digunakan meliputi data getaran dan kecepatan rotasi (RPM). Sistem ini diimplementasikan dalam lingkungan industri dan dirancang untuk mengidentifikasi kondisi kerusakan komponen secara otomatis berdasarkan data real-time dari sensor. Klasifikasi kerusakan gearset dilakukan dengan metode KNN menggunakan parameter RPM motor dan RPM roll. Sementara itu, deteksi kerusakan bearing dilakukan dengan pendekatan berbasis rumus frekuensi karakteristik (BPFO, BPFI, BSF, dan FTF) yang dihitung dari data getaran. Sistem diuji menggunakan sekitar 6.500 data yang diperoleh selama kegiatan magang di PT Petrokimia Gresik dan dievaluasi berdasarkan akurasi klasifikasinya. Penelitian ini diharapkan dapat membantu proses identifikasi kerusakan komponen secara lebih cepat dan tepat serta mendukung penerapan teknologi Industry 4.0 di sektor manufaktur.
================================================================================================================================
Bearings in conveyor systems play a crucial role in ensuring smooth operations, particulary in the Robo Palletizer Fuji used at the Warehousing and Packaging Department of PT. Petrokimia Gresik. This research aims to design and develop a classification system for bearing and gearset damage using the K-Nearest Neighbors (KNN) algorithm. The system utilizes key parameters such as vibration data and rotational speed (RPM) to determine the condition of each component. The system is implemented in an industrial setting and is designed to automatically classify the condition of components based on real-time sensor data. Gearset damage classification is conducted using KNN with input features including motor and roll RPM, while bearing damage detection is performed through a formula-based approach using characteristic fault frequencies (BPFO, BPFI, BSF, and FTF) derived from vibration signals. Approxiamately 6,500 data samples were collected during an internship at PT.Petrokimia Gresik and used to evaluate the accuracy of the classification system. The research is expected to assist in faster and more precise identification of component fault and support the adoption of Industry 4.0 technologies in the manufacturing sector.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Bearing, Gearset, Konveyor, K-Nearest Neighbors, Klasifikasi Kerusakan, Conveyor, Industry 4.0, Fault Classification |
| Subjects: | T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.546 Computer algorithms |
| Divisions: | Faculty of Vocational > 36304-Automation Electronic Engineering |
| Depositing User: | Harris Fikri Satria Utama |
| Date Deposited: | 07 Aug 2025 04:07 |
| Last Modified: | 07 Aug 2025 04:07 |
| URI: | http://repository.its.ac.id/id/eprint/127915 |
Actions (login required)
 |
View Item |