Perancangan Sistem Monitoring Kinerja Welder pada Aktivitas Pengelasan Fillet Joint Berbasis Wearable Device

Gemilang, Argya (2023) Perancangan Sistem Monitoring Kinerja Welder pada Aktivitas Pengelasan Fillet Joint Berbasis Wearable Device. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 04111940000038-Undergraduated_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 September 2025. Download (5MB) | Request a copy

Abstract

Pada industri perkapalan pengelasan merupakan bagian yang penting dalam proses pembangunan kapal. Keterlambatan pembangunan kapal sering terjadi dikarenakan salah satu faktornya yaitu kinerja welder yang kurang baik. Pengawasan terhadap aktivitas pengelasan masih dilakukan secara manual dan tidak secara menyeluruh. Oleh karena itu penelitian ini dilakukan dengan tujuan merancang sistem monitoring kinerja welder pada aktivitas pengelasan fillet joint berbasis teknologi wearable device untuk monitoring aktivitas dan kinerja welder pada saat melakukan pengelasan. Hal yang dilakukan pertama adalah melakukan observasi terhadap kondisi terkini dari pengelasan fillet joint. Kedua melakukan eksperimen pengambilan data menggunakan alat wearable device MetaMotion, yang dilengkapi oleh sensor accelerometer, gyroscope, dan magnetometer. Ketiga melakukan pengolahan data dan analisis data menggunakan machine learning dengan algoritma support vector machine dan menggunakan deep learning dengan algoritma Convolutional Neural Network-Long Short Term Memory (CNN-LSTM) untuk data aktivitas pengelasan dan kinerja welder. Hasil pelatihan data aktivitas pengelasan melalui SVM memiliki tingkat akurasi sebesar >82% , hasil pelatihan data kinerja welder berdasarkan beberapa kriteria memiliki tingkat akurasi sebesar >90%. Hasil pelatihan data aktivitas pengelasan degan CNN-LSTM berdasarkan beberapa variasi data memiliki tingkat akurasi sebesar>90%, pelatihan data kinerja welder menggunakan CNN-LSTM memiliki nilai akurasi sebesar>95%
====================================================================================================================================
In the shipbuilding industry, welding plays a crucial role in the ship construction process. Delay in shipbuilding often occurs due to various factors, and one of them is the subpar performance of welders. Welding activities are currently supervised manually and not comprehensively. Therefore, this research aims to design a monitoring system for welder performance during fillet joint welding activities based on wearable device technology to monitor their activities and performance during welding.The first step involved in this research is to observe the current condition of fillet joint welding. Secondly, data collection experiments are conducted using the wearable device MetaMotion, equipped with accelerometer, gyroscope, and magnetometer sensors. Thirdly, data processing and analysis are performed using machine learning with the Support Vector Machine (SVM) algorithm for welding activity data and the Convolutional Neural Network-Long Short Term Memory (CNN-LSTM) algorithm for both welding activity and welder performance data.The results of training welding activity data through SVM show an accuracy rate of >82%, while the training of welder performance data based on several criteria achieves an accuracy rate of >90%. On the other hand, the training of welding activity data using CNN-LSTM with several variations of data achieves an accuracy rate of >90%, and the training of welder performance data using CNN-LSTM reaches an accuracy rate of >95%

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Sistem monitoring, kinerja welder, aktivitas pengelasan, Support vector machine, Convolutional Neural Network-Long Short Term Memory; Monitoring system, welder performance, welding activity, Support vector machine, Convolutional Neural Network-Long Short Term Memory
Subjects:	V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering > VM298.5 Shipbuilding industri. Shipyards
Divisions:	Faculty of Marine Technology (MARTECH) > Naval Architecture and Shipbuilding Engineering > 36201-(S1) Undergraduate Thesis
Depositing User:	Argya Gemilang
Date Deposited:	18 Aug 2023 03:51
Last Modified:	18 Aug 2023 03:51
URI:	http://repository.its.ac.id/id/eprint/104190

Actions (login required)

View Item