Sistem Deteksi Keamanan Manusia Terhadap Konveyor Teleskopik di Gudang Urea Menggunakan Metode Convolutional Neural Network (CNN)

Aji, Ardha Nurhaskara (2025) Sistem Deteksi Keamanan Manusia Terhadap Konveyor Teleskopik di Gudang Urea Menggunakan Metode Convolutional Neural Network (CNN). Other thesis, Institut Teknologi Sepuluh Nopember.

Text 2040211048-Undergraduate _Thesis.pdf Restricted to Repository staff only Download (7MB) | Request a copy

Abstract

Konveyor teleskopik merupakan alat bantu pemindahan barang yang berperan penting dalam proses pemuatan pupuk di gudang Urea 1B PT Petrokimia Gresik. Interaksi langsung antara pekerja dan konveyor yang terus bergerak menimbulkan potensi kecelakaan kerja yang signifikan. Penelitian ini bertujuan untuk merancang sistem peringatan keselamatan berbasis visi komputer menggunakan metode Convolutional Neural Network (CNN) dengan algoritma You Only Look Once versi 8 (YOLOv8) guna mendeteksi keberadaan manusia di zona bahaya konveyor secara waktu nyata (real-time). Sistem memanfaatkan kamera IP untuk menangkap citra video, yang kemudian dianalisis oleh model CNN guna mendeteksi objek manusia dan konveyor. Deteksi bahaya ditentukan berdasarkan tumpang tindih antara posisi manusia dan area konveyor. Jika kondisi bahaya teridentifikasi, sistem akan mengirimkan perintah melalui protokol MQTT ke mikrokontroler ESP32 untuk mengaktifkan lampu dan buzzer peringatan. Hasil pengujian menunjukkan bahwa model mampu mendeteksi manusia dan konveyor dengan akurasi masing-masing sebesar 90% dan 100%. Sistem juga mampu mengidentifikasi kondisi bahaya dengan akurasi 87% dan rata-rata waktu respons sebesar 0,32 detik. Dengan performa tersebut, sistem ini dapat digunakan sebagai langkah preventif untuk memperingatkan keberadaan manusia di sekitar area konveyor teleskopik.
===============================================================================================================================
The telescopic Conveyor is a material handling tool that plays a crucial role in the fertilizer loading process at the Urea 1B warehouse of PT Petrokimia Gresik. However, direct interaction between workers and the continuously moving Conveyor poses a significant risk of occupational accidents. This study aims to design a computer vision-based safety warning system using the Convolutional Neural Network (CNN) method with You Only Look Once version 8 (YOLOv8) algortihm to detect the presence of humans in hazardous zones around the Conveyor in real time. The system utilizes an IP Camera to capture video footage, which is then analyzed by a (CNN) model to detect human and conveyor objects. Hazard detection is determined based on the overlap between the detected human position and the Conveyor area. When a hazardous condition is identified, the system transmits a command via the MQTT Protocol to an ESP32 microcontroller to activate a warning light and buzzer. Test results show that the model can detect humans and conveyors with accuracies of 90% and 100%, respectively. The system also achieves 87% accuracy in hazard detection with an average response time of 0,32 seconds. With this performance, the system can be used as a preventive tool to provide warnings regarding human presence near telescopic Conveyor areas.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	CNN, deteksi objek, keselamatan kerja, konveyor teleskopik, sistem peringatan, YOLOv8
Subjects:	T Technology > T Technology (General) > T55 Industrial Safety T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques. Image analysis--Data processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.585 TCP/IP (Computer network protocol) T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK6592.A9 Automatic tracking. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Ardha Nurhaskara Aji
Date Deposited:	07 Aug 2025 09:08
Last Modified:	07 Aug 2025 09:08
URI:	http://repository.its.ac.id/id/eprint/127968

Actions (login required)

View Item