Claim Missing Document
Check
Articles

Found 1 Documents
Search

IoT-Based Image Segmentation and Fuzzy Mamdani for Ceramic Insulator Contaminant Identification Rahul Maryulis Putra; Yusreni Warmi; Arif Handika
Jurnal Ilmiah Telsinas Vol 9 No 1 (2026)
Publisher : Universitas Pendidikan Nasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.38043/telsinas.v9i1.7071

Abstract

Ceramic insulators on 150 kV transmission lines are prone to contamination from dust and moss, which can degrade insulation performance and increase the risk of system failure. This study aims to identify insulator contaminants using image processing techniques and Mamdani fuzzy logic integrated with an Internet of Things (IoT) platform to assess the dielectric strength of 150 kV ceramic insulators. The method involves capturing images of the insulator surface using a monocular digital microscope, applying OpenCV-based image processing for segmentation and quantification of contamination, and classifying contamination levels using Mamdani fuzzy logic. The results show that the system can accurately detect the percentage of dust and moss contamination. The processed image data are used as input to the fuzzy system to determine the dielectric strength of the insulator. Simulations demonstrate consistent classification of contamination into three levels: low, medium, and high. The system is integrated with the Ubidots IoT platform for real-time monitoring of insulator conditions. As contamination levels increase, the insulator's breakdown voltage decreases, with dust having a greater impact than moss, especially under high humidity conditions. The dielectric strength ranges from 45 kV to 110 kV, with values below 60 kV considered critical.