Irsyad, Azhar (2026) PREDIKSI KERUSAKAN TRANSFORMATOR STEP-UP BERDASARKAN HASIL DISSOLVED GAS ANALYSIS DENGAN FUZZY DAN KNN DI PT INDONESIA POWER UBP SURALAYA. S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.

	Text (Fulltext) Irsyad Azhar_3332210083_Fulltext.pdf Restricted to Registered users only Download (3MB) | Request a copy
	Text (Bab 1) Irsyad Azhar_3332210083_01.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (Bab 2) Irsyad Azhar_3332210083_02.pdf Restricted to Registered users only Download (879kB) | Request a copy
	Text (Bab 3) Irsyad Azhar_3332210083_03.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (Bab 4) Irsyad Azhar_3332210083_04.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (Bab 5) Irsyad Azhar_3332210083_05.pdf Restricted to Registered users only Download (381kB) | Request a copy
	Text (Daftar Refrensi) Irsyad Azhar_3332210083_Ref.pdf Restricted to Registered users only Download (273kB) | Request a copy
	Text (Lampiran) Irsyad Azhar_3332210083_Lamp.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text (Cek Plagiasi) Irsyad Azhar_3332210083_CP.pdf Restricted to Registered users only Download (5MB) | Request a copy

Abstract

Step-Up transformers are critical components in electric power systems and require condition monitoring to prevent faults and failures. This study aims to analyze and predict types of power transformer faults based on the results of Dissolved Gas Analysis (DGA). The objects of this study are transformer step-up units 1 and 2 with a rated power of 470 MVA. The fault analysis methods employed are Roger’s Ratio and the Duval Triangle, utilizing the concentrations of H₂, CH₄, C₂H₂, C₂H₄, and C₂H₆ gases as determining variables. The analysis results indicate that transformer units 1 and 2 are identified as experiencing Partial Discharge and Thermal Faults, characterized by an increase in the Total Dissolved Combustible Gas (TDCG) value up to 646 ppm. Validation using fuzzy logic and the K-Nearest Neighbor (KNN) method demonstrates diagnostic results consistent with applicable standards, achieving an accuracy of 97% to 98%. Therefore, the proposed methods are proven to be effective for accurately diagnosing power transformer faults.

Actions (login required)

View Item