Implementasi Logika Fuzzy Tipe-2 untuk Mengatasi Ketidakpastian dalam Identifikasi Gangguan Transformator Daya Berbasis DGA

Dachi, Ichsan Azis (2025) Implementasi Logika Fuzzy Tipe-2 untuk Mengatasi Ketidakpastian dalam Identifikasi Gangguan Transformator Daya Berbasis DGA. S2 thesis, Universitas Kristen Indonesia.

	Text (Hal_Judul_Abstrak_Daftar_Isi_Daftar_Gambar_Daftar_Tabel_Daftar_Lampiran) HalJudulAbstrakDaftarisiDaftarGambarDaftarTabelDaftarLampiran.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (BAB_I) BABI.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (840kB)
	Text (BAB_II) BABII.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (BAB_III) BABIII.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (962kB)
	Text (BAB_IV) BABIV.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (4MB)
	Text (BAB_V) BABV.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (744kB)
	Text (Daftar_Pustaka) DaftarPustaka.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (652kB)

Abstract

Transformator daya merupakan komponen vital dalam sistem tenaga listrik, di mana Analisis Gas Terlarut (DGA) menjadi metode utama untuk diagnosis dini kegagalan. Namun, interpretasi DGA secara tradisional dipenuhi oleh ketidakpastian dan ambiguitas data yang tinggi, yang menyebabkan metode konvensional dan Logika Fuzzy Tipe-1 (T1FLS) memiliki keterbatasan akurasi. Penelitian ini mengusulkan implementasi sistem diagnostik cerdas berbasis Interval Type-2 Fuzzy Logic System (IT2FLS) untuk mengatasi masalah ketidakpastian tersebut. Sistem ini dirancang dengan mengintegrasikan basis aturan dari metode standar industri (Duval Pentagon dan Key Gas) dan memanfaatkan konsep Footprint of Uncertainty (FOU) untuk memodelkan ketidakpastian data secara eksplisit. Kinerja sistem divalidasi menggunakan dua set data industri (total 94 sampel) dan dibandingkan secara langsung dengan T1FLS. Hasil penelitian menunjukkan keunggulan signifikan IT2FLS, yang mencapai akurasi diagnostik 100% pada data Industri 1, mengungguli T1FLS (91.7%). Pada data Industri 2, kedua sistem menunjukkan akurasi 97.1%. Secara kualitatif, IT2FLS terbukti lebih robust dalam menangani kasus-kasus ambigu, dengan memberikan diagnosis bernuansa yang benar di mana T1FLS gagal. Penelitian ini menyimpulkan bahwa IT2FLS adalah metode yang lebih andal dan kuat untuk interpretasi DGA, menjadikannya alat bantu keputusan yang superior untuk pemeliharaan transformator daya berbasis kondisi. Kata Kunci: Logika Fuzzy Tipe-2 (IT2FLS), Analisis Gas Terlarut (DGA), Transformator Daya, Diagnosis Gangguan, Ketidakpastian / Power transformers are vital components in electrical power systems, where Dissolved Gas Analysis (DGA) serves as a primary method for early fault diagnosis. However, traditional DGA interpretation is fraught with high data uncertainty and ambiguity, causing conventional methods and Type-1 Fuzzy Logic (T1FLS) to have accuracy limitations. This research proposes the implementation of an intelligent diagnostic system based on an Interval Type-2 Fuzzy Logic System (IT2FLS) to address this uncertainty problem. This system is designed by integrating rule bases from industrystandard methods (Duval Pentagon and Key Gas) and utilizing the Footprint of Uncertainty (FOU) concept to explicitly model data uncertainty. The system's performance was validated using two industrial datasets (94 samples total) and compared directly with T1FLS. The results show the significant superiority of IT2FLS, which achieved 100% diagnostic accuracy on Industrial Data 1, outperforming T1FLS (91.7%). On Industrial Data 2, both systems demonstrated 97.1% accuracy. Qualitatively, IT2FLS proved to be more robust in handling ambiguous cases, providing correct, nuanced diagnoses where T1FLS failed. This study concludes that IT2FLS is a more reliable and robust method for DGA interpretation, making it a superior decision-support tool for condition-based maintenance of power transformers. Keywords: Type-2 Fuzzy Logic (IT2FLS), Dissolved Gas Analysis (DGA), Power Transformers, Fault Diagnosis, Uncertainty

Actions (login required)

View Item