Studi Klasifikasi Hotspot Peralatan pada Bay Penghantar Waru 2 akibat Pemadaman Bay Penghantar Sambikerep-Waru 1 di GI Sambikerep Berbasis Metode Fuzzy Tsukamoto

Penelitian ini membahas klasifikasi hotspot pada peralatan Bay Penghantar Waru 2 di Gardu Induk Sambikerep akibat pemadaman Bay Penghantar Waru 1 yang menyebabkan pelimpahan beban ke Bay Waru 2. Kondisi tersebut berpotensi meningkatkan suhu pada titik klem dan konduktor. Pengukuran dilakukan menggunakan metode thermovisi dengan thermal imager Flir E50 untuk memperoleh data suhu klem, suhu konduktor, dan arus pengukuran. Data tersebut kemudian diolah menggunakan metode Fuzzy Tsukamoto melalui tahap fuzzifikasi, penyusunan aturan, perhitungan α-predikat, penentuan nilai z, dan defuzzifikasi. Hasil logika fuzzy menunjukkan bahwa 2,25% data termasuk kategori baik, 30,13%, kategori periksa saat pemeliharaan, dan 67,63% kategori rencana perbaikan. Tidak terdapat peralatan pada kategori perbaiki segera maupun darurat, sehingga kondisi Bay Penghantar Sambikerep–Waru 2 belum berada pada tingkat kritis. Setelah dilakukan pengujian akurasi dengan penilaian pakar, didapatkan bahwa akurasi logika fuzzy yang dibuat yaitu 88,95%.

==================================================================================

This study discusses hotspot classification on equipment in the Waru 2 Feeder Bay at Sambikerep Substation due to the shutdown of the Waru 1 Feeder Bay, which caused the load to be transferred to Waru 2. This condition has the potential to increase the temperature at clamp and conductor points. The measurement was carried out using the thermovision method with a Flir E50 thermal imager to obtain clamp temperature, conductor temperature, and measurement current data. The data were then processed using the Fuzzy Tsukamoto method through fuzzification, rule
formation, α-predicate calculation, z-value determination, and defuzzification. The fuzzy logic results show that 2.25% of the data were classified as good, 30.13% as inspection during maintenance, and 67.63% as repair planning. No equipment was classified as immediate repair or emergency, indicating that the condition of the Sambikerep–Waru 2 Feeder Bay equipment was not yet critical. After an accuracy test was conducted using expert assessment, the fuzzy logic model achieved an accuracy of 88.95%.