Analisa Indeks Kesehatan Transformator Daya Berbasis Neural Network Untuk Mereduksi Jumlah Test Pada Trafo

Sutaryono, Gaffar Lutfi (2015) Analisa Indeks Kesehatan Transformator Daya Berbasis Neural Network Untuk Mereduksi Jumlah Test Pada Trafo. Undergraduate thesis, Institut Technology Sepuluh Nopember.

[img]
Preview
Text
2209100015-Undergraduate Theses.pdf - Published Version

Download (2MB) | Preview

Abstract

Transformator daya merupakan salah satu peralatan listrik yang mempunyai peran sentral dan kritis pada sistem tenaga listrik. Untuk menghindari kegagalan pada transformator daya, dibutuhkan antisipasi dengan pemantauan dan pemeliharaan kondisi transformator. Penelitian ini menyajikan informasi mengenai diagnosis kondisi transformator berdasarkan metode Indeks Kesehatan Transformator dengan menggunakan Neural Network. Metode Indeks Kesehatan Transformator memberikan penilaian kondisi transformator secara komprehensif. Analisis gas-gas terlarut, minyak isolasi trafo, dan furan (kertas isolasi) dilakukan untuk membagi penilaian kondisi transformator daya menjadi beberapa kategori, sesuai dengan prediksi umur operasi transformator dan level penurunan kondisi komponen transformator. Target dari tugas akhir ini adalah mendesain sistem diagnosis berbasis Neural Network dengan input salah satu dari ketiga test yang dilakukan pada trafo. Diharapkan dengan hanya menggunakan satu dari tiga jenis test, maka proses diagnosis kondisi trafo menjadi lebih mudah dan murah. ============================================================================================================= Power transformer is one of the electrical appliances that have a central and critical role in power system. To avoid failure in the power transformer, it takes anticipation with transformer condition monitoring and maintenance. This study presents information on the diagnosis of the condition of the transformer using Transformer Health Index method by based on Neural Network. Transformer Health Index Method provides a comprehensive transformer’s condition assessment. This method divides the power transformer condition assessment into several categories, according to the prediction of the operating life of the transformer and the level of deterioration of transformer components. Analysis of dissolved gases, transformer insulating oil, and furans (paper insulation) was conducted to determine the type of failure that occurred in the transformer. The results of this analysis are operating life prediction, the type of the possibility of failure, and action recommendations for next transformer maintenance. The target of this final project is to design a neural network-based diagnosis system with the input one of the three tests conducted on the transformer. Using only one of three types of test, the diagnosis of the condition of the transformer is expected to become easier and cheaper.

Item Type: Thesis (Undergraduate)
Additional Information: RSE 621.314 Sut a
Uncontrolled Keywords: Analisis, diagnosis, indeks kesehatan transfomator, neural network, transformator daya
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK201 Electric Power Transmission
Divisions: Faculty of Industrial Technology > Electrical Engineering > (S2) Master Theses
Depositing User: Mr. Tondo Indra Nyata
Date Deposited: 28 Jun 2018 02:00
Last Modified: 28 Jun 2018 02:00
URI: http://repository.its.ac.id/id/eprint/52051

Actions (login required)

View Item View Item