الفهرس | Only 14 pages are availabe for public view |
Abstract The main purpose of this thesis is to determine the incipient faults that may occur within power transformers during operation. Power transformer during operation is subjected to different stresses such as electrical stress and thermal stress which lead to liberation of gases from the hydrocarbon mineral oil. In this thesis, two different proposed algorithms based on artificial intelligence techniques (fuzzy logic approach and neural network approach) are used to get the correct diagnosis of the incipient faults based on dissolved gas analysis. (DGA) in order to avoid the drawbacks of the conventional DGA methods. Each approach is built using MATLAB software. Design steps of each approach are explained. In this thesis. In order to examine the accuracy of the proposed artifical intelligence fault diagnosis techniques, various power transformers DGA results are tested. The accuracy of each approach is then calculated and compared with the accuracy of the conventional DGA methods. Five cases for study obtained from the Egyptian ministry of electricity is then used to test the two approaches |