Title: A novel method for fault detection, classification and location in Ilam province's power transmission network
Authors: Yavar Azarakhsh; Jafar Tavoosi
Addresses: Department of Electrical Engineering, Islamic Azad University, Ilam Branch, Ilam, Iran ' Department of Electrical Engineering, Ilam University, Ilam, Iran
Abstract: Due to the importance of continued energy supplying to subscribers, especially industrial subscribers, who are severely affected by any disruption in network parameters and network outages, we decided to use new technology and software to reduce network outages and even reducing the shutdown time. In this study, equipment and relays are used to diagnose and clear faults in transmission networks based on a novel Hopfield neural network. Then, the 230 kV power transmission network of Ilam province and its ten substations (63 kV) and several power plants located in the province are modelled using DIgSILENT Power Factory software. In this regard, first, the load distribution in normal network and the values of network parameters are recorded, then the values of network parameters are recorded in the faulty network. These values are fed to a novel fuzzy Hopfield neural network as training data, the results are compared with initial values and it is observed that fuzzy Hopfield neural network can quickly and accurately locate network faults.
Keywords: fuzzy neural network; transmission network; fault diagnosis; fault location.
International Journal of Applied Pattern Recognition, 2021 Vol.6 No.4, pp.308 - 321
Received: 05 Jan 2021
Accepted: 02 May 2021
Published online: 11 Nov 2021 *