Neural Network Based Fault Detector and Classifier for Synchronous Generator Stator Windings.

Hatata, Ahmad; Helal, Ahmed; El Dessouki, Hesien; El-Saadawi, Magdi Mohamed Ali; Tantawy, Mohammed

doi:10.21608/bfemu.2020.122340

Neural Network Based Fault Detector and Classifier for Synchronous Generator Stator Windings.

Document Type : Research Studies

Authors

¹ Electrical Engineering Department., Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.

² Assistant Professor., Electrical and Control Engineering Department., Faculty of Engineering., Arab Academy for Science and Technology., Alex., Egypt.

³ Dept of Elec. Engineering and Control, Faculty of Engineering, Arab Academy for Science and Technology, Alex. Egypt.

⁴ Professor of Electrical Engineering Department., Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.

⁵ Professor of Electrical Engineering Department, Faculty of Engineering., El-Mansoura University., Mansoura., Egypt.

10.21608/bfemu.2020.122340

Abstract

This paper presents an application of multilayer feedforward neural network (MFNN) as a differential protection for synchronous generators. Two MFNN are designed, trained, and tested in this paper. The first one has two outputs which detect the internal and external fault state. The other neural network has four outputs to classify the faulty phases. The proposed neural fault detector and classifier were trained using various sets of data available from a selected synchronous model and simulating different fault scenarios (fault type, fault location, fault resistance and fault inception angle). The results show very good behavior of the MFNN and it was more reliable and accurate than conventional methods. It shows that MFNN offer the possibility to be used for on line synchronous generator protection and give satisfactory results.

Keywords

Main Subjects