Development of Intelligent Fault Analysis and Reporting System for High Voltage System

Mohd Najeeb, Muhammad Addin (2024) Development of Intelligent Fault Analysis and Reporting System for High Voltage System. Final Year Project (Bachelor), Tunku Abdul Rahman University of Management and Technology.

Text MUHAMMAD ADDIN BIN MOHD NAJEEB-Full Text.pdf Restricted to Registered users only Download (1MB)

Abstract

As electrical transmission systems becoming increasingly complex and adapting digitalization, traditional methods such as the travelling wave method and impedance-based techniques for fault detection are facing limitations and drawbacks. This paper explores and presents the applications of using a hybrid intelligent system of an Adaptive Neuro-Fuzzy System (ANFIS) that combines learning capabilities of neural networks with the reasoning abilities of fuzzy logic to detect, classify and locate the fault within a transmission line. This advanced method has the potential to outperform traditional approaches in terms of accuracy and reliability. The study utilizes a 11kV transmission system with a 100km transmission line to collect data for developing the ANFIS model. The system is simulated in MATLAB/Simulink under various fault scenarios at different locations. Two separate ANFIS models are developed for fault detection and localization and their performance is tested with another 11kV network. To validate the models, mean absolute error for each type of fault is calculated and compared for all faults which includes single line to ground fault (SLG), double line to ground fault (DLG), line to line fault (LL) and three phase faults (3LG). The results demonstrate that model can accurately detect the fault type with high level of accuracy and can correctly identify all types of faults. The fault localization also achieved an accuracy of 97.63% with an average fault location error of 3%. Consequently, the proposed ANFIS model show promise for implementation in real-time protection systems.