High impedance fault detection in distribution system using signal processing techniques

Abstract

Currently, the primary objective of power system engineers is to modernize the power system by building it more intelligent, consistent and sustainable. The integration of static, dynamic loads and constantly increasing electronic device based non-linear loads are changing the behavior of the power system network. Over the years, conventional protection schemes have been successfully used to detect and protect against the faults. However, some of these cases, faults cannot be detected by the protection system, since these types of faults are created when an energized conductor contacts high impedance object and takes low magnitude of current with random behavior. This type of fault is generally known as a High Impedance Fault (HIF). Fast and consistent detection of these faults in distribution system is a continual challenge for utility companies due to their behaviour. The objective of this thesis is to investigate the characteristics of the HIFs and develop a detection method. The initial phase of the work is to understand the HIF and find novel indicators that reveal the characteristics of the HIFs by analyzing fault current signal. Then, an electric power distribution system has been studied and accurately modelled with MATLAB/SIMULINK. This model is simulated over 1000 different cases which include HIF and non-HIF (Linear and Non-linear Load switching, Capacitance switching, Transformer inrush current) conditions to observe the current signal. The next phase is focused on the development of an HIF detection algorithm. newline