سيدسروش اقائي ميبدي

A high-impedance fault refers to a type of fault in power distribution systems that occurs when an energized conductor comes into contact with a high-resistance object, such as a tree branch, dry ground, concrete, or other insulating materials. In these conditions, the fault current is very small, and due to its low magnitude and the nonlinear and complex nature of this current, detecting it using traditional protective methods becomes very challenging. In this thesis, an algorithm is examined to detect high-impedance faults in power distribution systems, based on mathematical morphology and using the zero-sequence voltage signal of the network as input. In the field of high-impedance fault detection, mathematical morphology is used to process and analyze voltage or current signals. This technique allows for extracting key features from the signal that indicate the occurrence of a fault, and distinguishing small changes or nonlinear disturbances caused by faults from noise and normal variations in the system. Due to the similarity of some primary characteristics of high-impedance faults with several natural events in the power distribution system, this algorithm may result in false detection. To enhance its accuracy, various filters are employed to eliminate frequencies above 700 Hz and odd harmonics from the system. This method has been tested on the IEEE 13-node test system, which includes overhead lines, underground cables, and unbalanced loads at a voltage of 4.16 kV. This system enables the simulation of real-world conditions in distribution networks, and the performance of the algorithm in detecting high-impedance faults and nonlinear conditions has been comprehensively eva‎luated. The enhanced algorithm performs correctly during fault conditions on various surfaces and at different distances from the detection point, while natural system events such as switching of nonlinear loads and capacitors, unbalanced loads, and transformer energization do not affect its operation. Additionally, noise introduced to the input signal does not lead to false detection of high-impedance faults. Due to its low sampling rate for time and frequency domain signal analysis, this method has a low computational burden, and the calculation and fault detection times are acceptable.

تشخيص خطا امپدانس بالا , ريخت‌شناسي رياضي , جريان توالي صفر , هارمونيك فرد , رويدادهاي طبيعي سيستم , حوزه زمان

High Impedance Fault Detection , Mathematical morphology , Zero-sequence current , Odd harmonics , Natural system events , Time domain

Seyyed Soroosh Aghaei

Dr. Seyed Mohammad Shahrtash