ACCURATE FAULT LOCATION IN UNDERGROUND CABLES USING ADVANCED SIGNAL PROCESSING TECHNIQUES

Abstract

Underground cables are vital links in power systems that achieve the essential continuity of service for reliable transmission and distribution. From view of economics and power quality, the ultimate goal for a power system is to supply a reliable and stable power to customs. However, it is inevitable that underground power cables are exposed to faults. The importance of acc\lrate fault location techniques in power cables is increasing for fast repair and power system restoration. Recently, a significant amount of research work has been directed to address the need for gentler and more accurate cable fault locating methods. In this thesis, two power cable fault location schemes are proposed. The main advantage of the proposed schemes is solving the problem of cable changing parameters over cable age. The principle of the first proposed method is based on the successive identification of the fault generated high frequency travelling waves of the voltage signals arriving at both cable ends. The high frequency components are processed using Wavelet Transform to recognize the changes corresponding to the discontinuity points on the cable line. Correct fault position will be determined by analyzing the relationship between the characteristics of the transieni sequences detected. The second adaptive fault location scheme extracts the fundamental power frequency current and voltage measurements from both ends of the cable line using Discrete Fourier Transform, then estimates the fault location using the distributed line model. The method considers the line cable sections to be transposed and no assumptions were made for fault boundary conditions or fault resistance. In both proposed schemes, the processed relaying signals are the modal coordinates signals. Extensive simulation studies and results using ATP/EMTP are outlined in the thesis to demonstrate the validity of the two suggested techniques in locating faults with accepted accuracy. Both of the schemes respond very well irrespective to fault position, fault type, fault resistance, or fault inception angle. In .addition, the results prove that both of the two proposed schemes solve the problem of cables changing parameters. Both of the schemes are adopted to address the problem of locating faults in aged cables in commonly system configurations: radial aged cable system, multi­ end aged cable system, and aged cables combined with overhead line.