| The safe and reliable operation of power system is an important guarantee for the development of national economy. High voltage transmission lines are the key to build the power system and their reliable operation is the premise of the stable operation for the power system. Due to the influence of external environment and other factors, sometimes, faults may happen on the high voltage transmission lines. Therefore, it is important to calculate fault distance in order to repair the fault and eliminate the hidden trouble in time.At present, the fault location methods are mainly classified into fault analytical method,traveling wave ranging method and intelligent distance ranging method. In this dissertation,the traveling wave ranging method is studied from the aspects of traveling wave head extraction method, the identification of the wave head polarity and the fault section. First of all, the introduction of fault transient traveling wave generation process, the phase mode transformation theory of three phase transmission lines, the phenomenon of refraction and reflection during the traveling wave transmission, and the basic principle of traveling wave fault location are expounded. Secondly, the relevant theory of wavelet transform and S transform is described. Based on the theory of S transform, an improved S transform method is obtained by the modified deduction on the Gauss window function. The ranging results based on the two signal extraction methods of wavelet transform and improved S transform are compared and analyzed by the error analysis factors, such as the average deviation, the standard deviation and the range. In the view of the problem that the polarity discrimination method could be invalid in the process of single-terminal fault location, the method of artificial neural network based on genetic algorithm optimization is used to make a preliminary judgment on the fault location. According to the preliminary fault distance result the ranging formula is selected. Then the accurate fault distance result is calculated with the arrival time of wave head. Aiming at the problem that the complexity of the algorithm is increased because of the distinguishment of wave head polarity in the process of single-terminal fault location, the method of two-terminal fault location is studied in this dissertation. Based on the theory of two-terminal fault location method, the transmission line model is extended. The fault branch discrimination method of the teed and multi-terminal transmission lines is deduced and analyzed. And a flexible threshold is introduced to modify the fault branch discrimination matrix.The simulation results show that the improved S transform method is more stable and accurate than the wavelet transform method in the calibration process of the wave head; the preliminary fault distance results obtained by intelligent algorithm can make up the the deficiency of polarity discrimination method in order to improve the reliability and accuracyof the single-terminal fault location; the matrix analysis method of fault branch discrimination can rapidly achieve fault section of the teed and multi-terminal transmission lines, so that the matrix analysis method can improve the efficiency of complex circuit fault location; the applicability of fault branch discrimination matrix is improved by the threshold correction method. |