Research On Fault Location Of High Voltage Transmission Line

With the development of smart grid project, UHV AC transmission network has been developed vigorously. Meanwhile, the reactive power compensation device installed in the power grid, high voltage direct current transmission (HVDC), flexible AC transmission(FACTS), and a variety of clean energy grid connected, fault location becomes more complex,Existing fault location methods will also be affected. The transmission line is the lifeline of the power system, at the same time, it is the most frequently occurring part of power system fault,research on fault location algorithm for transmission line quickly and accurately, it is of great significance to the stable and efficient operation of the power grid. Under this background, in this paper two kinds of transmission lines (high voltage transmission lines and series compensated transmission lines) are studied, the corresponding fault location algorithms are proposed for different lines. Specific work as follows:First, transient traveling wave occurs when high voltage transmission line is short circuited,transient traveling wave contains abundant fault information. According to the characteristics of transient traveling wave and the precise extraction of traveling wave head, the TT transform is applied to the traveling wave head extraction. On this foundation, the TT transform is applied to the T-type transmission lines, a fault branch discrimination method is derived, and the fault branches are identified and located according to the fault judgment method.Second, Series compensated transmission lines is widely used, it can improve the transmission line distance, improve the stability of the system. Because of series capacitor protection MOV is nonlinear, the original fault location algorithm is no longer suitable for series compensated lines. This paper presents a fault location method for series compensated transmission lines based on wavelet neural network. This method first extract the zero sequence current of the fault line, then the discrete wavelet transform is used to extract the fault feature vector, then the fault feature vector and corresponding fault distance target value are the input of BP neural network for training, the training result is the fault distance.Simulation results verify: in the fault location of double ended and T-type transmission lines, the TT transform can accurately capture the traveling wave head and realize the fault location; in fault location of series compensated transmission lines, wavelet neural network is not affected by the main protection MOV, sample test certificate that the method can accurately locate the fault. In addition, these two algorithms are not affected by the line length, fault type and transition resistance, and have a high positioning accuracy.