|
چکیده
|
Recently, the development and progress in the technology and manufacturing of traveling wave relays have caused algorithms based on this method to receive more attention than traditional relay algorithms such as phasor and impedance-based methods. So far, various researches in the field of comprehensive protection including all three stages of fault detection, classification, and location (FDCL) in high-voltage transmission lines based on traveling wave theory have been widely published. Most of the published references in this field have worked on each of the stages of comprehensive protection separately, and few references have designed all three stages in the form of an algorithm. Solving them simultaneously in the form of a flowchart has been a major challenge for various reasons in terms of execution time and accuracy of results. On the other hand, references that have solved all three stages in the form of an algorithm have encountered various challenges, including the need for model training data, lack of appropriate accuracy in critical conditions, and high sensitivity to specific conditions. Therefore, in this paper, a comprehensive method for FDCL in high-voltage transmission lines is presented based on three-phase voltage signals of both line terminals of the line and also using the unique features of time–time (TT) transform and discrete wavelet transform (DWT) in the form of a flowchart. The proposed method uses the output results of the TT transform to detect the arrival time of the traveling wave caused by the fault to both terminals on both sides of the line, as well as the accurate and detailed coefficients of the fourth level of wavelet transform and predetermined threshold values, to implement the proposed algorithm. The method is implemented in the Simulink/MATLAB software and is programmed in an m-file of this software. The results obtained from 960 simulated scenarios and eight critical sensitivity analyses demonstrated that the proposed algorithm achieves a fault detection delay of only 0.188 ms, 100% accuracy in faulty phase identification, and an average fault location error of 0.66%. The comparative results with previously reported methods in the literature indicate that the proposed approach provides superior robustness, speed, and accuracy under various operating and noise conditions.
|