نوع مقاله : مقاله پژوهشی
نویسنده
دانشکده مهندسی برق، واحد شبستر، دانشگاه آزاد اسلامی، شبستر، ایران
چکیده
کلیدواژهها
موضوعات
عنوان مقاله [English]
نویسنده [English]
In this paper, a novel method based on the Adaptive Short-Time Fourier Transform (ASTFT) is proposed for reliable detection of faults under power swing conditions. The core of this approach lies in analyzing the frequency content of current signals and calculating their spectral energy. The innovation of the proposed algorithm is its dynamic adaptation of the time-frequency analysis window width: under normal and power swing conditions, a wider window is used to achieve high frequency resolution, while at the moment of fault occurrence, a narrower window is employed to enhance time resolution and enable faster detection. This adaptive mechanism is governed by estimating the instantaneous frequency of the signal and calculating its local stationarity length. To evaluate the proposed method, simulations were conducted in PSCAD/EMTDC and MATLAB environments on a standard power system network. Results demonstrate that ASTFT can detect symmetrical and asymmetric faults during power swings regardless of swing frequency (ranging from 0.5 to 5 Hz), fault location, fault resistance, or fault initiation time. Notably, the detection time of ASTFT is on average 15% faster than the classical STFT method (less than 20 milliseconds), and it maintains robust performance even in the presence of disturbances. Due to its high speed, accuracy, and insensitivity to varying network conditions, ASTFT offers a practical and effective solution for enhancing the reliability of distance relays in modern power systems.
کلیدواژهها [English]
)