Harmonic Fingerprinting for Power Quality Source Identification Using Tunable-Q Wavelet Transform

The reliable operation of modern electrical grids heavily depends on the accurate identification of power quality disturbances caused by various sources such as non-linear loads, faults, and electric vehicle (EV) chargers. This study introduces an innovative approach for harmonic source identification using the Tunable-Q Wavelet Transform (TQWT), which delivers high-resolution time-frequency analysis to effectively capture and isolate harmonic components. By applying TQWT to voltage and current signals, we extract distinctive harmonic features that serve as unique signatures for different disturbance sources, including high-order harmonics from non-linear loads, low-frequency distortions from faults, and switching harmonics from EV chargers. To enhance classification accuracy, a dual-task Convolutional Neural Network (CNN) is employed, simultaneously addressing both power quality event classification and location identification. Our results demonstrate that the proposed TQWT-based method, in conjunction with the dual-task CNN, outperforms traditional Fourier-based techniques, achieving higher accuracy, faster convergence, and reduced loss. This approach offers a promising solution for real-time power quality monitoring, providing valuable insights into grid disturbances and enhancing overall system reliability and operational efficiency.