Short-Term Power Quality Disturbance Detection Method Based on S-Transform with Source Code

Voltage sag represents one of the most critical dynamic power quality concerns. With the extensive application of various new power electronic devices, power quality requirements for grid systems have become increasingly stringent. Meanwhile, voltage sags exhibit propagation characteristics within power networks, making research and management of voltage sags urgently needed from both utility and consumer perspectives.

This research paper initiates the study of voltage sags through detection methodologies. The paper first analyzes the current research landscape of short-term power quality disturbance detection globally, with particular focus on voltage sag characteristics under three disturbance scenarios: short-circuit faults, induction motor starting, and transformer excitation. Based on this analysis, the study adopts an S-transform-based short-term power quality disturbance detection approach. The implementation algorithm extracts required data from both S-magnitude matrices and S-complex matrices obtained through S-transform processing, enabling comprehensive analysis of amplitude variations, phase jumps, duration characteristics, and harmonic content under various disturbance conditions. The core algorithm involves time-frequency analysis through windowed Fourier transforms with frequency-dependent resolution.

To better demonstrate research outcomes, simulation experiments were conducted using MATLAB implementations. Results indicate that the S-transform-based detection method effectively identifies key voltage sag characteristics including amplitude variations, phase jumps, duration parameters, and harmonic components. The method employs feature extraction algorithms that calculate statistical parameters from time-frequency matrices, making it suitable for real-time monitoring applications. Therefore, this approach provides substantial support for maintaining and improving power quality in electrical grid systems.