Electricity-driven industries require high and regulated power quality, but increasing high-power nonlinear loads pose challenges to the distribution system. To address these issues, this research aims to improve power quality (PQ) in distribution networks by developing a novel hybrid control method for multi-level inverter-driven shunt active power filters (SAPF). The major goals are to minimize total harmonic distortion (THD) of the supply current and reduce inverter-related power losses. To achieve this, the proposed technique combines the Black-Winged Kite Algorithm (BWKA) with a Quantum Self-Attention Neural Network (QSANN) to intelligently control the inverter switching angles. The BWKA is employed to optimize switching angles globally, while the QSANN predicts and fine-tunes these angles adaptively for dynamic load conditions. The performance is evaluated through MATLAB simulations on a 415 V, 50 Hz system under both balanced and unbalanced nonlinear load conditions. Simulation outcomes establish that the proposed technique minimizes THD of the supply current to 1.4% and minimizes power losses to 0.3 W, outperforming conventional techniques such as CNN, MNSGA-II, and BOA in terms of harmonic suppression, inverter efficiency, and dynamic response. These results highlight the potential of BWKA-QSANN for adaptive and effective power quality enhancement in simulated distribution system scenarios.
Raja et al. (Fri,) studied this question.