Enhancing power quality of pv grid-connected system through mantis shrimp optimization algorithm for optimal Dc bus voltage control.

The nonlinear and intermittent nature of Photovoltaic (PV) systems introduces dynamic disturbances that negatively impact the stability of the DC bus voltage (Vdc) between PV sources and shunt active power filters (SAPFs). These fluctuations pose significant challenges to the performance of SAPFs, especially when the reference DC bus voltage (Vdc*) is constant and not adapted to the instantaneous operating conditions. In this study, a Perturb and Observe (P&O) algorithm is employed within the PV subsystem to perform Maximum Power Point Tracking (MPPT), further contributing to the time-varying behavior of Vdc. To address this problem, this paper proposes a real-time optimization strategy based on the Mantis Shrimp Optimization Algorithm (MShOA) for continuous Vdc* adjustment. This method relies on real-time Total Harmonic Distortion (THD) feedback to dynamically determine the optimal Vdc*, thereby improving harmonic mitigation and maintaining voltage stability. Simulation results demonstrate that the proposed MShOA-based approach effectively reduces THD from 3.59% to 2.85% obtained with conventional methods to 2.33% before PV injection, and maintains 4.19% after PV injection, remaining within the IEEE 519 - 92 standard limits. To confirm its superiority, a comparison with the Whale Optimization Algorithm (WOA) was performed, which achieved 2.65% before and 5.78% after PV injection. These findings validate the higher accuracy, faster convergence, and better adaptability of the proposed MShOA in ensuring robust voltage regulation and improved power quality under PV injection conditions.