Dual-Objective Accelerated Linear Convergence Spotted Hyena Optimization for Power Enhancement in Partially Shaded PV Arrays

Abstract

Partial shading (PS) presents a significant concern in PV arrays due to its substantial impact on system performance, causing reduced power output, distorted PV characteristics, and power loss. Therefore, this paper introduces a novel accelerated linear convergence factor-based spotted hyena optimization for dynamic PV array reconfiguration (DPVAR). The key aim of the accelerated linear convergence factor is to reduce row difference in each tier while minimizing the need to relocate modules, within a significantly reduced iteration count. The algorithm pursues two primary objectives: enhancing power generation under PS and efficiently finding the global maximum power within a minimum period while minimizing steady-state oscillations. Furthermore, the proposed algorithm is adaptable to dynamic PS conditions and applicable for symmetric and asymmetric PV arrays of any size. The effectiveness of the technique is evaluated through simulation and experiments. In addition, the performance of the proposed technique is compared to the particle swarm optimization (PSO) based reconfiguration method. The investigation reveals that the proposed reconfiguration technique demonstrates an average power enhancement of 17.57% compared to the before-reconfiguration method and 7.88% compared to the PSO reconfiguration method.