Enhanced Grey Wolf Optimisation for Sustainable PV-DG Allocation Under Seasonal Uncertainty

Improving energy efficiency and renewable-energy hosting capacity in distribution networks is essential for sustainable power-system development. Conventional networks face power losses, voltage deviations, thermal stress, and limited flexibility to accommodate increasing renewable penetration. Accordingly, optimal placement and sizing of photovoltaic-based distributed generation (PV-DG) has become an effective approach for enhancing network performance. This study applies the Grey Wolf Optimiser (GWO) to PV-DG allocation considering three objectives: minimisation of active power loss and total voltage deviation, and maximisation of the voltage stability index. To mitigate the premature convergence of standard GWO, Opposition-Based Learning (OBL) is integrated to enhance exploration through opposite candidate solutions. The proposed OBL-GWO framework is evaluated on the IEEE 33-bus system under three PV-DG penetration scenarios using five-year irradiance and load data from Riyadh, including seasonal uncertainty and long-term load growth. Compared with standard GWO, it achieves average additional gains of 11.24, 10.0, 6.79, and 2.74 percentage points in power-loss reduction, TVD reduction, VSI improvement, and minimum-bus-voltage improvement, respectively. Sensitivity analysis confirms limited dependence on moderate changes in objective weights. A persistent cloudy-week case demonstrates robust performance against GWO, PSO, and GA under adverse irradiance. Scalability tests on IEEE 69- and 118-bus systems confirm that OBL-GWO retains its solution-quality and convergence advantages on larger networks.