Optimal distributed generation allocation considering renewable and load uncertainties

Abstract

The integration of distributed generation (DG) into electrical networks offers technical and environmental benefits, including reduced power losses, improved voltage stability, and enhanced reliability. Improper DG placement or sizing, however, can degrade network performance. This study addresses optimal DG allocation under uncertainty, focusing on voltage profiles and stability indices in distribution networks. Variations in photovoltaic (PV) and wind turbine (WT) generation due to solar irradiance and wind speed are considered. Seven optimization algorithms—FVIM, SBO, SCSO, PSO, WOA, ALO, and Harmony Optimization—are applied and compared. Two DG types (active power only and active-reactive power) and a hybrid scenario with capacitor banks are analyzed. The IEEE-33 bus network is used as a test system. Results show reductions in total annual cost ($7.654 M → $2.614 M), voltage deviations (38.376 p.u. → 10.826 p.u.), and power losses (4043.462 kW → 2500.466 kW). Minimum voltage improved from 0.9065 to 0.9538 p.u. WOA achieves the lowest overall cost ($2,614,363), followed by PSO ($2,648,914) and FVIM ($2,699,308). FVIM demonstrates consistent technical performance, with total VDDT of 11.3144 p.u., VSIT of 730.6071, minimum bus voltage of 0.9527 p.u., and total active/reactive losses of 2789.703 kW and 2029.58 kVAr, maintaining a strong balance between economic and operational objectives.