Mini-grid sizing in tropical rural communities using particle swarm optimization

This study makes a significant contribution by quantitatively demonstrating the critical impact of incorporating hourly solar radiation, temperature, and electricity demand data into MG sizing, revealing substantial implications for system capacity and cost metrics in tropical rural communities. The findings show that neglecting hourly temperature variations leads to a considerable underestimation of the required PV and battery capacities, alongside a dramatic increase in the Levelized Cost of Electricity (LCOE) and Net Present Cost (NPC). Specifically, the analysis across three Mozambican communities indicates that accounting for temperature necessitates an increase in PV capacity ranging from 74% to 114% and battery storage from 85% to 122%. This translates directly into a significant increase in project costs, with LCOE increasing by 116% to 165% and NPC by 116% to 147%. These percentage increases are not minor adjustments but rather highlight a fundamental oversight in mini-grid sizing methodologies that do not account for temperature effects.