Loss of load probability minimization for stand-alone photovoltaic system using elephant herding optimization

Abstract

Stand-alone photovoltaic power systems are commonly deployed for rural electrification. Notwithstanding this, non-optimal sizing has been a constant issue in implementing such systems as non-optimal system sizing reduces the system’s reliability in supplying the required power to the loads. Hence, an Elephant Herding Optimization (EHO) is presented for optimal system sizing by searching for the optimal photovoltaic module, solar battery, charge controller and inverter so that the loss of load probability is minimized. Before developing EHO for sizing, an iterative-based approach was formulated to size the system by testing all possible design options. The optimal sizing results obtained from this iterative approach was then used for the validation of optimal sizing solution obtained from EHO-based sizing algorithm. EHO was found to produce same loss of load probability as presented by the iterative approach but with approximately 2.86 times lower in computation time. In addition, it was discovered to be better than evolutionary programming as it had lower loss of load probability with slightly faster computation.