A Comparative Study of Metaheuristics Algorithms Applied for Optimal Reactive Power Dispatch Problem Considering Load Uncertainty

The power grid is facing increased power losses and voltage instability due to the rising electricity demand. In addition, the unpredictability of energy demand caused by consumer behavior and seasonal changes creates difficulties in accurately forecasting and planning for the power system. This variability in load requirements can negatively impact decision-making in the power industry. However, incorporating this variability in demand can improve planning by allowing the power system to adapt to changing electrical demands. This chapter focuses on solving the optimal reactive power dispatch (ORPD) while considering the load uncertainty. The main objective of ORPD is to achieve minimal power losses by adjusting system control variables that involve both continuous and discrete control variables while satisfying the system's equality and inequality constraints. The Monte Carlo simulation (MCS) and scenario-based reduction (SBR) approaches are utilized for load uncertainty representation. The experimentation is carried out on IEEE-30 bus systems, and the results are evaluated in a comparative study between four metaheuristic optimization methods, namely black widow optimization (BWO) gray wolf optimization (GWO), particle swarm optimization (PSO), and harmony search (HS) algorithms. Through computational analysis, it is demonstrated that the GWO algorithm outperforms other reported algorithms.