Monte Carlo–Based EV Load Modelling and Optimal EVCS–DG Planning Using Pelican Optimization Algorithm

Abstract

The stochastic nature of electric vehicles poses significant challenges to the radial distribution systems. In this paper, the electric vehicle load is modelled using a Monte Carlo Simulation (MCS) framework by considering key parameters such as initial state of charge, final state of charge, and charging start time. The peak EV demand obtained from the MCS analysis is considered as the electric vehicle charging station (EVCS) load, and three optimal EVCS locations are determined using a new optimization algorithm called the pelican optimization algorithm (POA). A multi-objective optimization problem is formulated and tested on the IEEE-69 and IEEE-85 bus radial distribution systems. Coordinated placement of three EVCSs with three types of distributed generation (Type-I, Type-II, and Type-III) is carried out and benchmarked against Particle Swarm Optimization (PSO). Turning to the result highlights, the MATLAB analysis reveals notable outcomes. Specifically, in the IEEE-69 bus system, power losses reduced by 97.93%, AVDI decreased to 0.000002 p.u, and the VSI increased to 0.976979 p.u. In the IEEE-85 bus system, losses reduced by 92.49% and AVDI decreased to 0. 000049 p. u. Meanwhile, VSI reached 0.939066 p.u. Furthermore, POA exhibits improved convergence behaviour and overall performance, highlighting its effectiveness for EV-integrated distribution network planning.