For the purpose of minimizing passive and receptive losses and giving the lowest possible voltage deviation, the IEEE standard 33-bus network is utilized in the current research to evaluate the capacity and position of EV charging stations. Therefore, according to their location, the current passing through the lines, the voltage of the phases, and, of course, the location of the loads, the voltage of the buses changes and can change the network expenses and the network voltage deviation profile. A charge is applied to the network since electric car owners show their likelihood of recharging based on a number of behavioral criteria, including the average charging time, the distance to the charging location, whether the charging location is busy or quiet, and other considerations. Therefore, this probabilistic behavior is dynamically modeled by queuing theory (modeled using Poisson's function), The uncertainty of other loads except for charging stations is modeled using the Monte Carlo distribution function, and finally, with the help of two cumulative algorithms, PSO and genetics, the ideal station placement is addressed, and the simulation's outcomes show appropriate operation. In terms of optimal positioning, particle algorithms outperform genetic algorithms.
