چكيده به لاتين
In this thesis, charging schedule of electric vehicles (EVs) at battery swapping stations (BSSs) is done which seeks to optimize the charging cost and cost of energy loss considering power system operational security. To this end, two kinds of variables, named as battery swapping location and battery charging priority, are employed for all of the optimization problems. The approaches are presented in the form of three scenarios. For the first one, charging schedule of relatively large number of EVs through a transmission network is performed as a preliminary template. For the second scenario, instead, charging schedule of relatively much less number of EVs through a region of a radial distribution network is accomplished. The problem is designed as a single objective which the goal of charging process is to minimize the cost of charging and energy loss. Considered constraints of the problem include maximum voltage deviation value of network buses and maximum permitted values for apparent power flowing through network branches. The last constraint of the problem is maximum permitted power consumption of the BSSs. For the first two scenarios, a modified combination of genetic algorithm (GA) and particle swarm optimization (PSO) is used as solution method. In the third scenario, a modified and completed form of EVs charging management is designed. Scheduling process is modeled as a multi-objective problem which the objective functions consist of the cost pertaining to charging and energy loss, voltage deviation of buses and apparent power flowing through branches. Besides that, the two latter objective functions are also modeled and considered as the power system operational security constraints. Two other constraints of this scenario are maximum number of battery swapping for each of the BSSs and charging all of the swapped batteries. Charge timing process of this scenario is designed in which no interruptions will occur for each of the batteries while charging. Furthermore, a dynamic pricing is depicted in order to applicably model the cost charging of the BSSs and also hinder high magnitude alternations in power consumption of the BSSs which makes the load profile smoother. Solution method of this scenario is non-dominated sorting genetic algorithm type II (NSGA-II) accompanied by analytical hierarchy process (AHP) in order to choose the best answer among the whole choices which approximate the Pareto front of the problem.
Keywords: Electric vehicles; battery swapping stations; scheduling by dynamic pricing strategy and non-interruptible charging method; non-dominated sorting genetic algorithm type II; modified combination of genetic algorithm and particle swarm optimization
