Optimal schedule for electric vehicle charging at parking lots and battery swapping stations

Abstract

Electric vehicles (EVs) are now being widely adopted, not only to reduce the amount of fossil fuels used, but also to reduce greenhouse gas emissions and to protect the environment. However, many car owners are still willing to buy traditional vehicles due to certain well-known drawbacks of EVs, such as long charging time, short battery life, limited travel distance per charge and expensive EV batteries. In order to solve the EV charging problem in urban cities where the drivers may not have private charging poles at home, two approaches are examined in this thesis: charging at a parking lot (PL), or swapping battery at a battery swapping station (BSS). Currently, most of the PLs manage the EV charging requests on a first-in-first-serve strategy, which causes long queuing time and low charging efficiency. With the rapid growth of EV populations, it is essential to employ a resource allocation system to schedule the EV charging and satisfy the electric requirements for power infrastructures. A PL dynamic resource allocation system for recharging EVs is introduced in this thesis. For scheduling purposes, the scheduling period is divided into sequential timeslots. At the beginning of each timeslot, the dynamic system can determine an optimal charging schedule for that timeslot, as well as plan for subsequent timeslots. Considering the variation in electricity prices during the day, the objective is to minimize the cost of electricity used to charge EVs by scheduling optimal electric quantities at the parking timeslots of each EV. The optimal solution satisfies the EV’s charging rate limit and the PL’s transformer limit. However, considering the limitation of PL charging model, such as long charging time and limited travel distance per charge, two BSS models are designed as an alternative approach for providing energy to EVs. In both BSS models, the charging schedule is to determine an optimal charging method for the swapped batteries, and the charging methods have different charging rates and charging damages. In the first BSS model, the aim is to maximize an integrated objective function, which maximizes the BSS’s battery stock level and minimizes the average charging damage. In the second model, the aim is to minimize the BSS’s operation cost, which includes the number of batteries taken from stock, potential charging damage, and electricity cost. A mathematical model is also formulated for the charging process following the constant-current/constant-voltage charging strategy. The above scheduling problems are formulated as different mathematical models, and then some optimization algorithms such as the particle swarm optimization (PSO) algorithm, genetic algorithm (GA), differential evolution (DE) algorithm are implemented to solve the proposed models. After studying the properties of each algorithm, some new algorithms are designed to improve the performance of the typical algorithms for each model. All the proposed algorithms are implemented to solve the scheduling problems and the performances are compared with the typical algorithms and other scheduling strategies.