Optimisation of electric vehicles charging station

Abstract

As the number of electric vehicles grows, the need for charging stations and the relevance of the limited resources problem increases. Many charging stations are powered only by a limited amount of electricity. Therefore, not all arriving cars can charge to their maximum potential. The distribution of electricity power to consumers must therefore be optimised. To ensure a positive charging experience for the customer, it is important to consider their needs when designing the charging station: while some prioritise fast charging, others aim for the lowest cost possible. Our research focused on developing a methodology for optimising electric car charging, which divides users into three categories based on their priorities (time and money costs). Whenever there are several electric cars of the same category at the charging station, a linear programming problem is created in order to distribute electric power according to arrival time and battery capacity. Experimental studies were conducted to assess whether the arrival time of an electric car or its battery capacity is more important in a linear programming problem. In order to conduct the research, we used a set of generated data, which includes the technical characteristics of the vehicles, the user category selected, and the time of arrival of the vehicles. Kruskal-Wallis test and Dunn’s statistical test were applied in order to determine whether the difference in the obtained models results is statistically significant. It was found during the research that the best-performing model is based on equal weights assigned to two parameters: the arrival time of the electric vehicle and its battery capacity