An optimal solution for a smart charging station of light electric vehicles

This paper presents an optimal management tool for a smart charging station (SCS) of Light Electric Vehicles (LEV), i.e., bicycles and scooters, integrated in a public lighting system. Transitioning this line to LEDs reduced capacity on the line so there is sufficient excess capacity on the system to handle the addition of the LEVs. The features of the SCS include both smart and vehicle-to-grid (V2G) nodes. The tool implemented in Python consists of two main components: (1) a forecasting module which predicts both the time of arrival in the charging station, and the charging nodes state profile, empty or occupied; and (2) an optimization module to determine the optimal power profile of each smart charging node at the SCS. The optimization problem is a mixed integer linear programming model (MILP) that uses time-of-use electric tariff rates, expected energy demand of the street where the SCS is installed, and predicted profiles from the forecasting module to minimize the cost function over a time horizon of 24-hours. This paper includes a case study based on public data from Barcelona’s bicycle sharing system, public lighting demand of a street, and energy prices from the Spanish energy market. The results show how is it possible to integrate this type of SCS within public lighting systems without the need to expand the electrical network thanks to the use of optimal management strategies.