This paper presses a smart charging decision-making criterion that significantly contributes in enhancing the scheduling of the electric vehicles (EVs) during the charging process.The proposed criterion aims to optimize the charging time, select the charging methodology either DC constant current constant voltage (DC-CCCV) or DC multi-stage constant currents (DC-MSCC), maximize the charging capacity as well as minimize the queuing delay per EV, especially during peak hours. The decision-making algorithms have been developed by utilizing metaheuristic algorithms including the Genetic Algorithm (GA) and Water Cycle Optimization Algorithm (WCOA). The utility of the proposed models has been investigated while considering the Mixed Integer Linear Programming (MILP) as a benchmark. Furthermore, the proposed models are seeded using the Monte Carlo simulation technique by estimating the EVs arriving density to the EVS across the day. WCOA has shown an overall reduction of 13% and 8.5% in the total charging time while referring to MILP and GA respectively.
This paper demonstrates an experimental attempt to prototype electric vehicle charging station’s (EVCS) decision-making unit, using artificial neural network (ANN) algorithm. The algorithm acts to minimize the queuing delay in the station, with respect to the vehicle state of charge (SOC), and the expected arrival time. A simplified circuit model has been used to prototype the proposed algorithm, to minimize the overall queuing delay. Herein, the worst-case scenario is considered by having number of electric vehicles arriving to the station at the same time greater than the charging points available in the station side. Accordingly, the optimization technique was applied to reduce the mean charging time of the vehicles and minimize queuing delay. Results showed that this model can help in reducing the queuing delay by around 20% of the mean charging time of the station, while referring to a bare model without ANN algorithm as a reference.
The current study proposes a smart decision-making algorithm to be utilized in electric vehicle stations. The suggested approach emphasizes the prediction of queuing delay seeking for minimum total charging time. For this purpose, artificial neural network (ANN) model is used, where a dataset is pre-generated to be seeded into the model. The proposed model effectiveness can be proven when the number of arriving vehicles at the station exceeds the maximum number of charging points at the station. The model accuracy was recorded to reach 89%. For validity, the proposed ANN model was evaluated with respect to a meta-heuristic optimizer, showing a reduced total charging time by 2.5%, and 23.9% with respect to a bare model with no optimization. As a final validation step, a physical realization of the ANN model was conducted by emulating a vehicle as a transmitting node and the station as a receiving node.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.