•  
  •  
 

Keywords

electric vehicle; orderly charging; V2G; mountainous city; wind-solar forecasting; Bi-LSTM

Abstract

To address the current energy abandonment phenomenon of new energy sources such as wind power and photovoltaic power, an orderly charging and discharging strategy for electric vehicles (EVs) is proposed to promote wind-solar consumption. This strategy uses the vehicle-to-grid (V2G) interaction technology and aims to maximize the regional wind-solar consumption rate, minimize power load fluctuation, and maximize the power company’s electricity sales benefit in the context of mountainous cities by establishing a multi-objective charging model. The output of wind power and photovoltaic power is predicted using variational mode decomposition combined with a bidirectional long short-term memory (Bi-LSTM) network. Based on the predicted outputs, the output periods are divided, and dynamic electricity prices are set. The problem is solved using the adaptive particle swarm optimization algorithm, Yalmip + Cplex, and CVX toolbox. Case results show that when the user V2G responsiveness is 30%, 60%, and 100%, the wind-solar consumption rates are 83.73%, 89.12%, and 97.11%, respectively, power load fluctuations are decreased by 41.89%, 44.46%, and 47.32%, respectively, while ensuring the electricity sales benefits of the power company.

DOI

10.19781/j.issn.1673-9140.2025.03.017

First Page

154

Last Page

162

Share

COinS