This paper proposed an improved vehicle-to-grid (V2G) scheduling approach for the frequency control with the advantage of protecting the batteries hence saving the battery lifetime during grid connected service. The proposed methodology is improved in two ways. Firstly, to give a prediction of the available electric vehicle (EV) battery capacity in the control time-step for the V2G service, a deep learning based prediction is developed. Secondly, this study advances the previous V2G method by adding the quantitative analysis of the battery cycle life into the V2G optimization. The accurate prediction of the schedulable battery capacity based on the LSTM algorithm is shown very effective in the power system frequency control. Also, compared with the previous method that without battery lifetime control, the proposed method benefits in the reduction of charge/discharge cycles.
Bibliographical noteFunding Information:
This work was supported by the National Nature Science Foundation of China with Grant Number 51807008 and Grant Number U1864202 .
© 2020 Elsevier Ltd
Copyright 2020 Elsevier B.V., All rights reserved.
FunderNational Nature Science Foundation of China with Grant Number 51807008 and Grant Number U1864202
- Deep learning
- Electric vehicles
- Frequency control
- Vehicle to the grid
ASJC Scopus subject areas
- Civil and Structural Engineering
- Building and Construction
- Mechanical Engineering
- Industrial and Manufacturing Engineering
- Electrical and Electronic Engineering