An Energy Management Strategy of Hybrid Electric Vehicles based on Deep Reinforcement Learning | |
---|---|
( Volume 5 Issue 12,December 2019 ) OPEN ACCESS | |
Author(s): | |
Chengzhao Yang, Changcheng Zhou | |
Abstract: | |
An energy management strategy (EMS) plays important roles on the performance of hybrid electric vehicles (HEVs). However, the EMS based on rules is difficult to achieve an optimized result, whereas the EMS based on optimization theories cannot achieve the adaptability to different driving cycles. An energy management strategy of HEVs based on the deep reinforcement learning (DRL) is proposed in this research, which is fully data-driven and learning-enabled and does not rely on any experience of experts and accurate mathematical models. The proposed strategy is verified under a co-simulation environment, and the result shows that the proposed strategy achieves a 3.51% fuel saving compared to a rule-based strategy. |
|
Paper Statistics: | Total View : 690 | Downloads : 681 | Page No: 1-4 | |
Cite this Article: | Click here to get all Styles of Citation using DOI of the article. |