Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning
Abstract Smart transportation is an important application scenario in the field of urban computing. As the popularity of electric vehicles increases, the demand for fast charging is growing rapidly. In response to this, battery swapping stations are being proposed as a solution, but their operationa...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-02-01
|
| Series: | Computational Urban Science |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s43762-025-00166-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823862853516394496 |
|---|---|
| author | Hongwei Yi Huan Zhang Jianghong Li Yanling Zhao |
| author_facet | Hongwei Yi Huan Zhang Jianghong Li Yanling Zhao |
| author_sort | Hongwei Yi |
| collection | DOAJ |
| description | Abstract Smart transportation is an important application scenario in the field of urban computing. As the popularity of electric vehicles increases, the demand for fast charging is growing rapidly. In response to this, battery swapping stations are being proposed as a solution, but their operational efficiency is challenged by factors such as battery life, vehicle queues, and grid load management. In this paper, a mixed intelligent optimization strategy combining the proximal policy optimization (PPO) algorithm from reinforcement learning and the goat swarm optimization (GSO) algorithm is proposed. The GSO-PPO algorithm is constructed, where PPO algorithm learns the optimal scheduling strategy for the battery swapping station in a dynamic environment, and the GSO algorithm optimizes the hyperparameters of PPO and adjusts the weight of the reward function to achieve the multi-objective optimization of minimizing battery life, shortening vehicle waiting time, and efficiently managing grid load. The experimental results show that compared with random strategies and traditional PPO algorithms, GSO-PPO reduces vehicle waiting time and improves service efficiency, making the overall operation of the battery swapping station more stable. The study demonstrates the potential of combining reinforcement learning and swarm intelligence algorithms in smart energy infrastructure and solving multi-objective optimization problems. |
| format | Article |
| id | doaj-art-e879199d109540e29c1f7b3914185661 |
| institution | Kabale University |
| issn | 2730-6852 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Springer |
| record_format | Article |
| series | Computational Urban Science |
| spelling | doaj-art-e879199d109540e29c1f7b39141856612025-02-09T12:25:00ZengSpringerComputational Urban Science2730-68522025-02-015111310.1007/s43762-025-00166-0Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learningHongwei Yi0Huan Zhang1Jianghong Li2Yanling Zhao3Smart City College, Beijing Union UniversitySmart City College, Beijing Union UniversitySmart City College, Beijing Union UniversitySmart City College, Beijing Union UniversityAbstract Smart transportation is an important application scenario in the field of urban computing. As the popularity of electric vehicles increases, the demand for fast charging is growing rapidly. In response to this, battery swapping stations are being proposed as a solution, but their operational efficiency is challenged by factors such as battery life, vehicle queues, and grid load management. In this paper, a mixed intelligent optimization strategy combining the proximal policy optimization (PPO) algorithm from reinforcement learning and the goat swarm optimization (GSO) algorithm is proposed. The GSO-PPO algorithm is constructed, where PPO algorithm learns the optimal scheduling strategy for the battery swapping station in a dynamic environment, and the GSO algorithm optimizes the hyperparameters of PPO and adjusts the weight of the reward function to achieve the multi-objective optimization of minimizing battery life, shortening vehicle waiting time, and efficiently managing grid load. The experimental results show that compared with random strategies and traditional PPO algorithms, GSO-PPO reduces vehicle waiting time and improves service efficiency, making the overall operation of the battery swapping station more stable. The study demonstrates the potential of combining reinforcement learning and swarm intelligence algorithms in smart energy infrastructure and solving multi-objective optimization problems.https://doi.org/10.1007/s43762-025-00166-0Urban computingElectric vehicleBattery swapping stationReinforcement learningHybrid optimizationTarget balance |
| spellingShingle | Hongwei Yi Huan Zhang Jianghong Li Yanling Zhao Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning Computational Urban Science Urban computing Electric vehicle Battery swapping station Reinforcement learning Hybrid optimization Target balance |
| title | Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning |
| title_full | Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning |
| title_fullStr | Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning |
| title_full_unstemmed | Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning |
| title_short | Hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning |
| title_sort | hybrid intelligent optimization strategy of battery swapping station for electric vehicles based on reinforcement learning |
| topic | Urban computing Electric vehicle Battery swapping station Reinforcement learning Hybrid optimization Target balance |
| url | https://doi.org/10.1007/s43762-025-00166-0 |
| work_keys_str_mv | AT hongweiyi hybridintelligentoptimizationstrategyofbatteryswappingstationforelectricvehiclesbasedonreinforcementlearning AT huanzhang hybridintelligentoptimizationstrategyofbatteryswappingstationforelectricvehiclesbasedonreinforcementlearning AT jianghongli hybridintelligentoptimizationstrategyofbatteryswappingstationforelectricvehiclesbasedonreinforcementlearning AT yanlingzhao hybridintelligentoptimizationstrategyofbatteryswappingstationforelectricvehiclesbasedonreinforcementlearning |