AI-Driven Adaptive Load Balancing Strategy for Electric Vehicle Charging Pile Networks
Main Article Content
Abstract
With the widespread use of EVs, new strains have been added to urban power systems. In-coordinated connections to charging station networks result in uneven load distribution, the separation of peak and valley demand, and less overall operational efficiency. In this paper, an AI-based adaptive load balancing approach to EV charging station networks is proposed. A three-layer distributed architecture is built, including perception and data collection layer, edge computing layer and decision-making layer in the cloud. For short-term load forecasting, a model combining long short-term memory networks with an attention mechanism is designed. A multi-agent proximal policy optimization algorithm is then used to realize the adaptive load balancing decisions. A simulated network of 20 charging stations and 200 charging piles was used for the experiments in an urban environment. The outcome is evident. The proposed strategy sets the Peak-to-valley load reduction rate at 63.2% across the network, and it reduces the load standard deviation to 31.4kW. The average wait time for users reduces from 18.7 minutes to 6.3 minutes. Violation rate of node voltage drops from 8.64% to 0.47%. Decision latency is still within 1.24 seconds even if a massive 50 stations are used in scenarios. The approach will ensure the safety and stability of the grid and optimize the use of the resources within the charging network without compromising the system’s integrity. It provides a theory and reference guide to sophisticated management of smart charging.
Downloads
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
References
K. Park and I. Moon, “Multi-agent deep reinforcement learning approach for EV charging scheduling in a smart grid,” Appl. Energy, vol. 328, p. 120111, 2022, doi: 10.1016/j.ap energy.2022.120111. [Online].
Z. Wan, H. Li, H. He, et al., “Model-free real-time EV charging scheduling based on deep reinforcement learning,” IEEE Trans. Smart Grid, vol. 10, no. 5, pp. 5246–5257, Sept. 2019, doi: 10.1109/TSG.2018.2879572. [Online].
L. Yan, X. Chen, J. Zhou, et al., “Deep reinforcement learning for continuous electric vehicles charging control with dynamic user behaviors,” IEEE Trans. Smart Grid, vol. 12, no. 6, pp. 5124–5134, Nov. 2021, doi: 10.1109/TSG.2021.3098298. [Online].
H. Li, Z. Wan, and H. He, “Constrained EV charging scheduling based on safe deep reinforcement learning,” IEEE Trans. Smart Grid, vol. 11, no. 3, pp. 2427–2439, May 2020, doi: 10.1109/TSG.2019.2955437. [Online].
X. Sun, W. Wang, S. Su, et al., “Coordinated charging strategy for electric vehicles based on time-of-use price,” Autom. Electr. Power Syst., vol. 37, no. 1, pp. 191–195, Jan. 2013, doi: 10.7500/AEPS201209274. [Online].
Z. Xu, Z. Hu, Y. Song, et al., “Coordinated charging strategy for PEV charging stations based on dynamic time-of-use tariffs,” Proc. CSEE, vol. 34, no. 22, pp. 3638–3646, Nov. 2014, doi: 10.13334/j.0258-8013.pcsee.2014.22.008. [Online].
Y. Zhang, Q. Yang, D. An, et al., “Multistep multiagent reinforcement learning for optimal energy schedule strategy of charging stations in smart grid,” IEEE Trans. Cybern., vol. 53, no. 7, pp. 4292–4305, July 2023, doi: 10.1109/TCYB.2022.3165074. [Online].
T. Qian, C. Shao, X. Li, et al., “Multi-agent deep reinforcement learning method for EV charging station game,” IEEE Trans. Power Syst., vol. 37, no. 3, pp. 1682–1694, May 2022, doi: 10.1109/TPWRS.2021.3111014. [Online].
S. Wang, S. Bi, and Y. A. Zhang, “Reinforcement learning for real-time pricing and scheduling control in EV charging stations,” IEEE Trans. Ind. Informat., vol. 17, no. 2, pp. 849–859, Feb. 2021, doi: 10.1109/TII.2019.2950809. [Online].
D. Qiu, Y. Wang, T. Zhang, et al., “Hybrid multiagent reinforcement learning for electric vehicle resilience control towards a low-carbon transition,” IEEE Trans. Ind. Informat., vol. 18, no. 11, pp. 8258–8269, Nov. 2022, doi: 10.1109/TII.2022.3166215. [Online].
D. Hu, Z. Ye, Y. Gao, et al., “Multi-agent deep reinforcement learning for voltage control with coordinated active and reactive power optimization,” IEEE Trans. Smart Grid, vol. 13, no. 6, pp. 4873–4886, Nov. 2022, doi: 10.1109/TSG.2022.3185975. [Online].
D. Liu, P. Zeng, S. Cui, et al., “Deep reinforcement learning for charging scheduling of electric vehicles considering distribution network voltage stability,” Sensors, vol. 23, no. 3, p. 1618, 2023, doi: 10.3390/s23031618. [Online].
C. Zhang, J. Liu, Y. Xiang, et al., “Electric vehicle charging facilities planning and its two-stage optimal charging based on data mining,” Proc. CSEE, vol. 38, no. 4, pp. 1054– 1064, Apr. 2018, doi: 10.13334/j.0258-8013.pcsee.170472. [Online].
L. Zhang, C. Sun, G. Cai, et al., “Two-stage optimization strategy for coordinated charging and discharging of EVs based on PSO algorithm,” Proc. CSEE, vol. 42, no. 5, pp. 1837–1851, May 2022, doi: 10.13334/j.0258-8013.pcsee.211150. [Online].
Z. Ye, Y. Gao, and N. Yu, “Learning to operate an electric vehicle charging station considering vehiclegrid integration,” IEEE Trans. Smart Grid, vol. 13, no. 4, pp. 3038– 3048, July 2022, doi: 10.1109/TSG.2022.3165479. [Online].
M. Alqahtani and M. Hu, “Dynamic energy scheduling and routing of a large fleet of electric vehicles using multi-agent reinforcement learning,” Comput. Ind. Eng., vol. 169, p. 108180, 2022, doi: 10.1016/j.cie.2022.108180. [Online].
J. Shi, W. Zhang, Y. Bao, et al., “Load forecasting of electric vehicle charging stations: attention based spatiotemporal multi-graph convolutional networks,” IEEE Trans. Smart Grid, vol. 15, no. 3, pp. 3016–3027, May 2024, doi: 10.1109/TSG.2023.3321116. [Online].
P. Aduama, Z. Zhang, and A. Al-Sumaiti, “Multifeature data fusion-based load forecasting of electric vehicle charging stations using a deep learning model,” Energies, vol. 16, no. 3, p. 1309, 2023, doi: 10.3390/en16031309. [Online].
W. Chen, H. Sun, R. Xu, et al., “Cloud-edge collaboration based hierarchical dispatch technology for EV participating in frequency regulation service,” Proc. CSEE, vol. 43, no. 3, pp. 914–926, Mar. 2023, doi: 10.13334/j.0258-8013.pcsee.220646. [Online].
Z. J. Lee, G. Lee, T. Lee, et al., “Adaptive charging networks: a framework for smart electric vehicle charging,” IEEE Trans. Smart Grid, vol. 12, no. 5, pp. 4339–4350, Sept. 2021, doi: 10.1109/TSG.2021.3074437. [Online].
D. Qiu, Y. Wang, W. Hua, et al., “Reinforcement learning for electric vehicle applications in power systems: a critical review,” Renew. Sustain. Energy Rev., vol. 173, p. 113052, 2023, doi: 10.1016/j.rser.2022.113052. [Online].
C. Bi, Y. Tang, Y. Luo, et al., “Review on critical problems in reinforcement learning methods applied in power system optimization and control scenarios,” Proc. CSEE, vol. 44, no. 1, pp. 1–21, Jan. 2024, doi: 10.13334/j.0258-8013.pcsee.223433. [Online].