Optimizing the utilization of new energy vehicle charging stations via integrated demand response and load balancing.
Saved in:
| Title: | Optimizing the utilization of new energy vehicle charging stations via integrated demand response and load balancing. |
|---|---|
| Authors: | Guo, Yanli1, Xue, Yun1 yun.xue@buu.edu.cn, Kong, Qingxue1, Gao, Shuli1 |
| Source: | Advances in Transportation Studies. Nov2026, Vol. 70, p275-298. 24p. |
| Subjects: | Electric vehicle charging stations, Time-based pricing, Demand forecasting, Spatiotemporal processes, Electric vehicles, Mathematical optimization, Energy demand management, Resource allocation |
| Geographic Terms: | Shenzhen (Guangdong Sheng, China : East), China |
| Abstract: | In response to issues such as uneven utilization efficiency of charging infrastructure, spatial-temporal mismatches between supply and demand, and peak congestion under the rapid development of new energy vehicles, this paper takes Shenzhen, China, as the research object. Based on the UrbanEV multi-source open dataset and integrating charging operation data, spatial data, weather data, and POI data, an hourly-level charging station occupancy analysis dataset is constructed, and a systematic study on spatiotemporal feature analysis, demand forecasting, and optimization strategy evaluation is conducted. First, key spatiotemporal features of charging station occupancy are extracted through data cleaning, multisource information integration, and feature engineering; second, three types of prediction models - multiple linear regression, ARIMA, and TimesNet - are constructed, and model performance is uniformly evaluated using RMSE, MAPE, and RAE; third, sensitivity analysis is conducted using the control variable method to identify the main driving factors affecting occupancy; finally, optimization strategies are designed from three dimensions: dynamic pricing, station layout optimization, and peak-shifting charging guidance, and their implementation effects are evaluated through scenario simulation. Results show significant spatiotemporal heterogeneity in Shenzhen's charging station occupancy. Electricity price has the strongest negative impact, catering POI density is a key spatial driver, while weather effects are weak. Multiple linear regression outperforms ARIMA and TimesNet in accuracy and interpretability. Off-peak guidance yields the highest improvement (17.05%), followed by layout optimization (16.89%), while dynamic pricing is least effective. The research results can provide a theoretical basis and decision support for the planning, operation management, and time-based pricing optimization of urban charging infrastructure, and are of reference value for improving charging service efficiency and promoting the coordinated development of new energy vehicles and urban energy systems. [ABSTRACT FROM AUTHOR] |
| Copyright of Advances in Transportation Studies is the property of Advances in Transportation Studies and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| Abstract: | In response to issues such as uneven utilization efficiency of charging infrastructure, spatial-temporal mismatches between supply and demand, and peak congestion under the rapid development of new energy vehicles, this paper takes Shenzhen, China, as the research object. Based on the UrbanEV multi-source open dataset and integrating charging operation data, spatial data, weather data, and POI data, an hourly-level charging station occupancy analysis dataset is constructed, and a systematic study on spatiotemporal feature analysis, demand forecasting, and optimization strategy evaluation is conducted. First, key spatiotemporal features of charging station occupancy are extracted through data cleaning, multisource information integration, and feature engineering; second, three types of prediction models - multiple linear regression, ARIMA, and TimesNet - are constructed, and model performance is uniformly evaluated using RMSE, MAPE, and RAE; third, sensitivity analysis is conducted using the control variable method to identify the main driving factors affecting occupancy; finally, optimization strategies are designed from three dimensions: dynamic pricing, station layout optimization, and peak-shifting charging guidance, and their implementation effects are evaluated through scenario simulation. Results show significant spatiotemporal heterogeneity in Shenzhen's charging station occupancy. Electricity price has the strongest negative impact, catering POI density is a key spatial driver, while weather effects are weak. Multiple linear regression outperforms ARIMA and TimesNet in accuracy and interpretability. Off-peak guidance yields the highest improvement (17.05%), followed by layout optimization (16.89%), while dynamic pricing is least effective. The research results can provide a theoretical basis and decision support for the planning, operation management, and time-based pricing optimization of urban charging infrastructure, and are of reference value for improving charging service efficiency and promoting the coordinated development of new energy vehicles and urban energy systems. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 18245463 |
| DOI: | 10.53136/97912218296861 |
