Xuming Chen, Le Liu, Xiaoning Kang
The transition to a low-carbon transportation system is crucial for urban sustainable development, requiring the integration of electric vehicles, renewable energy, and intelligent transportation systems. The photovoltaic energy storage-charging station (PV-ES-CS), as an effective integration of these key technologies, plays a critical role in the planning of the low-carbon city. This paper considers the dynamic changes in traffic impedance and addresses the issues of siting, sizing of PV-ES-CS and traffic flow distribution in the PTNs, using data mining to assess the impact of multiple uncertainties. First, the dynamic changes in the transportation network are considered by establishing road impedance and user behavior models. The uncertainty of user route choices is incorporated into the traffic flow distribution, and the siting and sizing of PV-ES-CS are determined based on the flow distribution results. Next, a data mining model based on SHAP is proposed, The model identifies the components of the objective function that are most sensitive to uncertainty. Additionally, it extracts the key factors among multiple uncertainty sources. Then, sensitivity analysis is conducted on the impacts of these uncertainties on traffic flow and carbon emission costs. Finally, the proposed model is applied to a coupled network of the IEEE-33 node and a specific traffic network in Xi’an, Shaanxi Province, China. The results show that the proposed model effectively reduces road congestion and carbon emission costs, and the data mining framework can identify key variables, providing a comprehensive solution for low-carbon transportation city planning. • A PV-ES-CS planning method considering multiple uncertainties. • Considering dynamic changes in traffic networks and refining driver behavior. • Using a SHAP-based data mining method for sensitivity analysis. • Reducing carbon emission cost and alleviating traffic congestion.