This paper addresses the accurate prediction of charging demand for freight electric vehicles (FEVs) using GPS data, crucial for optimizing charging infrastructure planning. We introduce the Developing Geographic PageRank (DGPR) model to improve prediction accuracy. DGPR utilizes GPS data to process origin-destination (OD) information, considering directional factors and social dimensions like points of interest (POI) and land use. It divides the study area into grid cells, creating a directed graph from OD data, and employs the PageRank algorithm to determine charging demand intensity for each area. The results underscore the DGPR model’s significant impact on precise FEV charging demand prediction. This research facilitates efficient charging infrastructure planning, promoting the sustainable development and effective operation of freight electric vehicle transportation systems.
Freight Electric Vehicle Charging Demand Forecasting Using GPS Data by Developing Geographic PageRank Model
24th COTA International Conference of Transportation Professionals ; 2024 ; Shenzhen, China
CICTP 2024 ; 632-641
11.12.2024
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Electric vehicle charging demand forecasting using deep learning model
Taylor & Francis Verlag | 2022
|Urban-range electric vehicle charging demand prediction method considering geographic information
Europäisches Patentamt | 2020
|Urban freight demand forecasting: A mixed quantity/delivery/vehicle-based model
Online Contents | 2014
|Forecasting Short-Term Freight Transportation Demand: Poisson STARMA Model
British Library Conference Proceedings | 1998
|