Open AccessAnalysis on Charging Demand of Shared Vehicle Based on Spatiotemporal Characteristic Variable Data Mining
Author(s) -
Haolin Wang,
Yongjun Zhang,
Haipeng Mao
Publication year - 2019
Publication title -
iop conference series. earth and environmental science
Language(s) - English
Resource type - Journals
eISSN - 1755-1307
pISSN - 1755-1315
DOI - 10.1088/1755-1315/345/1/012002
Subject(s) - grid , computer science , variable (mathematics) , shared space , set (abstract data type) , monte carlo method , order (exchange) , simulation , space (punctuation) , mathematical analysis , statistics , geometry , mathematics , finance , economics , programming language , operating system
The wide application of shared vehicles in the future will bring about tremendous importance to the power grid and planning of charging facilities. At present, there are flaws in the prediction methods for shared vehicles charging demand. Based on data mining of national household travel survey(NHTS), this paper constructs a two-dimensional dynamic traffic behaviour model supported by spatiotemporal feature variables. Then, in order to explore the characteristics of continuous charging and centralized charging of shared vehicles, two charging scenarios are set to construct a charging behaviour model. Finally, the Monte Carlo method is used to simulate the shared vehicle traffic charging behaviour, and get the result of the shared vehicle charging demand at different times and regions. The impact of the load on the grid is analyzed in the same time. The results show that the interactive spatial-temporal characteristic variables can reasonably describe the characteristics of time-space two-dimensional uncertain changes in shared vehicles and the method can make a scientific prediction of the shared vehicle charging demand.