Scholarlink

<p>The availability of fast charging stations (FCSs) is one of the key motivating factors for the widespread adoption of electric vehicles (EVs) by the community. However, as the development cost of FCSs is remarkably high, the optimal allocation and planning of FCSs are critically important. Moreover, EV charging is inherently influenced by several significant uncertainties, including the EV charging start time, charging duration, and the energy supplied. In this paper, a bi-stage stochastic model is proposed to address these uncertainties and is designed for the synthetic generation of transportation data by mining large-scale real-world historical data gathered from FCSs. To achieve this, the uncertainties in the aforementioned transportation parameters are accounted for, and different stochastic random variables (RVs) are correlated using the best-fitted copula, i.e., Student-t copula, in the first stage. In the second stage, artificial EV charging start times are effectively generated over a 5-year planning time horizon, a crucial parameter for FCS planning and operational studies. Furthermore, EV charging demand is forecasted stochastically with respect to the three RVs of interest, achieving an acceptable level of accuracy. Simulations are carried out in R Programming Language and MATLAB.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Probabilistic multivariate modeling of electric vehicles charging start time in fast charging stations

  • 期刊论文

  • 发表日期：

    2026年2月23日

查看全文
  • Payam Farhadi,
  • Seyed-Masoud Moghaddas-Tafreshi,
  • Amir Shahirinia

摘要

The availability of fast charging stations (FCSs) is one of the key motivating factors for the widespread adoption of electric vehicles (EVs) by the community. However, as the development cost of FCSs is remarkably high, the optimal allocation and planning of FCSs are critically important. Moreover, EV charging is inherently influenced by several significant uncertainties, including the EV charging start time, charging duration, and the energy supplied. In this paper, a bi-stage stochastic model is proposed to address these uncertainties and is designed for the synthetic generation of transportation data by mining large-scale real-world historical data gathered from FCSs. To achieve this, the uncertainties in the aforementioned transportation parameters are accounted for, and different stochastic random variables (RVs) are correlated using the best-fitted copula, i.e., Student-t copula, in the first stage. In the second stage, artificial EV charging start times are effectively generated over a 5-year planning time horizon, a crucial parameter for FCS planning and operational studies. Furthermore, EV charging demand is forecasted stochastically with respect to the three RVs of interest, achieving an acceptable level of accuracy. Simulations are carried out in R Programming Language and MATLAB.