A Data Driven Performance Analysis Approach for Enhancing the QoS of Public Charging Stations

The gaining momentum of Electric Vehicles' (EV) market is hindered mainly due to the range anxiety. Accordingly, a ubiquitous charging station (CS) network is becoming indispensable. However, due to the lack of reservations or check-in policies in EV charging, users and operators are not provid...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on intelligent transportation systems 2022-08, Vol.23 (8), p.11116-11125
Main Authors: Antoun, Joseph, Kabir, Mohammad Ekramul, Atallah, Ribal F., Assi, Chadi
Format: Article
Language:English
Subjects:
Analytical models
charging station
Charging stations
Data acquisition
Data analysis
Data models
Electric vehicle charging
Electric vehicles
Measurement
modeling
Quality of service
Queueing analysis
queuing
Queuing theory
real charging data
State of charge
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The gaining momentum of Electric Vehicles' (EV) market is hindered mainly due to the range anxiety. Accordingly, a ubiquitous charging station (CS) network is becoming indispensable. However, due to the lack of reservations or check-in policies in EV charging, users and operators are not provided proper information regarding waiting time at public CSs. This renders users reluctant to use public CSs. In addition, this incomplete information, creates difficulties in the deployment and operation of CSs. Evidently, there is a need to improve the Quality-of-Service (QoS) such as minimizing the waiting time or blocking probability. Therefore, CS owners rely during the designing stage on some theoretical distribution for the associated parameters assumption ( i.e. , battery capacity, charging demand, charging time, waiting time, etc.). To alleviate this situation, instead of depending on theoretical assumptions, real CSs usage data for EV charging are analyzed to acquire data driven distributions. Moreover, since the charging rate is dependent on the State of Charge (SoC), instead of a constant charging rate, a SoC dependent charging model based on real experimental data is proposed and evaluated with real data. Finally, exploiting the acquired distributions and charging model, variations of the M/G/k queuing system to approximate the waiting time, reneging probability and blocking probability is implemented. A detailed simulation is placed and the findings provide a direction for CS owners in determining the capacity ( i.e. , number of outlets) or parking area size to enhance the QoS.
ISSN:1524-9050
1558-0016
DOI:10.1109/TITS.2021.3100875