TY - JOUR
T1 - Artificial Intelligence for Electric Vehicle Infrastructure
T2 - Demand Profiling, Data Augmentation, Demand Forecasting, Demand Explainability and Charge Optimisation
AU - Sumanasena, Vidura
AU - Gunasekara, Lakshitha
AU - Kahawala, Sachin
AU - Mills, Nishan
AU - De Silva, Daswin
AU - Jalili, Mahdi
AU - Sierla, Seppo
AU - Jennings, Andrew
N1 - Funding Information:
This study was funded by the Victorian Higher Education State Investment Fund (VHESIF) for “Electrifying Victoria’s Future Fleet: Barriers and Opportunities” and the La Trobe University Net Zero Carbon Emissions Project.
Publisher Copyright:
© 2023 by the authors.
PY - 2023/3
Y1 - 2023/3
N2 - Electric vehicles (EVs) are advancing the transport sector towards a robust and reliable carbon-neutral future. Given this increasing uptake of EVs, electrical grids and power networks are faced with the challenges of distributed energy resources, specifically the charge and discharge requirements of the electric vehicle infrastructure (EVI). Simultaneously, the rapid digitalisation of electrical grids and EVs has led to the generation of large volumes of data on the supply, distribution and consumption of energy. Artificial intelligence (AI) algorithms can be leveraged to draw insights and decisions from these datasets. Despite several recent work in this space, a comprehensive study of the practical value of AI in charge-demand profiling, data augmentation, demand forecasting, demand explainability and charge optimisation of the EVI has not been formally investigated. The objective of this study was to design, develop and evaluate a comprehensive AI framework that addresses this gap in EVI. Results from the empirical evaluation of this AI framework on a real-world EVI case study confirm its contribution towards addressing the emerging challenges of distributed energy resources in EV adoption.
AB - Electric vehicles (EVs) are advancing the transport sector towards a robust and reliable carbon-neutral future. Given this increasing uptake of EVs, electrical grids and power networks are faced with the challenges of distributed energy resources, specifically the charge and discharge requirements of the electric vehicle infrastructure (EVI). Simultaneously, the rapid digitalisation of electrical grids and EVs has led to the generation of large volumes of data on the supply, distribution and consumption of energy. Artificial intelligence (AI) algorithms can be leveraged to draw insights and decisions from these datasets. Despite several recent work in this space, a comprehensive study of the practical value of AI in charge-demand profiling, data augmentation, demand forecasting, demand explainability and charge optimisation of the EVI has not been formally investigated. The objective of this study was to design, develop and evaluate a comprehensive AI framework that addresses this gap in EVI. Results from the empirical evaluation of this AI framework on a real-world EVI case study confirm its contribution towards addressing the emerging challenges of distributed energy resources in EV adoption.
KW - artificial intelligence
KW - charge optimisation
KW - demand explainability
KW - demand forecasting
KW - demand profiling
KW - electric vehicles
KW - EV data augmentation
UR - http://www.scopus.com/inward/record.url?scp=85149755191&partnerID=8YFLogxK
U2 - 10.3390/en16052245
DO - 10.3390/en16052245
M3 - Article
AN - SCOPUS:85149755191
SN - 1996-1073
VL - 16
JO - Energies
JF - Energies
IS - 5
M1 - 2245
ER -
