Vehicle-to-grid (V2G) technology has proven to be a promising solution for integrating electric vehicles (EVs) into the electricity grid, offering benefits such as grid stabilization and demand response. Predicting the aggregate available capacity (AAC) of EVs is crucial for effectively utilizing their energy storage capabilities. Here, a comprehensive methodological framework for predicting AAC in V2G systems is presented. It mainly includes data preprocessing and feature selection methods tailored to manage complex datasets with multiple data sources such as GPS, weather, vehicle characteristics, historical data, and calendar information. In addition, data augmentation methods are presented to address the problem of data scarcity that is typical of EV infrastructures. The core of such a framework then focuses on interpretable predictive models based on explainable machine learning or a state-space representation. The discussion on the framework under development aims to highlight the importance of interpretable models in V2G systems and provide insights into future research directions for such a prominent area, considering the evolution of the energy sector.

Patanè, L., Sapuppo, F., Comi, A., Napoli, G., Xibilia, M.g. (2024). An explainable model framework for vehicle-to-grid available aggregated capacity prediction. In 2024 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) (pp.652-657). New York : IEEE [10.1109/metroxraine62247.2024.10796291].

An explainable model framework for vehicle-to-grid available aggregated capacity prediction

Comi, Antonio;
2024-01-01

Abstract

Vehicle-to-grid (V2G) technology has proven to be a promising solution for integrating electric vehicles (EVs) into the electricity grid, offering benefits such as grid stabilization and demand response. Predicting the aggregate available capacity (AAC) of EVs is crucial for effectively utilizing their energy storage capabilities. Here, a comprehensive methodological framework for predicting AAC in V2G systems is presented. It mainly includes data preprocessing and feature selection methods tailored to manage complex datasets with multiple data sources such as GPS, weather, vehicle characteristics, historical data, and calendar information. In addition, data augmentation methods are presented to address the problem of data scarcity that is typical of EV infrastructures. The core of such a framework then focuses on interpretable predictive models based on explainable machine learning or a state-space representation. The discussion on the framework under development aims to highlight the importance of interpretable models in V2G systems and provide insights into future research directions for such a prominent area, considering the evolution of the energy sector.
2024 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE)
St Albans, London, UK
2024
IEEE
Rilevanza internazionale
contributo
2024
Settore ICAR/05
Settore CEAR-03/B - Trasporti
English
Vehicle-to-grid; Data-driven predictive model; Time-series prediction; Machine; Interpretable models; Explainable AI
Intervento a convegno
Patanè, L., Sapuppo, F., Comi, A., Napoli, G., Xibilia, M.g. (2024). An explainable model framework for vehicle-to-grid available aggregated capacity prediction. In 2024 IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering (MetroXRAINE) (pp.652-657). New York : IEEE [10.1109/metroxraine62247.2024.10796291].
Patanè, L; Sapuppo, F; Comi, A; Napoli, G; Xibilia, Mg
File in questo prodotto:
File Dimensione Formato  
An_Explainable_Model_Framework_for_Vehicle-to-Grid_Available_Aggregated_Capacity_Prediction_published.pdf

solo utenti autorizzati

Descrizione: full paper
Tipologia: Versione Editoriale (PDF)
Licenza: Copyright dell'editore
Dimensione 2.63 MB
Formato Adobe PDF
2.63 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2108/398084
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact