Μέθοδοι προβλεπτικής αναλυτικής με εφαρμογές σε ηλεκτρικά οχήματα
Predictive analytics methods with applications οn electric vehicles
View/
Keywords
Ηλεκτρικά οχήματα ; Ηλεκτρική ενέργεια ; Διαθεσιμότητα σταθμών φόρτισης ; Ζήτηση ηλεκτρικής ενέργειας ; ΠρόβλεψηAbstract
This paper focuses on forecasting electric vehicle charging demand using Recurrent Neural Networks (RNNs) and Bidirectional Networks (Bi-RNNs). It examines the development of methods for
generating data that improve prediction accuracy. The study aims to forecast energy consumption
during charging sessions and solve the binary classification problem of predicting charging station availability through a Hybrid Long Short-Term Memory (Hybrid LSTM) model. The research
employs advanced machine learning techniques and investigates data generation methodologies,
including temporal and environmental variables that affect charging behavior. The results demonstrate the effectiveness of the proposed hybrid model in predicting charging station availability,
while highlighting the challenges in building accurate charging demand forecasting models. Additionally, the study provides valuable insights for future research directions and opportunities for
progress in predictive modeling methodologies.
Postgraduate Studies Programme
Κυβερνοασφάλεια και Επιστήμη ΔεδομένωνDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα ΠληροφορικήςNumber of pages
69Language
GreekCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα