Artificial Intelligence Based Optimal Operation of Green Hydrogen Refueling

KIRAT O., ÇİÇEK A.

32nd Telecommunications Forum, TELFOR 2024, Belgrade, Sırbistan, 26 - 27 Kasım 2024, (Tam Metin Bildiri) identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/telfor63250.2024.10819143
  • Basıldığı Şehir: Belgrade
  • Basıldığı Ülke: Sırbistan
  • Anahtar Kelimeler: Cloud computing, deep learning, fuel cell electric vehicle, green hydrogen, optimization, time-series prediction
  • Trakya Üniversitesi Adresli: Evet

Özet

This study presents an AI-based optimized energy management model for a hydrogen refueling station with a PV system and fuel cell (FC). Using the Time-series Dense Encoder (TiDE) model for day-ahead PV generation and electricity price forecasting, with real data from Spain, the station efficiently manages hydrogen demand and electricity sales. Despite lower-than-expected PV production (11,559.39 kWh vs. 12,991.71 kWh), real-time revenue rose to 1,144.84 Euros, demonstrating operational flexibility and profit.