Résumé: This comprehensive study reviews the latest and most popular artificial intelligence (AI) techniques utilised for estimating pan evaporation (Ep), an essential parameter for water resource management and irrigation planning. Through an extensive evaluation of 76 papers published between 2006 and 2022, this study analyses the input data categories, time steps, properties, and capabilities of different AI models used for estimating Ep across various regions. The reviewed papers offer partial and comprehensive observations, providing valuable insights for researchers looking to model Ep in similar studies. Furthermore, this study proposes innovative theories and approaches to enhance the efficacy of Ep modelling in the relevant analysis domain. While hybrid AI techniques have gained popularity due to their perceived superiority over standalone deep learning and machine learning approaches, they often pose significant operational and computational challenges for Ep forecasting. As such, the study strongly recommends the use of transformer neural networks for Ep estimation, given their unique architecture and promising performance across various fields. Overall, this study presents a comprehensive and up-to-date overview of the latest AI-based techniques for estimating Ep and highlights the most promising approaches for future research.

Langue: Anglais
Index décimal 006.3 .Intelligence artificielle (ouvrages généraux sur l'intelligence artificielle et la science cognitive, reconnaissance de formes comme outil de l'intelligence artificielle, systèmes de questions et réponses)
Thème Informatique

Mots clés:
Deep learning
Artificial intelligence
Machine learning
Pan evaporation
Transformer neural network