Application of artificial neural networks in agricultural product drying

Drying is a critical process in postharvest handling of agricultural products, significantly impacting their preservation, quality and market value. This review explores the application of artificial neural networks (ANNs) in the drying of agricultural products, focusing on four key areas: moisture ratio prediction, quality detection, process optimization, and control systems. ANNs have remarkable potential for accurately modeling the highly nonlinear drying processes, optimizing energy consumption and improving product quality. Despite their advantages, challenges, such as data dependency, computational complexity and model interpretability, remain. This review highlights recent advancements in ANN-based drying models, discusses their limitations, and envisions future directions, including the integration of ANNs with other artificial intelligence (AI) technologies and the development of hybrid models. Through literature research in recent years, it has been found that ANN have achieved development from standard ANN to machine learning and even deep learning. Deep learning is gradually becoming the mainstream direction in agricultural product drying. The findings underscore the transformative potential of AI in revolutionizing the agricultural drying industry, paving the way for more efficient, sustainable and intelligent drying systems.