Search

Application of machine learning (ML)-based algorithms in food drying is an exciting and innovative approach to advance the drying technology. In order to appropriately develop this novel approach in all aspects of food drying field, significant scientific research is required. The main aspects of food drying research are the determination of material properties, microstructural characterization, mathematical modeling, and process optimization. It is essential to express this fundamental information through ML-based algorithms to advance the food drying research. This paper aims to present a comprehensive review of the application of machine learning-based approaches in food drying modeling, property prediction, microstructural characterization, and process parameters optimization. Moreover, this paper discusses the possibilities and challenges to apply ML-based algorithms in multiscale modeling and microwave-based hybrid drying. It is expected that this review paper will be beneficial in advancing the machine learning-based food drying technology.

Drying is a complex simultaneous heat and mass transfer process where water is migrated from the sample to the drying air through continuous evaporation. The rate of evaporation can be accelerated by the intermittent application of microwave energy with convective drying, which is called intermittent microwave convective drying (IMCD). However, theoretical aspects of IMCD process is not clearly understood yet, which is required to obtain a better -quality product and for designing energy efficient IMCD drying systems. Therefore, several researchers have attempted to develop a theoretical modelling framework for IMCD. However, most of the existing models do not consider the spatial distribution of air velocity around the product, which is important as it significantly affects the heat and mass transfer coefficient throughout the sample and eventually alters the drying kinetics. This paper aims to address this research gap by developing an IMCD modelling framework through integrating a simultaneous heat and mass transfer model with a computational Fluid Dynamics (CFD) model (drying air flow) under the condition of volumetric heating. A 3D model was developed, and the simulated results were validated with the experimental results. It was found that the integration of fluid flow (CFD) with an IMCD model significantly affected the drying kinetics as heat and mass transfer coefficients spatially varied throughout the sample. Drying models may thus overpredict the drying kinetics like moisture content if the spatial distribution of air velocity is not taken into account. It was also found that the nonuniform microwave power absorption caused an inhomogeneous distribution of the sample temperature. The new findings and knowledge from this study will lead to the development of more accurate drying models.

Food materials, particularly plant-based food materials, are mostly cellular in structure with diverse structural heterogeneities and variable cellular properties. To uncover the fundamental drying process, the cellular-level understanding of simultaneous heat and mass transfer process and associated morphological changes is essential. Consequently, researchers have attempted to investigate microlevel transport and material properties and their spatial and temporal distribution using recent innovative and cutting-edge experimental methods and technologies. The aim of this paper is to review the latest experimental technologies that can be applied to investigate the micro-level features of food material during drying. A comprehensive overview of advanced experimental techniques including X-ray microtomography, nuclear magnetic resonance (NMR), Near Infrared (NIR) Spectroscopy, light microscopy, nanoindentation and atomic force microscopy (AFM) has been presented. Finally, a summary of the current challenges of micro-level investigation for food drying is presented at the end of this paper.