Status of Physics-Based Models in the Design of Food Products, Processes, and Equipment

Machine learning-based modeling in food processing applications: State of the art

Food processing is a complex, multifaceted problem that requires substantial human interaction to optimize the various process parameters to minimize energy consumption and ensure better-quality products. The development of a machine learning (ML)-based approach to food processing applications is an exciting and innovative idea for optimizing process parameters and process kinetics to reduce energy consumption, processing time, and ensure better-quality products; however, developing such a novel approach requires significant scientific effort. This paper presents and evaluates ML-based approaches to various food processing operations such as drying, frying, baking, canning, extrusion, encapsulation, and fermentation to predict process kinetics. A step-by-step procedure to develop an ML-based model and its practical implementation is presented. The key challenges of neural network training and testing algorithms and their limitations are discussed to assist readers in selecting algorithms for solving problems specific to food processing. In addition, this paper presents the potential and challenges of applying ML-based techniques to hybrid food processing operations. The potential of physics-informed ML modeling techniques for food processing applications and their strategies is also discussed. It is expected that the potential information of this paper will be valuable in advancing the ML-based technology for food processing applications.

Estimation of Isentropic Compressibility of Biodiesel Using ELM Strategy: Application in Biofuel Production Processes

Isentropic compressibility is one of the significant properties of biofuel. On the other hand, the complexity related to the experimental procedure makes the detection process of this parameter time-consuming and hard. Thus, we propose a new Machine Learning (ML) method based on Extreme Learning Machine (ELM) to model this important value. A real database containing 483 actual datasets is compared with the outputs predicted by the ELM model. The results of this comparison show that this ML method, with a mean relative error of 0.19 and R 2 values of 1, has a great performance in calculations related to the biodiesel field. In addition, sensitivity analysis exhibits that the most efficient parameter of input variables is the normal melting point to determine isentropic compressibility.

Modelling Volume Change and Deformation in Food Products/Processes: An Overview

Volume change and large deformation occur in different solid and semi-solid foods during processing, e.g., shrinkage of fruits and vegetables during drying and of meat during cooking, swelling of grains during hydration, and expansion of dough during baking and of snacks during extrusion and puffing. In addition, food is broken down during oral processing. Such phenomena are the result of complex and dynamic relationships between composition and structure of foods, and driving forces established by processes and operating conditions. In particular, water plays a key role as plasticizer, strongly influencing the state of amorphous materials via the glass transition and, thus, their mechanical properties. Therefore, it is important to improve the understanding about these complex phenomena and to develop useful prediction tools. For this aim, different modelling approaches have been applied in the food engineering field. The objective of this article is to provide a general (non-systematic) review of recent (2005-2021) and relevant works regarding the modelling and simulation of volume change and large deformation in various food products/processes. Empirical- and physics-based models are considered, as well as different driving forces for deformation, in order to identify common bottlenecks and challenges in food engineering applications.

Toward predictive food process models: A protocol for parameter estimation

Mathematical models, in particular, physics-based models, are essential tools to food product and process design, optimization and control. The success of mathematical models relies on their predictive capabilities. However, describing physical, chemical and biological changes in food processing requires the values of some, typically unknown, parameters. Therefore, parameter estimation from experimental data is critical to achieving desired model predictive properties. This work takes a new look into the parameter estimation (or identification) problem in food process modeling. First, we examine common pitfalls such as lack of identifiability and multimodality. Second, we present the theoretical background of a parameter identification protocol intended to deal with those challenges. And, to finish, we illustrate the performance of the proposed protocol with an example related to the thermal processing of packaged foods.

Sodium Chloride Diffusion in Low-Acid Foods during Thermal Processing and Storage

This study aimed at modeling sodium chloride (NaCl) diffusion in foods during thermal processing using analytical and numerical solutions and at investigating the changes in NaCl concentrations during storage after processing. Potato, radish, and salmon samples in 1% or 3% NaCl solutions were heated at 90, 105, or 121 °C for 5 to 240 min to simulate pasteurization and sterilization. Selected samples were stored at 4 or 22 °C for up to 28 d. Radish had the largest equilibrium NaCl concentrations and equilibrium distribution coefficients, but smallest effective diffusion coefficients, indicating that a greater amount of NaCl diffused into the radish at a slower rate. Effective diffusion coefficients determined using the analytical solution ranged from 0.2 × 10(-8) to 2.6 × 10(-8) m²/s. Numerical and analytical solutions showed good agreement with experimental data, with average coefficients of determination for samples in 1% NaCl at 121 °C of 0.98 and 0.95, respectively. During storage, food samples equilibrated to a similar NaCl concentration regardless of the thermal processing severity. The results suggest that sensory evaluation of multiphase (solid and liquid) products should occur at least 14 d after processing to allow enough time for the salt to equilibrate within the product.