Simulating preservation processes of carcasses using real geometries

GOÑI, S. M.; PURLIS, E.; SALVADORI, V. O.

Viña del Mar, Chile

Congreso; Tenth International Congress on Engineering and Food; 2008

Chilling (or freezing) of carcasses after slaughter is important from many points of view. These processes affect microbial growth, weight loss, energy consumption, and quality factors such as meat tenderness. Furthermore, whole carcasses or large cuts of them may be cooked or heat treated in a later step, where microbiological and quality aspects must be satisfied. Therefore, the design of preservation processes should rely on accurate physical models. To solve this problem, various methods have been proposed. For instance, simplified geometric models, which make use of a simple shape to represents parts of a carcass, and the well-known shape factor approach, have been applied. Although they are easy-to-use techniques, they can not describe heat and mass transfer in an accurate manner. The objectives of this work were: (i) to construct real geometric models for different carcasses and (ii) to simulate common preservation processes such as chilling or cooking of carcasses. Firstly, geometries of carcasses (lamb, pork and chicken) were constructed. For this aim, magnetic resonance images from cross-sections of each carcass were acquired. Irregular boundaries of each section were approximated by B-Spline curves, which were then assembled using a lofting technique (COMSOL Multiphysics) to obtain a closed interpolating surface. This surface was finally converted in a 3D solid object representing the real carcass. Secondly, chilling processes were simulated by solving the heat transfer problem using the finite element method, over the constructed real geometries previously constructed. High realistic geometric models were obtained. All complex shapes and details of carcasses surfaces could be extracted using the developed technique. Moreover, image acquisition was non-destructive and effective in terms of time and noiseless output. Respect to the simulated processes, chilling times and complete information such as temperature profiles could be obtained. The constructed geometric models could be used to simulate other processes, or could be re-scaled to represent various sizes of the same kind of animal. Also, the fact of solving a problem of heat transfer over very complex domains extends the horizon of possibilities in the food engineering field.