< Back to previous page

Publication

Advanced Computational Fluid Dynamics modelling for a more sustainable and energy efficient postharvest cold chain of pome fruit

Book - Dissertation

The demand for fresh fruit is high and expected to increase significantly in the near future due to consumers targeting a healthy diet. This inherently leads to the need for high-quality fruit being available year-round for fresh consumption. However, freshly harvested fruit solely rely on their internal reserves to maintain their respiration metabolism. By lowering fruit temperature, the respiration rate can be reduced, thus retarding the onset of fruit senescence and extending their storage life. The latter is usually in combination with controlled atmosphere conditions or other postharvest treatments. In order to be flexible in reply to market demands while ensuring year-round availability of high quality produce to the consumers, improving current cooling and storage strategies is necessary. However, possibilities for experimental tests are limited due to costs and fruit availability, while the number of design variables within the cold chain is high. A cost-effective alternative to experiments is numerical modelling with computational fluid dynamics (CFD). The main objective of this PhD research was to develop, apply and analyse CFD models to simulate and improve cooling processes of pome fruit in different steps of the postharvest cold chain. In particular, the aim was to improve predictions of transient and spatially resolved cooling processes at different spatial scales; from individual fruit to large industrial cool rooms. To reach this goal, various advanced and novel numerical approaches were explored to model the fruit stacks and when possible, validated with experiments. Ultimately, these models were employed to evaluate practical postharvest cooling applications. During long-term storage of apples, the main costs are associated with the energy consumption of the cooling process. In the first research chapter, a CFD model was developed to allow quantitative evaluation of energy saving measures during long-term storage of apples in industrial cool rooms, while considering effects on fruit quality. The porous medium approach was adopted to simulate cold storage of large fruit bins stacked in industrial cool rooms. A transient CFD model was used to evaluate three cooling scenarios with a different temperature differential around the optimal apple storage temperature. The dynamic coolstore behaviour was captured by extending the CFD model with an evaporator and controller model. In addition, this model was coupled with a kinetic model of apple firmness to predict the time and spatial evolution of apple quality in the coolstore during long-term storage. Implementing a small temperature differential was shown to result in a better overall performance in terms of uniformity, final product quality and energy consumption. The cooling rate and uniformity, and energy consumption of forced air precooling greatly depend on the aerodynamic design and configuration of the fruit packages. In the second research chapter, a CFD model was developed and applied to compare apple fruit cooling performance of a conventional corrugated fibreboard cardboard (CFC) box to that of an alternative reusable plastic crate (RPC). Since the porous medium approach previously used is only applicable in configurations where the crate to fruit dimension is sufficiently large, an explicit modelling approach was used instead. In this model, fruit were approximated as regular spheres with a fixed diameter. Computations were verified with experiments and captured trade-offs between cooling rates and energy consumption. A mismatch between the position of vent holes and fruit-supporting trays created isolated regions for airflow, resulting in large temperature heterogeneities in the standard CFC box. Although both cooling uniformity and energy use were found to be the best for RPCs, high airflow rates at low temperatures might cause chilling injury to the apples. Fruit shape and size show considerable variability, and thus affect the cooling process significantly. In the third research chapter, a new explicit CFD modelling approach that considers random stacking and variable fruit shapes was developed and evaluated against simpler approaches. Variable 3D apple and pear models were created by means of a validated geometric model generator based on X-ray computed tomography images of individual fruit. The fruit were randomly stacked into a geometrical model of a CFC box using a Discrete Element Method. A horizontal forced-air cooling process was simulated for three such apple filling patterns using CFD and results were compared to those obtained with fruit represented by equivalent spheres. No significant difference in average aerodynamic resistance between the real apple shape and its spherical representation was found. However, the degree of cooling uniformity between individual fruit was overestimated: cooling uniformity decreased when realistic fruit shapes were used. This difference between real and simplified product shapes was even larger for a box filled with pear fruit that are more different from a spherical shape. Validation of explicit CFD models requires one-to-one comparison of predicted and measured temperature profiles in specific fruit stacks. In the final research chapter, CFD models were developed from X-ray computed tomography scans of boxes filled with pear fruit. This allowed direct validation against experimental measurements in the same stacking geometry and assessing cooling performance differences caused by variable fruit shapes. The actual filling patterns of different pear cultivars, with significant shape differences, in a standard cardboard package were reconstructed and implemented in an explicit CFD model of a horizontal forced-air cooling process. The simulated cooling profiles were successfully compared with experiments with only a maximum difference of 9 % in cooling time. The contribution of the filling pattern to the overall pressure drop over the box was quantified to be only 3 %, and intercultivar differences were negligible. The specific stacking arrangement of the more elongated 'Conference' pears obstructed vertical airflow in the box causing a larger fruit temperature heterogeneity during cooling than boxes filled with the more spherically shaped 'Durondeau' and 'Doyenné' pears. The variable size and shape of the pear cultivars resulted in a different filling pattern which clearly affected the cooling uniformity, but not the average cooling rate. With the continuous increase and improvement in computational resources, it can be expected that explicit CFD modelling will become the rule instead of the exception. Towards this end, the explicit modelling procedures elaborated in this dissertation have laid the ground works to develop more detailed, realistic and accurate CFD models for postharvest cooling applications. Although focussed on pome fruit, it can easily be extended towards other horticultural produce types where dedicated temperature control schemes during its postharvest life are required. Also other aspects of long-term storage (such as the gas conditioning, postharvest treatments) can be optimised in a more efficient way with the developed models.
Publication year:2019
Accessibility:Open