Design and evaluation of a prototype solar energy driven onion curing system using CFD modeling

In developing countries like Ethiopia, where the time of harvesting coincides with the dry season, farmers cure onion bulbs naturally on the field. Field curing generally takes longer than artificial curing and results in more losses and reduced quality of the bulbs due to increased risks for infestations and uncontrolled suboptimal drying conditions. Large-scale artificial curing systems are expensive and electrical energy supply is limited in rural areas. A CFD model was employed to design an alternative and sustainable onion curing system that can be deployed on the field in rural areas. The developed CFD model was validated by comparing the predicted air velocity, temperature, and mass loss to measured values on a prototype curing system operated on a field in Ethiopia. A good agreement between the model and experimental value was observed for the time profiles of temperature at different positions in the bulk of onions during curing, expressed by a root mean square error of 1.1°C in the temperature range from 28 to 47°C, 0.16 m s−1 in the velocity range from 0.1 to 2.5 m s−1, and 0.565% for the mass loss that ranged up to 6.35%. The developed model was used to assess the air velocity, temperature, and relative humidity distribution in order to get an insight into the uniformity of curing of onion bulbs using the develop alternative curing system. For all of the examined curing durations, the drying air temperature variation inside 80% of the porous medium was < 3°C. In the remaining 20% of the porous medium, a temperature variation of up to 6°C was observed. Thus, the newly designed and developed curing system was found to cure the onion bulbs uniformly. Moreover, its performance was evaluated experimentally and the onions were cured to a desirable level of curing for long-term storage within a total curing duration of 48 h. It is vital to consider bulb shrinkage, particularly in the neck, in order to further improve the model mass loss prediction.