Mathematical modelling of bulk stored onions in transport containers : a thesis presented in partial fulfilment of the requirements for the degree of Master of Horticultural Science in Agricultural Engineering at Massey University

Abstract

Export onion bulbs are predominantly transported from New Zealand loose in sacks which are bulk loaded into intermodal transport containers. Product respiratory heat, water vapour, and volatiles are dispensed of by a fan unit installed in the end of the container, ventilating the stow by forcing ambient air from a false floor up through the crop and exhausting the air from a head space. The objective of this study was to mathematically model this system with respect to onion bulb temperature and weight loss, and internal container air temperature and relative humidity. These product and flowfield variables were predicted at different locations within the transport vessel. Bulb temperature and weight loss were simulated as dynamic variables using ordinary differential equations, and air temperature and relative humidity were simulated as quasi steady state variables using algebraic equations. A validation experiment was conducted to evaluate the simulation model by placing temperature and humidity sensors throughout the product and flowfield space measuring the respective properties. Onion and air temperatures were predicted with satisfactory accuracy in almost all measured locations of the container. Prediction of relative humidity varied considerably throughout the container, although excessive sensor errors were identified casting suspicion on some validation measurements. Simulated relative humidity could not therefore be fully verified. Bulb weight loss was predicted with variable levels of accuracy. Significant variability in the validation data was evident in the upper and lower regions of the container preventing complete model validation. Central regions of the container were simulated with satisfactory accuracy. A model sensitivity analysis revealed that container ventilation rate strongly influenced model performance with respect to temperature and relative humidity. The mass transfer coefficient, as expected, was most influential over product weight loss.