Juice as a healthy drink is currently being consumed by a large percentage of the world's population. Juices are considered an essential component of the diet. In addition, the first step towards the production of fruit-based products such as marmalade, juice, soft drinks, alcoholic drinks, pastilles, sauce, and jelly is the extraction of juice. The applicability of the filter-press method for the extraction of black mulberry juice was investigated and optimised in this study. Considering cylinder diameter, layer height, and loading rate as independent variables, juice percent, juice density, final bulk density of pomace, final true density of pomace, final porosity of pomace, juice point time, elasticity modulus of bulk berries, specific consumed energy, and mean required power were determined. The force-deformation behaviour during the loading stage, stress-time behaviour during the stress-relaxation stage, and momentary juice percent during the test were modelled. The system's performance was optimised and verified practically based on minimising mean required power and maximising juice percent, capacity index (multiplication of input variables), and energy efficiency. The initial moisture content of the black mulberries was 86.36%, while the highest amount obtained for the percentage of extracted juice was 78.27%. Juice density, specific consumed energy, and mean required power ranged between 1016 and 1048 kg m −3 , 163.55–453.66 J, and 0.18–1.68 W, respectively. The optimisation and verification results showed that the maximum achievable desirability was 56.75%, and the maximum absolute error between predicted and practical values was less than 2%. • Black mulberry fruits were subjected to the axial compression load to extract juice. • Cylinder diameter, layer height, and loading rate were considered operating factors. • An exponential function modeled force-deformation behavior during the loading stage. • One-term Maxwell model was used to model stress-relaxation behavior. • Momentary juice mass (juice percent) behavior was modeled.
Read full abstract