Mathematical Modelling of the Mechanical Properties of Four Varieties of Brine Pickles Using Neural Networks

The study assesses the impact of selected fruit properties and the souring process parameters on the mechanical properties of four varieties of field cucumbers (Śremianin, Śremski, Polan and Izyd) harvested from 2014 to 2016. The analyses focused on the changes in the values of the puncture strength parameters of the peel and the mesocarp, the deformation from the moment of puncture and the energy needed to puncture the peel and the mesocarp of the selected cucumber varieties relative to the year of harvest, fruit size, type of brine, puncture location and souring time. Neural networks were used to model the relationships studied. Sensitivity analysis of the obtained models showed that the length fraction had the greatest impact on the puncture strength and the energy needed to puncture the peel and the mesocarp. On the other hand, deformation was most affected by the souring time and the brine composition.