Mathematical modelling and analysis of sorption isotherms of onions (Allium cepa L.)

The equilibrium moisture content (EMC) of onion slices was investigated under various drying conditions characterized by different temperatures and relative humidity levels. To identify the most suitable model for representing the experimental data, six EMC models were evaluated. The adsorption isotherms demonstrated a decreasing trend in EMC with increasing temperature at a constant relative humidity. Among the models assessed, the Modified Exponential and Modified Oswin models provided the most accurate predictions of EMC for onion slices. The Modified Exponential model was chosen as the optimal model for drying onion slices due to its superior performance, evidenced by a high coefficient of determination (R²; = 0.9935) and lower values of error metrics, specifically the reduced chi-square (χ²; = 0.00038), root mean square error (RMSE = 0.01807), and mean relative deviation modulus (P = 8.26). To further validate these findings, an artificial neural network (ANN) model was implemented in MATLAB, which corroborated the results obtained from the mathematical models. The comparative analysis between the mathematical and ANN models indicated a high degree of agreement with the sorption characteristics of sliced onions.