Answers are grounded in this paper and cite their sources.
Experimental study demonstrates superior accuracy of artificial neural networks for predicting persimmon fruit leather drying, indicating optimized processing parameters.
An engineering perspective using computer-aided comparison were introduced in this research to investigate the effects of starch concentration, product thickness and drying temperature on the drying time, effective moisture diffusivity and drying kinetics of fifteen differently treated persimmon fruit leather samples. The best descriptive mathematical model of the drying behaviors of the leather products was found by testing the equations of Page, Midilli and others, Two-term and Logarithmic models. Fick’s second law model was used to calculate the water diffusion coefficient and the diffusivity mechanism was modeled with feed forward backpropagation artificial neural network and response surface methodologies. The product thickness and drying temperature were the most effective operating parameters. The drying process for persimmon leather production took place in a falling rate period over a minimum of 120 minutes. The values of effective moisture diffusivity varied from 1.1˟10-7 to 13.8˟10-7 m2/s. Semi-theoretical Midilli and others model presented a better fit depending on the statistical analyses. The artificial network produced a more powerful prediction than the response surface methodology with higher R-squared value (0.9846), lower mean absolute error (0.0567) and root mean square error (0.0715).
No takes yet. Share an insight, caveat, or question.
Derya Dursun (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: