Use of Artificial Neural Network to Predict Temperature, Moisture, and Fat in Slab-Shaped Foods with Edible Coatings During Deep-Fat Frying

ABSTRACT: An artificial neural network (ANN) was developed to predict heat and mass transfer during deep-fat frying of infinite slab-shaped foods coated with edible films. Frying time, slab half-thickness, film thickness, food initial temperature, oil temperature, moisture diffusivity of food and film, fat diffusivity through food and film, thermal diffusivity of food, heat transfer coefficient, initial moisture content of food, and initial fat content of food (mf_o) were inputs. Temperature at the center (T₁), average temperature (T_ave), fat content (mf_ave), and moisture content (m_ave) of food were outputs. Four ANNs with 50 nodes each in 2 hidden layers with learning rate = 0.7 and momentum = 0.7 provided most accurate outputs, that is maximum absolute errors for T₁ and T_ave were < 1.2 °C, < 0.004 db for m_ave, and < 0.003 db for mf_ave. The predictions of mf varied linearly with mf.