Modeling the effect of ultrasound on viscosity, consistency coefficient, and flow behavior index of different concentrations of xanthan gum

The use of ultrasonic waves to change the structure of the gums leads to the modification and improvement of their functional characteristics and rheological properties. In this research, the effects of ultrasonic intensity and treatment time on apparent viscosity, consistency coefficient, and flow behavior index of different concentrations of xanthan gum were investigated and modeled. Genetic algorithm-artificial neural network method with three inputs (ultrasonic power, treatment time and gum concentration) and three outputs (viscosity, consistency coefficient, and flow behavior index) was used to model the process. The apparent viscosities of the xanthan gum control sample (untreated) at concentrations of 0.1, 0.15, and 0.2% were 21.0, 39.9, and 66.5 mPa.s, respectively. The results of this research showed that gum viscosity decreased with increasing intensity and duration of ultrasound application. Ultrasonic treatment for 20 min significantly reduced the apparent viscosity of xanthan gum from 39.9 to 23.2 mPa.s (p< 0.05). The genetic algorithm-artificial neural network modeling results showed that the network with 3-5-3 structure in a hidden layer and using the hyperbolic tangent activation function can predict the rheological parameters of xanthan gum with high correlation coefficient and low error value. Values of mean squared error (MSE), normalized mean squared error (NMSE), mean absolute error (MAE), and correlation coefficient (r) to predict the apparent viscosity of xanthan gum were 73.17, 0.20, 6.48, and 0.90, respectively. Based on the results of the sensitivity analysis test, ultrasonic treatment intensity was the most effective factor in changing the apparent viscosity, consistency coefficient, and flow behavior index of xanthan gum.