Neural computing efforts for integrated simulation of ultrasound-assisted hydration kinetics of wheat

• Published in: Information processing in agriculture 2019-09, Vol.6 (3), p.357-374
• Main Authors: Shafaei, S.M., Nourmohamadi-Moghadami, A., Rahmanian-Koushkaki, H., Kamgar, S.
• Format: Article
• Language: English
• Subjects: Hydration rate; Hydration temperature; Hydration time; Moisture content; Water absorption
• ISSN: 2214-3173; 2214-3173
• DOI: 10.1016/j.inpa.2019.01.001

[Display omitted] •Neural computing simulation environments were developed to predict ultrasound-assisted hydration kinetics of wheat.•Results of the ANN strategy were compared to those of the ANFIS strategy.•The distinguished ANFIS environment was more accurate than the distinguished ANN environment.•The ANFIS results showed positive effect of ultrasound technology on water absorption.•The ANFIS simulation results improved the state of art in this domain. This study is dedicated to examine predictive ability of neural computing environments, based on artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) strategies, for integrated simulation of ultrasound-assisted hydration kinetics of wheat kernel. Hydration process was accomplished at five hydration temperatures of 30, 40, 50, 60 and 70 °C in ultrasonication conditions named control (without ultrasound treatment), US1 (25 kHz, 360 W) and US2 (40 kHz, 480 W). The hydration temperature, ultrasonication condition, and hydration time were used as input variables and moisture content was taken as output variable in the neural computing simulation environments. On account of statistical performance criteria, the distinguished ANFIS simulation environment with coefficient of determination of 0.991, root mean square error of 2.478% d.b., mean relative deviation modulus of 4.301% and average of absolute values of simulation residual errors of 1.863% d.b. was better performed than the distinguished ANN simulation environment. The ANFIS simulation results showed that individual or simultaneous increment of hydration temperature and hydration time caused nonlinear increment of moisture content at any given ultrasonication condition. Moreover, physical perception obtained from the integrated ANFIS simulation results indicated congruency effect (sponge and acoustic cavitation) of cutting-edge ultrasound technology on water absorption. The ANFIS simulation results improved the state of art in domain of studying ultrasound-assisted hydration process of wheat. Therefore, the distinguished ANFIS simulation environment is suggested to be served as an effective step towards management of ultrasound-assisted hydration process of wheat in seed priming, flour milling (tempering), making dough, and wet storage processes.