Optimum operating conditions for heat and mass transfer in foodstuffs drying by means of neural network inverse

Artificial neural network inverse (ANNi) is applied to optimize the operating conditions on heat and mass transfer during foodstuffs drying. This proposed method (ANNi) inverts the artificial neural network (ANN) and uses the Nelder-Mead simplex method of optimization to find the optimum parameter value (or unknown parameter) for given required conditions. In the aim to demonstrate this ANNi method, two separate feedforward networks (ANN) with one hidden layer reported by Hernández-Pérez, García-Alvarado, Trystram, and Heyd [Hernández-Pérez, J.A., García-Alvarado, M.A., Trystram, G., & Heyd, B. (2004). Neural networks for the heat and mass transfer prediction during drying of cassava and mango. Innovative Food Science and Emerging Technologies, 5, 56-64], were used in order to obtain temperature and moisture kinetics simulations during the drying of mango and cassava. These reported models take into account air temperature, air velocity, shrinkage as a function of moisture content, time and air humidity as well-known input parameters. Levenberg-Marquardt learning algorithm, hyperbolic tangent sigmoid transfer-function, linear transfer-function and three neurons in the hidden layer were considered in both reported models. Results of the ANNi showed a good agreement with the experimental and simulated data (error<0.001%). Then ANNi could be applied to determine the optimal parameters during mango and cassava drying with elapsed time minor to 0.3 s. In addition, this methodology can be used to controlling the drying process.