An adaptive neuro-fuzzy inference system for predicting the parameter of dryer system for shelled pistachios

Abstract

This paper presents an Adaptive Neuro-Fuzzy Inference System (ANFIS) model for predicting the parameter of dryer system for shelled pistachios. The most important parameter of drying system is drying time because a good drying for good quality product is provided with moisture content 5-7%. Effects of temperatures and air velocities on drying characteristics of SPs with initial moisture content (MC) of 30±0.2% (wet basis (wb)) were studied in three air velocities (i.e., 0.065, 0.1 and 0.13 m/s) and three air temperatures (i.e., 60, 80 and 100 °C). A number of experiments were performed in order to obtain data for sample weight versus drying time (DT). In order to train the ANFIS, experimental measurements were used as training data and test data. The most suitable membership function and number of membership function were found as Gauss2mf-3. For this number level, after the training, it was found that Root-mean square (RMS) value is 0.0868, and absolute fraction of variance (R2) value is 1.00. When these values are compared with ANN model, it seen that values of ANFIS model are better. © Sila Science.