Fuzzy logic, artificial neural network and mathematical model for prediction of white mulberry drying kinetics

The thin-layer convective- infrared drying behavior of white mulberry was experimentally studied at infrared power levels of 500, 1000 and 1500W, drying air temperatures of 40, 55 and 70 degrees C and inlet drying air speeds of 0.4, 1 and 1.6m/s. Drying rate raised with the rise of infrared power levels at a distinct air temperature and velocity and thus decreased the drying time. Five mathematical models describing thin-layer drying have been fitted to the drying data. Midlli et al. model could satisfactorily describe the convective-infrared drying of white mulberry fruit with the values of the correlation coefficient (R-2=0.9986) and root mean square error of (RMSE= 0.04795). Artificial neural network (ANN) and fuzzy logic methods was desirably utilized for modeling output parameters (moisture ratio (MR)) regarding input parameters. Results showed that output parameters were more accurately predicted by fuzzy model than by the ANN and mathematical models. Correlation coefficient (R-2) and RMSE generated by the fuzzy model (respectively 0.9996 and 0.01095) were higher than referred values for the ANN model (0.9990 and 0.01988 respectively).