Rosa damascene Mill. moisture modeling in hot air dryer using mathematic models and artificial neural network

Damask rose with scientific name of Rosa damascena Mill. contains essential oils with large medicinal properties. Qualitative and quantitative extraction of essential oils as well as its economic justification depends on appropriate methods of drying. Appropriate method of drying reduces loss and damage during storage and helps maintain product quality. The purpose of this study was to predict Rose moisture content during the drying process with hot air flowing as a function of temperature at four levels (40، 50، 60 and 70°C) and air velocity at three levels (0. 5، 1 and 1. 5 m/s)، using artificial neural networks. The average initial and final moisture contents were calculated to be 78% and 9%، respectively. The drying process was modeled by mathematical models using matlab and then the moisture content graphs were achieved by excel. Then، the drying process was modeled using neural networks with three inputs including temperature، air velocity and time. Results showed that neural network was more accurate than mathematic models in modeling and predicting the drying process of damask rose and could be used in on-line controlling.