This paper deals with a neural network (NN) application to an agricultural
fixed-bed dryer. The aim of the study was to set-up a NN in order to determ
ine the relationship between the moisture distribution of the material to b
e dried and the physical parameters of the drying air temperature, humidity
and air flow rate. Input data was randomly changed, while output was gener
ated by O'Callaghan's model based on the input specifically for barley. A s
elected NN structure was used for studying the influence of sampling time,
randomised training, different back-propagation training algorithms and the
number of hidden neurones. It was concluded that the artificial NN could b
e effective for modelling of the grain drying process. (C) 2000 Elsevier Sc
ience B.V. All rights reserved.