Glasnik Zaštite Bilja, Vol. 48. No. 4., 2025.
Professional paper
https://doi.org/10.31727/gzb.48.4.10
Application of artificial neural networks in the modelling of grain moisture and drying parameters
Ivan Brandić
; Agronomski fakultet, Sveučilište u Zagrebu, Zagreb, Hrvatska
*
Karlo Pleskalt
; Agronomski fakultet, Sveučilište u Zagrebu, Zagreb, Hrvatska
Karlo Špelić
; Agronomski fakultet, Sveučilište u Zagrebu, Zagreb, Hrvatska
Ana Matin
; Agronomski fakultet, Sveučilište u Zagrebu, Zagreb, Hrvatska
* Corresponding author.
Abstract
Cereals play an important role in human nutrition, animal feed production and industry. After harvesting, the quality and shelf life of the grain is maintained by drying, which reduces the moisture to a level that prevents the development of microorganisms and mould. In practise, various drying methods are used, with each technique having its own advantages in terms of uniformity, speed and preservation of grain quality. The optimum moisture content for grain storage is between 12%
and 14%, depending on the variety. Low temperatures and controlled humidity during storage are key to preserving nutritional value and preventing spoilage. Modern innovations in agriculture, in particular the use of machine learning and artificial neural networks, enable precise modelling and optimisation of the drying process. These models analyse large amounts of data, recognise patterns and predict optimal operating conditions, thus contributing to greater efficiency, quality preservation and lower drying costs.
Keywords
drying, cereals, modelling, artificial neural networks
Hrčak ID:
333917
URI
Publication date:
22.7.2025.
Visits: 192 *