Real‐time equilibrium moisture content monitoring to predict grain quality of corn stored in silo and raffia bags

Abstract

AbstractThe use of silo and raffia bags for the temporary grain storage has been increasing in recent years. However, the methods for monitoring a stored product are limited to visual inspections and sampling. Thus, this research aimed to real‐time equilibrium moisture content monitoring to predict grain quality of corn stored in different conditions in silo and raffia bags using wireless sensor network prototype, Internet of Things (IoT) platform, and neural network algorithms. Experiments were conducted using corn grain with two initial water contents of 13% and 18% (w.b.), three storage environments with temperatures of 30, 23, and 17°C, and two types of packaging, that is, silo and raffia bags, for a 3‐month storage evaluation. During the monitoring of stored grain, variations in equilibrium moisture hygroscopic content were observed, which inferred changes in the corn quality. Water contents of 13% under a storage condition of 17°C showed the highest quality results, whereas storage in silo bags with water contents of 13% and 18% showed no differences at 23°C; however, at a temperature of 30°C, the grain suffered a high level of deterioration. The storage time influenced the reduction of grain quality for all factors. The physicochemical quality prediction results indicated a high coefficient of determination of the trained models, presenting itself as a promising perspective, mainly in developing embedded technologies for monitoring and predicting the qualitative variables of grain stored in silo and raffia bags.Practical ApplicationsThe application of sensor technology and the Internet of Thing (IoT) to monitor the temperature and relative humidity of intergranular air in real time for the determination of equilibrium moisture content became possible to predict the physical and physicochemical quality of grains stored in bag silos and raffia bags using artificial neural networks (ANN) algorithms. The results obtained were satisfactory and can replace the punctual sampling of the grains mass stored in hermetic packages. The application of a set of technologies possible to monitor the grain quality in real time and predict the grain storage time in bag and raffia silos to reduce losses.