Abstract:The 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 Applications The 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.

Equilibrium Moisture Content and Dioxide Carbon Monitoring in Real-Time to Predict the Quality of Corn Grain Stored in Silo Bags using Artificial Neural Networks

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

Monitoring and predicting corn grain quality on the transport and post-harvest operations in storage units using sensors and machine learning models

Sensor-cable-probe and sampler for early detection and prediction of dry matter loss and real-time corn grain quality in transport and storage

Monitoring of carbon dioxide and equilibrium moisture content for early detection of physicochemical and morphological changes in soybeans stored in vertical silos

A Model for Predicting and Grading the Quality of Grain Storage Processes Affected by Microorganisms under Different Environments.

Rapid Assessment of Rice Quality Traits Using Low-Cost Digital Technologies

Application of MOX Sensors to Determine the Emission of Volatile Compounds in Corn Groats as a Function of Vertical Pressure in the Silo and Moisture Content of the Bed

Research of real-time corn yield monitoring system with DNN-based prediction model

A Monitoring Device and Grade Prediction System for Grain Mildew

Modeling grain quality characteristics via dynamic models using sensing data

Convolutional Neural Networks Using Enhanced Radiographs for Real-Time Detection of Sitophilus zeamais in Maize Grain

RF-Based Moisture Content Determination in Rice Using Machine Learning Techniques

Use of Artificial Neural Network Model for Rice Quality Prediction Based on Grain Physical Parameters

Monitoring Grain Silos Instantly with IoT Based Control System

Temperature Forecasting for Stored Grain: A Deep Spatiotemporal Attention Approach

Embedded Yolo-Fastest V2-Based 3D Reconstruction and Size Prediction of Grain Silo-Bag

Effect of the Consolidation Level on Organic Volatile Compound Emissions from Maize during Storage

Evaluation and prediction of the physical properties and quality of Jatobá-do-Cerrado seeds processed and stored in different conditions using machine learning models

Rapid and Non-Destructive Prediction of Moisture Content in Maize Seeds Using Hyperspectral Imaging

Training Machine Learning Algorithms Using Remote Sensing and Topographic Indices for Corn Yield Prediction