Modeling grain quality characteristics via dynamic models using sensing data

Grain storage is a critical issue in the national economy and greatly affects the livelihood of the people, and inefficient management of granaries will downgrade the quality of stored grains. To achieve a quantitative evaluation of grain quality during its storage period, we focus on two of the important metrics (temperature and humidity), and fully use the data collected from sensor networks in granaries. Specifically, we develop an approach for modeling grain quality characteristics by incorporating the estimated temperature field with humidity data in the granary, and we characterize the grain quality via dynamic models. A real case study has been conducted to validate the performance of the proposed approach. Results indicate a deep insight into grain quality characteristics and provide a reference for developing grain quality inspection and maintenance strategies.