A prediction method for interior temperature of grain storage via dynamics models: A simulation study

Temperature monitoring in the interior of grain warehouse is one of the critical issues for the quality assurance in grain storage depot. Traditional methods, such as sampling inspection or sensor-based temperature data collection in grain warehouse, are usually time-consuming and ineffective because of the low reliability and precision of sensors imbedded in grain stacks. In this article, we develop an approach to predict the interior temperature of grain in a cube warehouse through combining thermodynamics model and spatial stochastic processes. To characterize the underlying spatial and temporal interaction structures, we integrate a 3-D unsteady heat transfer model into a Gaussian Markov random field for interior temperature estimation. The variations of ambient temperature, solar radiation and heat transmission within the wallboards of the warehouse are also considered in this approach. To validate our approach, a simulation study was conducted for temperature prediction by using real environmental data of a horizontal warehouse. The result based on this simulation study with the physical interpretations bring a deep insight of spatial-temporal variation of the interior temperature, and meanwhile provides a guidance for future sensor placement design and parameter calibration.