In the paper, the three-dimensional numerical model and the data interface between database and calculating program are established for the prediction of wheat temperature variation during the practical grain storage. Due to the temperature-dependence of thermal properties of wheat, the nonlinear transient finite element simulation of the wheat temperature field is carried out. The numerical results are compared with the measured results to verify the proposed numerical model, and it’s found that there is a good agreement between them. So, it can be said that the temperature variation for the practical wheat storage can be observed and analyzed in the visual type, instead of time and money-consuming experiment, if the numerical method developed in this paper can be put into practice. Furthermore, the present method of monitoring grain temperature only based on the data analysis can be further improved to the visual level.