Abstract: The precise mathematical model of the industrialized grain drying process is the prerequisite for realizing the dynamic tracking and closed-loop control of the process. Based on the 5 HNH-15 continuous grain dryer designed by this laboratory, an 8-11-1 BP neural network prediction model is constructed. The input of the model is the 8 drying influencing factors of the 5 HNH-15 continuous dryer, and the output of the model It is the moisture content of the exported grain. Using matlab software to establish and verify the BP neural network model, the results show that: after 67 iterations, the mean square error MSE of the constructed BP neural network model reaches 2.8361 e-6, the absolute error is less than ±0.1, and the average absolute error MAE= 0.0288, the relative error is less than 1.2%, the regression coefficient R=0.99996, and the coefficient of determination R~2 =0.9998. A new set of verification tests was added, and the results showed that the absolute error between the predicted value of the model and the actual value was less than ±0.1, the average absolute error was MAE=0.0121, and the relative error was less than 1.1%. The intelligent control of industrialized grain drying provides theoretical basis and technical support.