Research on average vertical velocity of rubber particles in vertical screw conveyor based on bp neural network

Incorrect formula of average particle vertical velocity can lead to incorrect derivation of the formula of conveying quantity, thus leading to incorrect design of the vertical screw conveyor. In this paper, the vertical velocity of rubber particles in vertical screw conveyor is analyzed by discrete element simulation, and the vertical velocity distribution of particle flow in vertical screw conveyor with different scale structure is analyzed. In order to reduce the computational burden of the training sample and improve the accuracy of the neural network prediction, the BP neural network method is employed. Firstly, the univariate vertical velocity of different influencing factors in microscale structure is trained and predicted, and then the average vertical velocity involving two important input variables, spiral speed and pipe diameter, is trained to get the predicted values of the average vertical velocity of neural network. Finally, the experimental verification is carried out. In this paper, the BP neural network model of the average vertical velocity of rubber particles is established, which not only provides a method for the establishment of other particle neural network models, but also avoids the repeated use of discrete element simulation, and saves a lot of time.