Control And Application Of Reflux Zone Based On BP Neural Network In Tobacco Flour Mill

Tobacco industry is one of the largest pillars in Yunnan Province, and the tobacco processing precision and requirement are in continuous improvement. The three phases of tobacco redrying include drying, cooling and resurgence. Drying zone and cooling zone are in the control of traditional PID, but temperature control and humidity control in resurging zone influence each other. With the development of artificial neural network technology, its ability of solving the nonlinear and uncertainty in the industrial process also constantly improves, so BP neural network is used in the resurging zone of tobacco baking machine.This paper first analyzes the problems occurring in humidity and temperature of the resurging zone with manual intervention, studies the principle and algorithm of BP artificial neural network, builds the structure of BP artificial neural network according to the influencing factors and industrial requirements, and then designs specific algorithms. With the support of Jiangchuan Tabacco Redrying Facroty, the writer has obtained the previous sample data needed in the network training, got the necessary network parameters obtained during training the artificial neural network, and finally put the neural network into the use of the tobacco baking machine.