Research on Intelligent Control of Moisture regain Machine

Clarify the impact of temperature and humidity on the moisture change during the loose moisture regain process. Method: K-means clustering analysis was used to divide the environmental temperature and humidity intervals in Zhangjiakou City. Gradient boosting decision trees were used to screen process parameters related to the change in loose moisture content as feature variables. A machine learning prediction model for the change in moisture content in loose moisture processes under different temperature and humidity intervals was constructed. At the same time, the influence of workshop temperature and humidity on moisture content was explored, and the final model was optimized. After using workshop temperature and humidity as characteristic variables, the prediction accuracy of the SVR and RF models for the moisture change in the loose rehydration process increased, with a decrease of 25% and 46% in R ² (the difference between R ² and 1), respectively. Conclusion: The change in moisture content during loose rehydration is influenced by external factors and workshop temperature and humidity. The moderate temperature and low humidity RF model and high temperature and high humidity SVR model constructed based on different temperature and humidity ranges have a good fitting effect on the moisture change, accurate prediction, and can be reliably applied to the prediction of moisture change in loose rehydration processes. They can also be extended to the prediction and control of moisture content in paper cigarette packaging materials, which has important theoretical significance and practical application value for improving the quality of tobacco rolling and packaging.