Research On Optimization Of Re-drying Control System Based On Data Mining And Predictive Auto-coupled PID

Re-drying production is the key treatment process of tobacco leaves in the re-drying process,which has the characteristics of complex system,such as large time delay,strong coupling,multi-interference and so on.In the face of the higher requirements for the quality of re-baked products put forward by cigarette factories,the traditional production mode and re-drying control system have been difficult to meet the characteristics,rapidity,stability,accuracy and other production requirements of tobacco processing.Therefore,the research on the optimization of re-drying control system is very urgent and important.Aiming at the optimization of re-drying control system,this paper mainly does the following two aspects of work:The main results are as follows:1)in the aspect of process parameter optimization of redrying system,based on the historical production data of tobacco processing enterprises,the relationship among process parameters,environmental factors and product quality indexes of tobacco leaves from different producing areas and grades was explored.Regression modeling was carried out by using data mining algorithms such as decision tree and random forest,and the importance order of process parameters was obtained,based on which the orthogonal experiment scheme was designed.The optimal parameter combination of the processing process was obtained by combining the prediction results of the regression model and the orthogonal test results.When the parameter combination is applied to the field production test,it is found that the parameter combination can well meet the production requirements in terms of accuracy.in the same processing environment,the time for stably reaching the advanced production index is 55%less than that of the factory's existing parameter combination.In the aspect of the optimal control of the re-drying process,the object mechanism characteristics of the temperature and moisture in the drying zone of the re-drying machine are described,and the corresponding mathematical model is established.according to the actual engineering requirements of the re-drying plant,a predictive auto-coupled PID controller is designed to control the temperature and moisture control object of the re-drying machine,and the simulation verification is carried out by using Matlab/Simulink to build a model.Then the predictive auto-coupled PID controller is applied to the field test in the production line of the re-baking plant.The results show that under the action of the controller,the absolute deviation between the actual value and the set value of moisture content at the outlet of tobacco leaves is 0.17%,and the relative deviation is reduced to 1.3%.Compared with the national tobacco export standard,the deviation range is reduced by 56%.It shows that the controller has a good control effect on the control objects with large time delay and strong coupling,such as temperature and moisture,and can be applied to the actual re-drying production.