Development Of Predictive Control System Of Cement Combine Grinding Particle Size

According to the data of the National Bureau of statistics of China,the national cement output in 2019 is about 2.33 billion tons,an increase of 6% year-on-year.It has been the first in the world for 35 consecutive years,and the industry profit has maintained a good level.Under the background of supply side reform,environmental protection and peak shifting production in the cement industry,the capacity utilization rate of domestic cement is currently maintained at about 60%.Therefore,it is of great significance to improve the quality and output of cement by improving the automation level of cement production process.In this paper,the cement particle size,an important index in the process of cement grinding,is taken as the research object.Using the on-line laser particle size analyzer as the detection means,combined with the mechanism of the combined grinding process,the research and development of the predictive control system for the particle size of the combined grinding of cement are carried out.The main research contents are as follows:(1)According to the actual situation of a cement production enterprise in Luoyang,based on the study of the mechanism of the on-site combined grinding process,combined with the analysis of historical data and the operation experience of on-site workers,Taking the cement particle size content less than 45 um as the model output and the speed of the powder concentrator as the model input,the mathematical model of cement particle size is established by linear regression and Forgetting Factor Recursive Least Square(FFRLS).The simulation results show that the model of cement particle size established by Forgetting Factor Recursive Least Square(FFRLS)has better consistency with the dynamic change of cement particle size.Therefore,the model of cement particle size established by Forgetting Factor Recursive Least Square(FFRLS)is selected as the mathematical model of the predictive control system of cement combined grinding particle size,which lays a foundation for the follow-up research of cement particle size control algorithm.(2)For cement particle size control,the traditional control algorithm is still PID control,because cement grinding is a non-linear,strong coupling and large lag industrial process,the traditional PID control effect is general.GPC has strong robustness and low requirement for model parameters,so this paper designs PI + GPC controller by combining the advantages of PI controller;Due to the general predictive control assumption that the controlled process is unconstrained,But there are a lot of constraints in the particle size control of cement combined grinding,such as the speed constraint of powder concentrator,the speed increment constraint of powder concentrator and the particle size constraint of cement,in order to solve this problem,the improved particle swarm optimization algorithm(PSO)is introduced;The improved PSO is applied to the rolling optimization of implicit GPC,and a PI+GPC controller based on the improved PSO is designed,the simulation results verify the effectiveness of the method.(3)Based on the mixed compilation of C# and MATLAB,combined with the research content of this paper,adding the Bang-Bang control,the predictive control system of cement grinding particle size is developed,which realizes the control of cement particle size,and carries out engineering application,and achieves good operation results.