Research And Development Of Cement Vertical Mill Control System Based On Fractional PID

Cement grinding is the last step before the cement leaves the factory.The stability of the grinding process directly affects the quality and performance of the finished cement.In order to improve grinding efficiency,the grinding equipment has been developed from single ball mill to cement combined grinding and vertical grinding.However,the cement vertical mill is a complex control system with many variables,strong coupling,large time delay,multi-disturbance and difficult modeling,which makes the automatic control of cement vertical mill extremely hard.Therefore,for the purpose of quality assurance,production increase and energy conservation,a control system software platform is established to realize stable operation of the cement vertical grinding on the basis of modeling and control of the key links of the cement vertical grinding process.The main research work is as follows:I.Based on the research background of a cement mill production line in Guangxi,this paper analyzes the grinding process and its working characteristics,and establishes the dynamic relationship of the key links of the vertical mill based on the data by methods of the least square recursion and particle swarm optimization.It includes: 1)Based on the abnormal working conditions and data analysis,the key factors affecting the operation of the vertical mill and cement quality are determined,namely,the pressure difference of the mill and the cement content less than 45 ?m;2)Low order difference method and mean filtering are used to reduce the influence of gross value and data noise on dynamic relationship fitting;3)Through the analysis of the dynamic characteristics such as lag and inertia between the pressure difference and the speed of the plate feeder,the model structure of the pressure difference with n-order is given.The particle swarm optimization algorithm is used to identify the gain,inertia coefficient and lag time under different order of the model.Three evaluation indexes,namely root mean square error,average absolute error and average relative error,are used to determine the best order and corresponding model parameters;4)Based on the analysis of the dynamic relationship between the cement content of less than 45?m and the rotation speed of powder concentrator,a transfer function model with second-order plus dead time of the cement content less than 45?m and the rotation speed of powder concentrator is proposed.According to the spatial distance between the separator and the particle analyzer,the lag time range is determined.The model gain and inertia coefficient under different lag time are fitted by the least-squares recursive algorithm,and the root mean square error evaluation index is used to determine the best fitting parameters.II.In order to complete the automatic control of the pressure difference and the cement content of less than 45?m,a vertical mill control method with fractional order PID based on the Bode ideal transfer function and amplitude-phase margin is proposed based on the transfer function established above.It includes: 1)the controller parameter setting method based on Bode ideal transfer function is used to design the fractional order PID controller of pressure difference in order to obtain the good robustness to the gain change;2)The amplitude and phase margin method is used to design fractional order PID controller for cement content with a large lag time of less than 45?m.Simulation results verify the effectiveness of the proposed method.III.Using VB programming language,OPC technology and SQL database,the software of cement vertical mill process monitoring and automatic control guidance decision system is developed.Specifically including: 1)According to the on-site communication architecture,database is designed to realize data interaction;2)The structure and function of the software are introduced,the interface is designed,and the software development process is described by combining the bang-bang control method and the proposed method;3)Based on the actual operation data,the effect of manual and automatic control is compared from the aspects of dynamic process,product quality and economic benefit indexes,such as output,power consumption,fluctuation of intermediate hopper weight,cement particle distribution and specific surface area.It is proved that the proposed control system can achieve an increase in output,a decrease in power consumption of main equipment and a stable quality of cement.