The Development Of Cement Pre-grinding Process Control System

The cement pre-grinding process plays an important role in cement combined grinding.The stable operation of the cement pre-grinding process can reduce the production pressure in the subsequent final grinding stage,thereby improving the production efficiency of the entire production line.There are many devices and parameters involved in the cement pre-grinding process.The coupling between parameters is strong,hysteresis and nonlinearity are serious,which increases the difficulty of control.At present,most of the control methods of the grinding process still remain in the manual control based on the DCS system.This production method consumes a lot of manpower,and due to the uneven technical level of the operator,the production and quality of the cement-made process products are extremely unstable,and the production efficiency of the production line is not high.Based on the in-depth analysis of the process mechanism of cement pre-grinding process,the coupling relationship between parameters is clarified,and the Fuzzy Cognitive Map(FCM)is established to describe the relationship between cement pre-grinding process parameters.An expert knowledge base was established by summarizing the operational experience of the excellent operators on site.The controller was designed with the Bang_Bang control thought.Based on the above results,the cement pre-grinding process control system was designed and applied in the field.Funded by the Key Technology Research and Application Demonstration Project of Intelligent Factory(Shandong Science and Technology Major Project: 2015ZDXX0101F01),the work of this paper is as follows:(A)The process mechanism of cement pre-grinding process was deeply analyzed,and the key parameters of the control system were determined,including: feeding amount,weight of steady flow silo,grinding hoist current,rolling machine flashboard opening degree,rolling machine current,circulating fan speed,circulation fan valve opening degree,and ball mill load.(B)The structure of the FCM model is determined based on the coupling relationship between the parameters.According to the Least Squares Techniques based Fuzzy Cognitive Map Learning Method(LS_FCM),the correlation parameters between the nodes are calculated,and the FCM model of cement pre-grinding process is established based on the determined structure and parameters.The model parameters are optimized based on the historical data of the field production.Through the analysis of the predicted values of the model,the validity of using the FCM model to predict the parameters of the cement pre-grinding process is verified.(C)Combining the operation experience of excellent operators on site,establishing the expert knowledge base and the reasoning mechanism,and designing the expert system;the Bang_Bang controller was designed based on the mechanism model;the cement pre-grinding process control system was designed based on the mechanism model prediction,expert system and Bang_Bang controller.(D)Based on SQL Server,the system supporting database was established,and the data interaction between the system and the on-site DCS system was realized by using OPC technology.Based on the Visual Studio,the cement pre-grinding process control system was developed by C#;the actual operation of the system verified the effectiveness of the system.