联合粉磨系统的模糊预测控制
发布时间：2019-06-15 03:57
【摘要】：众所周知,水泥在我国的生产行业中有着举足轻重的作用,联合粉磨系统作为水泥生产过程中重要的组成部分一直是国内外学者研究的热点。联合粉磨系统控制策略的研究也在不断进行,智能的控制方法不仅有利于提高水泥生产过程中智能控制的水平,而且对水泥产量和质量的提高也有很大地帮助。联合粉磨系统是一个非线性的系统,具有强耦合、滞后的特性,控制过程比较复杂。因预测控制具有预测模型、滚动优化、反馈校正的优点,适合各种工业现场。与此同时,模糊控制对模型精度要求不高的特点,也引起了广大学者的重视。而模糊预测控制集合了两者的优点,近年来发展更为迅速。针对这个情况,本文从水泥生产的工业现场着手,以联合粉磨系统为研究对象,对其进行了模糊预测控制的研究。本文依附某水泥厂联合粉磨系统生产1#线,选取合适的联合粉磨系统模型,采用喂料量、选粉机转速作为粉磨系统的输入,水泥磨机负荷流量、水泥粉磨生产合格品流量、水泥粉磨不合格品流量作为系统的输出。设计了联合粉磨系统模糊预测控制器,并开发了联合粉磨系统模糊预测控制软件来实现其运行。本课题的主要研究内容如下:1)通过分析水泥联合粉磨系统的生产工艺和机理研究,结合联合粉磨系统控制的要点和难点,确定联合粉磨系统的关键变量,选取合适的系统模型。提出联合粉磨系统模糊预测控制的具体方案。2)针对联合粉磨系统模糊预测控制的特点,对系统首先采用预测控制算法。对联合粉磨系统分别采用了线性参数时变(Linear Parameter Varying,LPV)预测控制、动态矩阵(Dynamic Matrix Control,DMC)预测控制、广义预测控制(Generalized Predictive Control,GPC)三种预测方法。对联合粉磨系统采用LPV预测控制引入了Lyapunov函数,采用线性矩阵不等式(Linear Matrix Inequality,LMI)的形式,将系统的控制问题转化为相应的凸优化问题,证明了设计的系统具有渐进稳定性;在采用动态矩阵预测控制时,对比了设置不同的磨机负荷流量期望值的仿真效果,并进行了结果分析。采用广义预测控制时,对联合粉磨系统进行了模型的转换,并进行了运行效果的仿真。章节最后对联合粉磨系统的三种预测控制方法进行了比较和总结。3)针对联合粉磨系统模糊预测控制的特点,对联合粉磨系统在预测控制基础上引入模糊控制算法,设计了模糊预测控制器,进行了仿真验证。4)通过VB 6.0语言和相应的软件完成了联合粉磨系统模糊预测控制软件的开发,可通过OPC接口与DCS实现对联合粉磨系统现场控制。本文对联合粉磨系统模糊预测控制进行了总结和展望,在联合粉磨系统的控制过程中有所创新,提出了新的控制算法。
[Abstract]:As we all know, cement plays an important role in the production industry of our country. As an important part of cement production process, combined grinding system has always been the focus of research by scholars at home and abroad. The research on the control strategy of the combined grinding system is also going on. The intelligent control method is not only beneficial to improve the level of intelligent control in the cement production process, but also helpful to the improvement of cement output and quality. The combined grinding system is a nonlinear system, which has the characteristics of strong coupling and lag, and the control process is complex. Because predictive control has the advantages of predictive model, rolling optimization and feedback correction, it is suitable for all kinds of industrial sites. At the same time, the low requirement of fuzzy control for model accuracy has also attracted the attention of many scholars. Fuzzy predictive control combines the advantages of the two and develops more rapidly in recent years. In view of this situation, this paper starts from the industrial site of cement production, takes the combined grinding system as the research object, and carries on the fuzzy predictive control research to it. In this paper, the 1 # line is produced by the combined grinding system of a cement plant, and the suitable model of the combined grinding system is selected. The feed quantity and the speed of the separator are used as the input of the grinding system, the load flow rate of the cement mill, the flow rate of qualified products produced by cement grinding, and the unqualified flow rate of cement grinding as the output of the system. The fuzzy predictive controller of the combined grinding system is designed, and the fuzzy predictive control software of the combined grinding system is developed to realize its operation. The main research contents of this paper are as follows: 1) by analyzing the production technology and mechanism of cement combined grinding system, combined with the key points and difficulties of combined grinding system control, the key variables of combined grinding system are determined, and the appropriate system model is selected. The specific scheme of fuzzy predictive control for joint grinding system is put forward. 2) according to the characteristics of fuzzy predictive control for combined grinding system, the predictive control algorithm is used for the system. Three predictive methods, linear parameter time-varying (Linear Parameter Varying,LPV (predictive control), dynamic matrix (Dynamic Matrix Control,DMC (predictive control) and generalized predictive control (Generalized Predictive Control,GPC), are used for the combined grinding system. The Lyapunov function is introduced into the combined grinding system by using LPV predictive control, and the control problem of the system is transformed into the corresponding convex optimization problem in the form of linear matrix inequality (Linear Matrix Inequality,LMI). It is proved that the designed system has progressive stability. When the dynamic matrix predictive control is used, the simulation results of setting different expected values of mill load flow rate are compared, and the results are analyzed. When the generalized predictive control is used, the model of the combined grinding system is transformed, and the running effect is simulated. At the end of the chapter, three predictive control methods of the combined grinding system are compared and summarized. 3) according to the characteristics of the fuzzy predictive control of the combined grinding system, the fuzzy control algorithm is introduced into the combined grinding system on the basis of predictive control, and the fuzzy predictive controller is designed and verified by simulation. 4) the development of the fuzzy predictive control software of the combined grinding system is completed by VB 6.0 language and the corresponding software. The field control of the combined grinding system can be realized by OPC interface and DCS. In this paper, the fuzzy predictive control of combined grinding system is summarized and prospected, some innovations are made in the control process of combined grinding system, and a new control algorithm is put forward.
【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ172.632
本文编号：2499948
[Abstract]:As we all know, cement plays an important role in the production industry of our country. As an important part of cement production process, combined grinding system has always been the focus of research by scholars at home and abroad. The research on the control strategy of the combined grinding system is also going on. The intelligent control method is not only beneficial to improve the level of intelligent control in the cement production process, but also helpful to the improvement of cement output and quality. The combined grinding system is a nonlinear system, which has the characteristics of strong coupling and lag, and the control process is complex. Because predictive control has the advantages of predictive model, rolling optimization and feedback correction, it is suitable for all kinds of industrial sites. At the same time, the low requirement of fuzzy control for model accuracy has also attracted the attention of many scholars. Fuzzy predictive control combines the advantages of the two and develops more rapidly in recent years. In view of this situation, this paper starts from the industrial site of cement production, takes the combined grinding system as the research object, and carries on the fuzzy predictive control research to it. In this paper, the 1 # line is produced by the combined grinding system of a cement plant, and the suitable model of the combined grinding system is selected. The feed quantity and the speed of the separator are used as the input of the grinding system, the load flow rate of the cement mill, the flow rate of qualified products produced by cement grinding, and the unqualified flow rate of cement grinding as the output of the system. The fuzzy predictive controller of the combined grinding system is designed, and the fuzzy predictive control software of the combined grinding system is developed to realize its operation. The main research contents of this paper are as follows: 1) by analyzing the production technology and mechanism of cement combined grinding system, combined with the key points and difficulties of combined grinding system control, the key variables of combined grinding system are determined, and the appropriate system model is selected. The specific scheme of fuzzy predictive control for joint grinding system is put forward. 2) according to the characteristics of fuzzy predictive control for combined grinding system, the predictive control algorithm is used for the system. Three predictive methods, linear parameter time-varying (Linear Parameter Varying,LPV (predictive control), dynamic matrix (Dynamic Matrix Control,DMC (predictive control) and generalized predictive control (Generalized Predictive Control,GPC), are used for the combined grinding system. The Lyapunov function is introduced into the combined grinding system by using LPV predictive control, and the control problem of the system is transformed into the corresponding convex optimization problem in the form of linear matrix inequality (Linear Matrix Inequality,LMI). It is proved that the designed system has progressive stability. When the dynamic matrix predictive control is used, the simulation results of setting different expected values of mill load flow rate are compared, and the results are analyzed. When the generalized predictive control is used, the model of the combined grinding system is transformed, and the running effect is simulated. At the end of the chapter, three predictive control methods of the combined grinding system are compared and summarized. 3) according to the characteristics of the fuzzy predictive control of the combined grinding system, the fuzzy control algorithm is introduced into the combined grinding system on the basis of predictive control, and the fuzzy predictive controller is designed and verified by simulation. 4) the development of the fuzzy predictive control software of the combined grinding system is completed by VB 6.0 language and the corresponding software. The field control of the combined grinding system can be realized by OPC interface and DCS. In this paper, the fuzzy predictive control of combined grinding system is summarized and prospected, some innovations are made in the control process of combined grinding system, and a new control algorithm is put forward.
【学位授予单位】：济南大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TQ172.632
【参考文献】
中国期刊全文数据库 前1条
1 赵敏;;基于离线状态观测器的LPV系统预测控制[J];控制工程;2013年06期
，本文编号：2499948
本文链接：https://www.wllwen.com/kejilunwen/huaxuehuagong/2499948.html
