球磨机系统先进控制算法研究及应用

发布时间：2018-03-16 20:01

本文选题：球磨机系统　切入点：前馈解耦　出处：《中国科学技术大学》2015年硕士论文　论文类型：学位论文

【摘要】：钢球磨煤机(球磨机)系统的稳定、经济运行是提高火电厂经济效益的基础。但球磨机系统是强耦合、大惯性、大滞后、强扰动对象,常规控制算法往往难以奏效。为改善其控制品质、提高其运行经济性,有必要研究球磨机先进控制算法。在分析球磨机系统各控制回路动态特性的基础上,本文设计了一种改进的广义预测解耦控制器,构建了一套球磨机先进控制与优化系统,并成功应用。应用结果表明,先进控制能有效地抑制球磨机系统各控制回路之间的耦合作用,增强系统的抗扰动能力,提高控制精度,降低单位磨煤电耗。本文的研究工作主要包括： (1)提出了一种改进的前馈解耦内模控制器。通过添加时滞补偿环节和反馈通道,可以有效地提高解耦控制器抑制扰动的能力和鲁棒性；理论分析表明此解耦控制器在模型匹配条件下可以实现完全解耦；球磨机系统存在白噪声型干扰量和乘性输出不确定性时的仿真实验结果表明此控制器的解耦效果良好。 (2)对球磨机系统解耦后的单变量回路设计了一种改进的阶梯式广义预测控制器。球磨机系统的大惯性大滞后特性会导致控制量在暂态过程中存在明显的振荡,这会影响控制系统的调节时间且造成执行器频繁动作,磨损增加,因此本文通过引入输出加权因子对输出预测进行调整,仿真结果表明此算法能有效减小控制量的振荡并降低调节时间和超调量。 (3)结合球磨机系统的对象特性,采用前述改进的广义预测解耦控制器,设计实现了一套球磨机先进控制与优化系统。完成了球磨机原DCS的逻辑和组态修改,设计了先进控制投切的判断逻辑、在线优化的方案、堵磨堵煤的判断及其应急处理方案,实现了自动检测、无扰切换、投切判断、实时监控等功能,并具有良好的可靠性和方便的人机交互界面；实际投运效果表明球磨机系统的控制品质得到显著改善,单位磨煤电耗大幅降低。
[Abstract]:The system of ball mill (ball mill) is stable and economical operation is the foundation to improve economic benefit of thermal power plant, but the ball mill system is strong coupling, large inertia, large lag and strong disturbance. In order to improve the control quality and improve the operation economy, it is necessary to study the advanced control algorithm of ball mill. On the basis of analyzing the dynamic characteristics of each control loop of ball mill system, an improved generalized predictive decoupling controller is designed, and a set of advanced control and optimization system of ball mill is constructed and successfully applied. Advanced control can effectively suppress the coupling action between control loops of ball mill system, enhance the anti-disturbance ability of the system, improve the control accuracy and reduce the power consumption of coal per unit mill. The research work of this paper mainly includes:. (1) an improved feedforward decoupling internal model controller is proposed. By adding time delay compensation link and feedback channel, the ability and robustness of decoupling controller to suppress disturbance can be improved effectively. Theoretical analysis shows that the decoupling controller can be decoupled completely under the condition of model matching, and the simulation results show that the decoupling effect of the controller is good when there are white noise disturbance and multiplicative output uncertainty in ball mill system. In this paper, an improved stepped generalized predictive controller is designed for the decoupled single-variable loop of ball mill system. The large inertia and large lag characteristic of ball mill system will lead to obvious oscillation of the control quantity in the transient process. This will affect the adjusting time of the control system and cause the actuator to move frequently and wear more. So this paper adjusts the output prediction by introducing the output weighting factor. The simulation results show that the proposed algorithm can effectively reduce the oscillation of the control quantity and reduce the adjustment time and overshoot. Combined with the object characteristics of ball mill system, a set of advanced control and optimization system of ball mill is designed and realized by using the above-mentioned improved generalized predictive decoupling controller. The logic and configuration modification of the original DCS of ball mill is completed. The judgment logic of advanced control switching, the scheme of on-line optimization, the judgment of blocking and blocking coal and its emergency treatment are designed. The functions of automatic detection, no-disturbance switching, cutting judgment, real-time monitoring and so on are realized. The practical operation results show that the control quality of ball mill system has been improved significantly and the power consumption per unit mill has been greatly reduced.
【学位授予单位】：中国科学技术大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TD453;TP273

【参考文献】