工业过程的预测控制与模糊PID控制的研究
发布时间：2018-12-21 08:31
【摘要】：在控制领域中,比例-积分-微分(PID)控制器是使用最多的一种控制器。因其简单结构、鲁棒性强以及便于实现等优点,所以在工业过程中被广泛使用。但是,由于工业过程变得越来越复杂,导致生产过程中不可避免的出现了时滞、非线性以及不确定性等问题,使得PID控制器越来越难满足需要的控制性能。预测函数控制(PFC)作为预测控制中被广泛使用的算法,所以被广泛应用于石油化工过程中。主要是PFC对模型的精度要求不高,并且具有较高的鲁棒性和跟踪性能。模糊控制是在智能控制算法中经常被使用到的一种算法,目前已经在工业过程控制中被广泛使用。这是因为其不依赖于数学模型,仅仅通过相关经验以及数据就能够很好的控制被控对象,所以在控制领域具有越来越重要的地位。如果能够将PID控制与模糊控制和PFC这两种算法的优点进行结合,将会对工业上的生产效率进行改善。本文通过总结前人优秀的成果,然后做了如下两方面的研究工作:一方面,通过将模糊理论与神经网络的相关性能进行结合,在此基础上提出了模糊网络PID控制器的一种设计方法。利用神经网络本身具有的自学能力与模糊推理能力相结合来对PID参数进行调整,从而提高了PID控制的自适应能力。最后,通过仿真来对模糊网络PID控制性能进行了验证。另一方面,通过结合预测控制、模糊控制和PID控制各自的优点,提出一种预测模糊PID控制器的设计方法。该方法是通过模糊控制达到自适应调整,并引入预测控制的预测模型来完成提前预测。从而在保证了该控制器具有了预测控制的预测能力外,还同时具有模糊控制的推理能力。最后将该控制方法应用到工业中的焦化炉被控对象上,并结合仿真来对该设计方法的有效性进行了验证。
[Abstract]:In the field of control, the proportional-integral-differential (PID) controller is one of the most widely used controllers. Because of its simple structure, strong robustness and easy to implement, it is widely used in industrial processes. However, as the industrial process becomes more and more complex, it is inevitable that there are some problems such as delay, nonlinearity and uncertainty in the production process, which makes the PID controller more and more difficult to meet the required control performance. Predictive function control (PFC) is widely used as an algorithm in predictive control, so it is widely used in petrochemical process. The main problem is that PFC has low precision and high robustness and tracking performance. Fuzzy control is often used in intelligent control algorithms and has been widely used in industrial process control. This is because it does not rely on the mathematical model, only through the relevant experience and data can control the controlled object very well, so it has more and more important position in the control field. If the advantages of PID control, fuzzy control and PFC can be combined, the industrial production efficiency will be improved. This paper summarizes the outstanding achievements of the predecessors, and then does the following two aspects of research: on the one hand, by combining the fuzzy theory with the correlation of neural networks, On this basis, a design method of fuzzy network PID controller is proposed. The self-learning ability of neural network and fuzzy reasoning ability are combined to adjust the PID parameters, thus improving the adaptive ability of PID control. Finally, the performance of fuzzy network PID control is verified by simulation. On the other hand, combining the advantages of predictive control, fuzzy control and PID control, a design method of predictive fuzzy PID controller is proposed. In this method, adaptive adjustment is achieved by fuzzy control, and the predictive model of predictive control is introduced to complete the prediction in advance. Thus, the controller has the predictive ability of predictive control and the reasoning ability of fuzzy control. Finally, the control method is applied to the controlled object of coking furnace in industry, and the validity of the design method is verified by simulation.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273
[Abstract]:In the field of control, the proportional-integral-differential (PID) controller is one of the most widely used controllers. Because of its simple structure, strong robustness and easy to implement, it is widely used in industrial processes. However, as the industrial process becomes more and more complex, it is inevitable that there are some problems such as delay, nonlinearity and uncertainty in the production process, which makes the PID controller more and more difficult to meet the required control performance. Predictive function control (PFC) is widely used as an algorithm in predictive control, so it is widely used in petrochemical process. The main problem is that PFC has low precision and high robustness and tracking performance. Fuzzy control is often used in intelligent control algorithms and has been widely used in industrial process control. This is because it does not rely on the mathematical model, only through the relevant experience and data can control the controlled object very well, so it has more and more important position in the control field. If the advantages of PID control, fuzzy control and PFC can be combined, the industrial production efficiency will be improved. This paper summarizes the outstanding achievements of the predecessors, and then does the following two aspects of research: on the one hand, by combining the fuzzy theory with the correlation of neural networks, On this basis, a design method of fuzzy network PID controller is proposed. The self-learning ability of neural network and fuzzy reasoning ability are combined to adjust the PID parameters, thus improving the adaptive ability of PID control. Finally, the performance of fuzzy network PID control is verified by simulation. On the other hand, combining the advantages of predictive control, fuzzy control and PID control, a design method of predictive fuzzy PID controller is proposed. In this method, adaptive adjustment is achieved by fuzzy control, and the predictive model of predictive control is introduced to complete the prediction in advance. Thus, the controller has the predictive ability of predictive control and the reasoning ability of fuzzy control. Finally, the control method is applied to the controlled object of coking furnace in industry, and the validity of the design method is verified by simulation.
【学位授予单位】：杭州电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273
【参考文献】
相关期刊论文 前10条
1 李佛W，
本文编号：2388665
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2388665.html
