基于PLC的智能PID方法研究与实现

发布时间：2018-03-26 15:29

本文选题：PLC　切入点：模糊控制　出处：《西安理工大学》2017年硕士论文

【摘要】：PLC作为目前自动化领域的主流控制器具有强大的技术优势,随着工业的不断发展,控制对象越来越复杂,常规PID控制往往由于参数整定不良而使控制系统的性能欠佳;将智能控制与常规PID控制相结合所形成的智能PID控制器,显示出强大的优势;同时,将智能PID控制算法嵌入PLC中也成为当今研究的热点。本文主要工作如下:(1)分析三容水箱液位控制系统实验装置的结构、工艺流程以及CompactLogix系列PLC的硬件与软件系统;结合串级PID控制的基本结构和特点,在CompactLogix上设计并实现了液位串级PID控制。(2)研究模糊控制系统的组成和模糊控制器的设计原理,结合常规PID控制算法,设计了模糊自适应PID控制器;借助MATLAB软件中模糊逻辑工具箱提供的图形界面(GUI)工具,建立离线控制表,并将生成的控制表存储在PLC的标签中;然后通过RSLogix5000编程软件实现输入量模糊化,根据得到的模糊量进行在线查表运算,最终实现PID参数的自适应调整;随后对模糊自适应PID算法进行改进,使得误差和误差变化的加权系数均可调,以适应不同控制过程的控制要求。(3)研究神经网络控制系统的结构和原理,利用RBF网络对被控对象进行在线辨识,BP网络根据系统运行状态及RBF网络提供的信息,通过自身权系数的调整,实现PID参数的在线自适应调整;针对BP网络存在收敛速度慢的问题,本文在算法中引入动量项进行改进;根据RSLogix5000编程软件的特点,在CompactLogix系列PLC上设计并实现了神经网络自适应PID控制算法。(4)设计了基于Factory Talk View Studio的三容水箱液位监控界面,具有显示系统当前运行状况和历史数据,并提供用于控制的操作界面;利用三容水箱液位监控界面对实验装置进行试验,验证了智能PID控制较常规PID控制具有更好鲁棒性和更好的动静态性能。
[Abstract]:As the mainstream controller in the automation field, PLC has strong technical advantages. With the development of industry, the control object is becoming more and more complex. The performance of the control system is often poor because of poor parameter setting in conventional PID control. The intelligent PID controller, which combines intelligent control with conventional PID control, shows powerful advantages, at the same time, Embedding intelligent PID control algorithm into PLC has become a hot research topic. The main work of this paper is as follows: 1) analyzing the structure, process flow and hardware and software system of CompactLogix series PLC. Combined with the basic structure and characteristics of cascade PID control, the composition of fuzzy control system and the design principle of fuzzy controller are studied on CompactLogix. (2) the composition of fuzzy control system and the design principle of fuzzy controller are studied. Combined with the conventional PID control algorithm, the structure of fuzzy control system and the design principle of fuzzy controller are studied. The fuzzy adaptive PID controller is designed, and the off-line control table is established with the help of the graphical interface tool provided by the fuzzy logic toolbox in MATLAB software, and the generated control table is stored in the label of PLC. Then the fuzzy input is fuzzied by RSLogix5000 programming software, and the fuzzy quantity is looked up on line. Finally, the adaptive adjustment of PID parameters is realized, and then the fuzzy adaptive PID algorithm is improved, and the fuzzy adaptive PID algorithm is improved. So that the error and the weighting coefficient of error change can be adjusted to meet the control requirements of different control processes.) the structure and principle of neural network control system are studied. The RBF network is used to identify the controlled objects on line. According to the system running state and the information provided by RBF network, the on-line adaptive adjustment of PID parameters is realized by adjusting the weight coefficient of the network. Aiming at the problem of slow convergence of BP neural network, the momentum term is introduced into the algorithm to improve the algorithm, and according to the characteristics of RSLogix5000 programming software, This paper designs and implements a neural network adaptive PID control algorithm on CompactLogix series PLC. It designs a three-tank liquid level monitoring interface based on Factory Talk View Studio, which can display the current running status and historical data of the system, and provide an operating interface for the control. The experimental device is tested by using the liquid level monitoring interface of the three-tank. It is proved that the intelligent PID control has better robustness and better dynamic and static performance than the conventional PID control.
【学位授予单位】：西安理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP273

【参考文献】