工业注塑过程的H_∞迭代学习控制器设计  

H+_∞ ILC Design for Industrial Injection Molding Process

作　　者：闫帅可 卜旭辉[1] YAN Shuai-ke;BU Xu-hui(School of Electrical Engineering ＆ Automation,Henan Polytechnic University,Jiaozuo 454000,China)

机构地区：[1]河南理工大学电气工程与自动化学院

出　　处：《控制工程》2018年第9期1610-1615,共6页Control Engineering of China

基　　金：国家自然科学基金项目(61203065,61573129);河南省高校科技创新人才支持计划(16HASTIT046);河南省高等学校青年骨干教师计划项目(2014GGJS-041);河南省高校基本科研业务费专项资金资助;河南理工大学青年骨干教师资助计划项目

摘　　要：注塑成型是一种典型的批次生产过程,保压压力是保证其产品质量的关键因素。已有注塑机保压压力迭代学习控制方法,是在数据传输完美情况下给出的。但在实际网络运行环境下,由于带宽的限制,数据在控制器和控制对象之间传输常发生网络拥堵,造成数据包丢失。因此,针对网络传输过程中数据丢失问题,提出了一种H∞迭代学习控制器的设计方法。首先,将数据丢失现象描述为随机的伯努利序列,在此基础上将模型转化为随机2D-Reosser模型。此时,迭代学习控制器的设计问题可以转化为一类随机2D-Roesser系统的H∞控制问题。最后,仿真结果验证所提方法的有效性。Injection molding is one of the typical batch production processes. And holding pressure is the most important factor to determine the quality of the product. So far ILC has been applied to pressure maintaining control with the assumption of perfect transmission of the system signal. However, there exists a phenomenon of packet loss between the controller and plant because network congestion often occurs in the actual network due to limited bandwidth. Hence one method to design H∞ iterative learning controller will be proposed for the question of pressure maintaining control process with data dropouts. Firstly, the above ILC control system can be transformed into 2D-Reosser model with the phenomenon of data loss modeled by Bernoulli random sequence. At this point, the design problem of H∞ iterative learning controller is converted to the H∞ control problem of a class of random 2D-Roesser system. Finally, the validity of the method proposed is verified by a numerical simulation.

关 键 词：注塑成型 迭代学习控制 数据丢失 H∞控制 2D-Roesser模型 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]