注塑机注射速度的零相位滤波迭代学习控制  

Injection Velocity Control in Injection Molding by Iterative Learning Control with Zero-phase Filtering

作　　者：李闯[1] 王福利[1] 董召然[1] 陈鹃[2] 

机构地区：[1]东北大学流程工业综合自动化教育部重点实验室,辽宁沈阳110004 [2]沈阳新松机器人自动化股份有限公司,辽宁沈阳110168

出　　处：《信息与控制》2009年第3期293-296,共4页Information and Control

摘　　要：针对注射速度控制,设计了一种基于零相位滤波迭代学习的前馈反馈控制方法。该控制算法中的前馈部分采用迭代学习控制,利用以往批次信息实现精确跟踪;反馈部分采用比例控制,用于克服当前批次扰动。为了消除学习算法应用时出现的坏学习瞬态,这里先推导出所提控制算法在频域下的收敛条件,再给出基于零相位滤波的单调递减学习瞬态。经实验验证,这种方法可以使学习控制具有单调递减的学习瞬态曲线,并能很好地满足实际的注射控制要求。A feedforward and feedback control method based on iterative learning control （ILC） with zero-phase filtering is designed for injection velocity control. In this control method, ILC is used as the feedforward part to realize accurate trajectory tracking with information of the previous cycle, and proportional control is used as the feedback part to avoid disturbances appearing in the current cycle. In order to overcome bad learning transients in ILC, the convergence condition in frequency domain is derived, and then the implementation method for monotonic decreasing learning transients is described with zero-phase filtering. The experiments prove that the proposed method provides the ILC with a monotonic decreasing learning transient curve and meets the actual needs of injection velocity control.

关 键 词：注射速度 迭代学习控制 零相位滤波 前馈反馈控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]