基于神经网络的自适应Dahlin数字控制器  

Adaptive Dahlin Digital Controller Based on Neural Network

作　　者：曲永印[1] 邵世煌[1] 段慧达[2] 王忠礼[2] 周振雄[2] 

机构地区：[1]东华大学信息工程学院,上海200051 [2]北华大学电气信息工程学院,吉林省吉林市132013

出　　处：《控制与决策》2006年第3期327-330,共4页Control and Decision

摘　　要：在常规Dahlin数字控制器的基础上,采用神经网络理论确定Dahlin控制器参数,提出了一种基于神经网络的自适应Dahlin数字控制器.该控制器具有收敛速度快、鲁棒性强的特点,既可用于参数时变和时滞未知的受控对象,亦适用于逆不稳定过程,不存在控制器的振铃现象,是解决具有慢时变和时滞未知工业过程控制的有效方法.An adaptive Dahlin digital controller based on neural network is presented. It is designed based on common Dahlin digital controller. Parameter of the controller is defined by using neural network method. It has features for fasting convergence and having good robust. It can be applied not only to the plant which is unknown or has slow time-varying and time-delay, but also to the plant which has unstable inverse process. It can als~ eliminate the ringing phenomena of controller. It is an effective method to solve the problem of controlling an industrial process with slow time-varying and time delayed. The simulation result shows the effectiveness of the proposed method.

关 键 词：DAHLIN算法 自适应控制 神经网络 逆不稳定 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]