Self-adaptation nonlinear time varying controller and controlling method thereof

The structure of control unit for the present invention is neural network with local connected recursion as the neural network is composed of input layer, hidden layer and output layer, of which the input layer has S number of neuron input node, the hidden layer has most of three neuron nodes as of...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SHUANG CONG, BEICHEN JI, GUODONG LI
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The structure of control unit for the present invention is neural network with local connected recursion as the neural network is composed of input layer, hidden layer and output layer, of which the input layer has S number of neuron input node, the hidden layer has most of three neuron nodes as of integral, differential and proportional node, the output layer has single neuron node. Total connection to each other is applied between each node of the input and hidden layer and each neuron node at the input layer is used as input end of network controller being connected with comparator of the controlled object with neuron node at output layer as output end being connected to input end of the controlled object. 非线性时变自适应控制器及其控制方法,涉及各行业中被控对象或生产过程的控制技术。控制单元的结构为带有局部连接递归的神经网络;所述神经网络由输入层、隐含层和输出层组成,其中:输入层具有s个神经元输入节点e#-[j],该节点数s为被控对象的可测输出变量数;隐含层具有最多3个神经元节点,分别为积分节点a#-[1]、微分节点a#-[3]和比例节点a#-[2];输出层具有单个神经元节点u;神经网络输入层各节点与隐含层各节点之间为相互全连接,输入层的各神经元节点还作为网络控制器1的输入端与被控对象输出变量Y(k)反馈后的比较器3相连接,输出层的神经元节点u还作为网络控制器1的输出端与被控对象2的输入端相连接。