冶金自动化
冶金自動化
야금자동화
METALLURGICAL INDUSTRY AUTOMATION
2012年
4期
16-19
,共4页
于蒙%邹志云%赵丹丹%郭宇晴
于矇%鄒誌雲%趙丹丹%郭宇晴
우몽%추지운%조단단%곽우청
单神经元%PID控制%RBF网络%自适应控制
單神經元%PID控製%RBF網絡%自適應控製
단신경원%PID공제%RBF망락%자괄응공제
single neural element%PID control%RBF neural networks%adaptive control
介绍了基于Delta学习规则的单神经元自适应PID控制器的控制算法,提出利用径向基函数(RBF)神经网络对其中未知部分进行辨识,构成单神经元和RBF网络组合控制器,并利用带有纯滞后环节的二阶模型进行仿真研究,结果显示改进型算法比原算法响应速度快、抗干扰能力强,具有更好的动、静态特性。
介紹瞭基于Delta學習規則的單神經元自適應PID控製器的控製算法,提齣利用徑嚮基函數(RBF)神經網絡對其中未知部分進行辨識,構成單神經元和RBF網絡組閤控製器,併利用帶有純滯後環節的二階模型進行倣真研究,結果顯示改進型算法比原算法響應速度快、抗榦擾能力彊,具有更好的動、靜態特性。
개소료기우Delta학습규칙적단신경원자괄응PID공제기적공제산법,제출이용경향기함수(RBF)신경망락대기중미지부분진행변식,구성단신경원화RBF망락조합공제기,병이용대유순체후배절적이계모형진행방진연구,결과현시개진형산법비원산법향응속도쾌、항간우능력강,구유경호적동、정태특성。
A control algorithm of single neural element adaptive PID controller based on Delta learning regulation is introduced. A combining controller is designed with single neural element and radial basis function (RBF) neural network. The RBF neural network is used to identify the undetermined portion of the single neural element self-adaptive PID controller. A second-order system with a pure delay seg- ment is used in simulation to validate the effectiveness of this method. The simulation results proves that this method has better adaptation, anti-disturbance ability, dynamic and static characteristics than the original algorithm.