摘要
结合神经网络方法和传统补偿方法,研究空间结构风振控制系统的时滞补偿问题.运用单个神经网络取代由两个神经网络组成的控制系统,有效减小系统时滞和产生时滞的环节.针对空间结构风振控制系统,综合运用神经网络方法与状态预测补偿法,构建基于神经网络的多步预测时滞补偿方法.研究显示,该时滞补偿方法克服了传统方法对多自由度系统不适用、数值计算困难等缺陷,可成功运用于空间结构风振控制系统,经时滞补偿后控制效果优于未经时滞补偿的系统.
The characteristics and shortcomings of traditional time delay compensation method were clarified,and the time delay in wind-induced vibration control system of spatial latticed structure was compensated with combination of neural network technique and other compensation methods.The single neural network system was applied to replace the combined control system consisting of two neural networks,in order to reduce the time delay in the control system.With the combination of neural network method and state pr...
出处
《空间结构》
CSCD
北大核心
2008年第2期3-7,13,共6页
Spatial Structures
基金
国家自然科学基金资助项目(10572091)
上海市基础研究重点项目资助(07JC14023)
关键词
空间结构
结构振动控制
神经网络控制
时滞补偿
spatial latticed structure
structural vibration control
neural network control
time delay compensation
