摘要
空间桁架结构的振动主动控制目前已成为振动控制领域研究的热点。为了研究空间桁架的主动控制方法 ,首先必须建立其准确的模型。神经网络固有的学习能力使其在模型辨识中得到广泛应用。针对空间桁架结构的非线性动态特性 ,本文采用了修正的 Elman递归网络进行了模型辨识。结果表明 ,带有自反馈增益修正的 El-man网络能很好地反映桁架结构的真实情况 ,适用于非线性动态系统的模型辨识。
Active vibration control for space truss structure becomes recently the hotspot in field of vibration control. To study the active control method of the space truss structure, the first step is to setup its model accurately. Due to the complexity of space truss structure, it is hard to setup its accurate mo del by using traditional method. Neural network modeling is widely used in model identifying due to its inherent learning ability. Considering the nonlinear dynamic characteristic of space truss structure, the improved Elman recurrent network is used to model the system. It proves that the improved Elman network with feed back gain is a good method for modeling of nonlinear dynamic system.
出处
《南京航空航天大学学报》
EI
CAS
CSCD
北大核心
2001年第5期445-448,共4页
Journal of Nanjing University of Aeronautics & Astronautics
基金
国家自然科学基金 (编号 :6970 40 0 7)
留学回国人员基金
骨干教师基金
优秀年轻教师基金资助项目
