摘要
利用时间序列动态数据处理与建模对结构有损状态进行线性和非线性过程处理，用这两种方法对结构的物理状态进行谱分析和模态频率的提取进行了比较，计算结果表明非线性 Ｓ Ｅ Ｔ Ａ Ｒ 模型在特征谱的识别与提取上比线性 Ａ Ｒ Ｍ Ａ Ｘ 模型更有效，对微小损伤更灵敏。说明了在参数识别中非线性 Ｓ Ｅ Ｔ Ａ Ｒ 模型对物体损伤情况下振动数据建模较为客观，更能反映振动体系的损伤信息，是损伤识别数学模型有效的方法。
This paper presents linear and nonlinear process of the structure with damage by the way of time sequence analysis and the modeling of dynamical data system,respectively.Compared with the two ways of spectrum analysis of the structural physical state and the identification of its modal frequence,the calcaulated shows that the nonlinear SETAR model is more effictive in character spectrum recognition and can reflect more damage information flexisible than the linear ARMAX model and it also confirms that SETAR model is more reseasonable in parameter recognition of data set obtained from the damaged structure.The way of modeling and settling is a promised and prospective course in the field of mathematical modeling in structure damage identification.
出处
《西安工程学院学报》
1999年第3期36-40,共5页
Journal of Xi'an Engineering University
基金
国土资源部九五重点项目
关键词
参数识别
结构振动
非线性
动态数据
结构损伤
parameter recognition
structure vibration
nonlinearity
dynamical data system
