基于位移时程的框架模型结构参数与损伤识别

本文提出了一种直接运用结构的振动位移响应时程的基于神经网络的结构物理参数和损伤识别的一般方法，该方法通过一个神经网络模拟器和一个参数评估用神经网络实现。在对该方法的理论基础进行论证的基础上，针对一个受基底激励的两层框架模型结构，运用由激光位移传感器所测量的结构位移响应时程，将该方法用于该结构的层间刚度识别和损伤结构的损伤识别。结果表明，神经网络模拟器能够准确地预测参考结构在基底激励下的位移响应，而参数评估用神经网络可以很好地识别出结构刚度参数，并反映出结构损伤的发生。本方法具有一般性，为实际工程结构的参数识别和模型修正提供了一种可行途径。

Parameter and Damage Identification Methodology for A Model Frame Structure Using Displacement Response Time Series

In this paper,a general structural parameter identification strategy based on neural networks is proposed and the theoretical base for the construction of a neural network emulator(NNE) and a parametric evaluation neural network(PENN) is explained.A two-story model frame structure on a shaking table is employed as an illustrative structure to validate the performance of the proposed approach for structural stiffness identification and damage detection using vibration displacement response measurement from laser displacement sensors.Results show that the NNE can forecast the displacement of the reference structure with high accuracy,and PENN can describe the mapping between an evaluation index and structural stiffness parameter.The proposed algorithm is a general and applicable way in practice for near real-time identification and structural model updating.