提出了一种直接运用加速度响应测量的结构参数识别方法。该方法通过两个神经网络对结构刚度和阻尼比进行直接识别。通过结构运动微分方程的离散解，阐明了该方法的理论基础以及构建两个神经网络的依据。首先，依据对目标结构参数的初步估计，假定一个参考结构，构建一个神经网络来描述该参考结构的加速度响应时间序列之间的映射关系，即建立该参考结构的非参数模型。然后，定义加速度响应预测差值均方根向量作为结构参数识别指标，并构建一个参数识别用神经网络来描述该指标与结构参数之间的关系。最后，基于一个框架结构模型振动台试验的加速度响应时间序列实测值，运用以上两个神经网络，在假定结构质量已知的情况下，对该框架模型的刚度和阻尼比进行了识别，并与扫频试验的结果进行了比较。结果表明该方法的结构参数识别结果可靠，所提出的准实时结构参数直接识别方法可行。

The use of sensitive higher modes in physical structural parameter identification of local members in a frame system is discussed in this paper. Analytical studies are supported by low-level vibration tests on a third-scale four-story reinforced concrete frame structure embedded in soil to represent a realistic foundation system. Preliminary numerical modal analysis are carried out to establish the sensitivity of higher modes to localized damage in the frame. Damage to one of the elements on the first story of the frame was simulated by adding a mass to the column and examining the vibration modes before and after the addition of the mass. Through forced vibration tests and selective placement of strain transducers, the Poly-reference least-squares complex frequency domain method (PolyMAX) is used to determine the existence of ‘highly sensitive higher modes’ (HSHMs). An evaluation of the identified higher modes further enables the identification of physical parameters of the target column element using a simple minimization scheme. Findings from the present study indicate that physical parameters of a local element in a frame structure can be identified effectively using HSHMs and that the higher modes of vibration are more sensitive to changes in local physical parameters than lower global modes. The identification of physical parameters as outlined in this study can be applied in structural damage detection.

A novel neural network-based strategyis proposed and developed for the direct identification of structural parameters (stiffness and damping coefficients) from the time-domain dynamic responses of an object structure without any eigenvalue analysis and extraction and optimization process that is required in manyidentification algorithms for inverse problems. Two back-propagation neural networks are constructed to facilitate the process of parameter identifications. The first one, called mulator neural network, is to model the behavior of a reference structure that has the same overall dimension and topologyas the object structure to be identified. After having been properly trained with the dynamic responses of the reference structure under a given dynamic excitation, the emulator neural network can be used as a nonparametric model of the reference structure to forecast its dynamic response with sufficient accuracy. However, when the parameters of the reference structure are modified to form a so-called associated structure, the dynamic responses forecast by the network will differ from the simulated responses of the associated structure. Their difference can be assessed with a proposed root mean square (RMS) difference vector for both velocityand displacement responses. With the associated structural parameters and their corresponding RMS difference vectors, another network, called parametric evaluation neural network, can be trained. In this study, several 5-story frames are considered as example object structures with simulated displacement and velocitytime histories that mimic the measured dynamic responses in practice. The performance of the proposed strategyhas been demonstrated quite satisfactorily; the error for the estimation of each stiffness or damping coefficient is less than 10% even in the presence of 7% noise. Numerical simulations show that the accuracyof the identified parameters can be significantlyimproved byinjecting noise in the training patterns for the parametric evaluation neural network. The proposed strategyis extremelyefficient in computation and thus has potential of becoming a practical tool for near real time monitoring of civil infrastructures. r 2004 Elsevier Ltd. All rights reserved.