Ambient vibration tests are conducted widely to estimate the modal parameters of a structure. The work proposes an efficient wavelet‐based approach to determine the modal parameters of a structure from its ambient vibration responses. The proposed approach integrates the time series autoregressive (AR) model with the stationary wavelet packet transform. In addition to providing a richer decomposition and allowing for an improved time–frequency localization of signals over that of the discrete wavelet transform, the stationary wavelet packet transform also has significantly higher computational efficiency than the wavelet packet transform in terms of decomposing time‐shifted signals because the former has a time‐invariance property. The correlation matrices needed in determining the coefficient matrices in an AR model are established in subspaces expanded by stationary wavelet packets. The formulation for estimating the correlation matrices is shown for the first time. Because different subspaces contain signals with different frequency subbands, the fine filtering property enhances the ability of the proposed approach to identify not only the modes with strong modal interference, but also many modes from the responses of very few measured degrees of freedom. The proposed approach is validated by processing the numerically simulated responses of a seven‐floor shear building, which has closely spaced modes, with considering the effects of noise and incomplete measurements. Furthermore, the present approach is employed to process the velocity responses of an eight‐storey steel frame subjected to white noise input in a shaking table test and ambient vibration responses of a cable‐stayed bridge.

Identification of modal parameters of a bridge from its earthquake responses is crucial for performing damage assessment of the structure. However, all the input base excitations of the bridge may not be measured because of economic concerns and sensor malfunctions. Consequently, evaluating the modal parameters of a bridge under the consideration of incomplete input measurements is a challenging and important task. An approach that combines the continuous Cauchy wavelet transform with an autoregressive time‐varying moving average with exogenous input (AR‐TVMA‐X) model is proposed in this study to identify the modal parameters of a multispan bridge under multiple support earthquake excitations with incomplete measurements. The efficiency and efficacy of the proposed approach are first validated using numerically simulated responses of a three‐span continuous beam subjected to multiple support nonstationary excitations. A standard procedure of using the proposed approach to identify the modal parameters is established according to comprehensive studies on the effects of noise in the data, the number of supports whose excitations are used in the AR‐TVMA‐X model, and the orders of the AR‐TVMA‐X model on the accuracy of identifying the modal parameters. This procedure is further applied to process the earthquake responses of a two‐span cable‐stayed 510‐m‐long bridge to demonstrate the engineering applicability of the proposed approach.

To determine its actual dynamic responses under the wind loads, modal identification from the field tests was carried out for the Kao Ping Hsi cable-stayed bridge in southern Taiwan. The dynamic characteristics of the bridge identified by a continuous wavelet transform algorithm are compared with those obtained by the finite element analysis. The finite element model was then modified and refined based on the field test results. The results obtained from the updated finite element model were shown to agree well with the field identified results for the first few modes in the vertical, transverse, and torsional directions. This has the indication that a rational finite element model has been established for the bridge. With the refined finite element model, a nonlinear analysis in time domain is employed to determine the buffeting response of the bridge. Through validation of the results against those obtained by the frequency domain approach, it is confirmed that the time domain approach adopted herein is applicable for the buffeting analysis of cable-stayed bridges.

No abstract