Mass normalized mode shape identification of bridge structures using a single actuator-sensor pair

Abstract

Summary Identifying mode shapes of bridge structures typically require a dense array of stationary sensors to accurately capture mode shapes with appropriate spatial resolution. An alternative approach is developed here, which requires only a single pair of actuator and sensor. The mode shape identification involves, first, identifying the natural frequencies and modal damping ratios, followed by an estimation of the mass normalized mode shapes components at the excited and measured degrees of freedom. An input–output balance is employed with a series of inputs and outputs obtained from a sequence of tests. The sequence of tests include exciting and measuring at different locations along the bridge, using either a roving actuator and/or a roving sensor; the requirement for a unique identification is that the roving actuator and sensor must be collocated in at least one of the tests. The performance of the proposed method using different types of responses, namely, displacement, velocity, and acceleration, is assessed using numerical simulations. The effect of different types of errors in the identification process is also studied. The method is finally applied to experimental data obtained from laboratory scale tests.

Bridge mode shape identification using moving vehicles at traffic speeds through non‐parametric sparse matrix completion

“…Different techniques have to be applied to the recorded mixed responses in order to extract the dynamic characteristics of the bridge. [25][26][27][28][29] In Yang et al, 24 the dynamic analysis of vehicle-bridge interaction was performed for an analytical bridge-vehicle system, consisting of a simple 2D beam representing the bridge, and a moving mass-spring system representing the vehicle. It was shown that the bridge frequency could be extracted from the vibration of the vehicle.…”

“…The concept of capturing the information of the bridge using a sensor installed in a moving vehicle was first proposed by Yang et al 24 The challenge of this concept lies in the fact that the data measured on the moving vehicle include coupled information from both the vehicle and the bridge. Different techniques have to be applied to the recorded mixed responses in order to extract the dynamic characteristics of the bridge 25–29 . In Yang et al, 24 the dynamic analysis of vehicle‐bridge interaction was performed for an analytical bridge‐vehicle system, consisting of a simple 2D beam representing the bridge, and a moving mass‐spring system representing the vehicle.…”

Damage Identification Method of Box Girder Bridges Based on Distributed Long-Gauge Strain Influence Line under Moving Load

“…Damage identification technology can be grossly divided into two species: frequency-domain signal (FDS)-based methods, and time-domain signal (TDS)-based methods [ 6 ]. FDS-based methods mainly use the dynamic characteristics of the structure, such as natural frequency, mode, etc., to identify the damage and parameters of the structure, and then to evaluate its working state and practical life; extensive literature reviews on FDS-based methods have been conducted [ 7 , 8 , 9 , 10 , 11 ]. However, this method is not sensitive to local structural damage in practical application, which leads to the deviation of identification.…”

Time‐varying frequency‐based scaled flexibility identification of a posttensioned concrete bridge through vehicle–bridge interaction analysis

“…Recently, He et al proposed a method to identify mass-normalized mode shapes using parking vehicle-induced frequency change, in which a two-axle vehicle should be parked at different locations of the bridge progressively. 23 This method can obtain mass-normalized modal shapes of the bridge from vehicle-induced vibrations, but it requires the vehicle to be parked at different locations of the bridge for obtaining dense modal shapes. In addition, abovementioned methods are not further utilized for identifying flexibility matrix and predicting static deflections of a bridge under arbitrarily distributed static loads.…”

“…Although bridge mode shapes can be indirectly identified by using responses of a passing vehicle, they are arbitrarily scaled and have a constant scaling factor with respect to mass‐normalized mode shapes, meaning that the identified mode shapes from indirect methods cannot be further adopted for FRFs reconstruction and flexibility identification. Recently, He et al proposed a method to identify mass‐normalized mode shapes using parking vehicle‐induced frequency change, in which a two‐axle vehicle should be parked at different locations of the bridge progressively 23 . This method can obtain mass‐normalized modal shapes of the bridge from vehicle‐induced vibrations, but it requires the vehicle to be parked at different locations of the bridge for obtaining dense modal shapes.…”