Flutter, buffeting response and vortex shedding are crucial factors when designing long-span bridges. An analysis of these phenomena requires experimental data, which can be provided by wind tunnel tests. The forced vibration method is chosen in this study because it is considered to be more reliable and better suited to provide data at high velocities, large amplitudes and more intense turbulence. The models currently used to describe self-excited forces in bridge engineering are linear. However, it is a well-known fact that the principle of superposition does not hold in fluid dynamics. Several case studies have shown that it is a fair approximation when predicting wind-induced dynamic response of bridges if the response is dominated by one vibration mode in each direction. Yet, it is uncertain how well the current models will be able to predict the self-excited forces for a more complicated motion. Currently developing experimental setup will enable the performance of forced vibration tests by applying an arbitrary motion. This paper focuses on extending three identification methods developed for single harmonic motion such that they can be applied in more complex motion patterns. Numerical simulations of forced vibration tests were performed to test the performance of those extended methods.
Keywords Forced vibration test • Arbitrary motion • Aerodynamics • Bridges • System identification
IntroductionThe aerodynamic properties of the bridge deck play an important role in long span bridge design. Wind tunnel tests are therefore commonly carried out in the design process. Currently, sectional model tests, where only a part of the bridge deck is considered, are the most commonly used [1][2][3]. Compared to other available testing techniques, sectional models are low cost, easy to build and are suitable for relatively small wind tunnels. Moreover, details of the bridge deck known to have significant impact on aerodynamic performance, such as handrails, can be easily reproduced. The main outputs from the tests are static force coefficients and aerodynamic derivatives, which define the self-excited and static wind forces and characterize the overall aerodynamic behavior of the bridge deck.There exist two main types of sectional model tests. Either the bridge deck is suspended in springs in a free vibration test or it is forced in an oscillatory motion in a forced vibration test. The free vibration method has been the most common method in the past. However, in recent years this trend seems to be changing. Data from forced vibration test are by some researchers considered to be much more reliable and repeatable, [4]. In particular, the forced vibration test outperforms the free vibration test when considering large reduced velocities, large motion amplitudes, high turbulence intensities and cross sections sensitive to vortex shedding [5,6].The load models most commonly used in bridge aerodynamics are linear engineering approximations. It has been shown in several case studies that the models are working wel...