MS3.2 - Bridge Dynamics

The behaviour of thin-walled members subjected to non-uniform bending cannot be properly predicted by the Euler-Bernoulli and Timoshenko models. The inadequacy of the classical beam theory depends on several phenomena occurring in real box girders, namely: (i) warping, (ii) shear flange deformation (shear-lag), (iii) shear web deformation, as well as (iv) distortion of the cross-section in its own plane. In this paper a free vibration analysis of a rectangular box girder is carried out; the problem is addressed in two steps: (i) first, the simplest case of a box girder with undeformable but warpable cross-section is approached; (ii) then, the most comprehensive case of a box girder with a deformable cross-section is dealt with. For the box girder with undeformable cross-section various approaches are followed. A simple but efficient method is proposed to study the combined effect of shear-lag (flange deformation) and web shear deformation. A new analytical expression for the effective flange width is derived by the equilibrium condition of the flange, under the hypothesis of transverse inextensibility. Moreover, the web deformation is taken into account considering an effective shear area, instead of the traditional shear area: indeed, the shear lag reduces the width of the flange and this, in turn, reduces the effective geometric characteristics. The effective flange width and the effective shear area are used to account, in approximate way, for shear flange and web deformation in the (no warping) Euler-Bernoulli and Timoshenko models. An alternative two d.o.f. model for the shear-flexural behaviour of box-girders with undeformable cross section, based on the Generalized Beam Theory, is formulated. Such a model takes in account warping, so that it does not require the use of the effective flange width and effective shear area concepts. Concerning box girder with deformable cross section, a four d.o.f. model is proposed. It is obtained by adding two degrees of freedom to the former GBT model, describing cross-section distortions, of symmetric or antisymmetric type with respect the flexural axis. The influence of the distortional modes on the overall mechanical behavior is investigated, as a function of the geometrical characteristics of the box-girder. All the analytical methods discussed here are validated by comparisons with finite element analyses.

Selected results collected in the continuous dynamic monitoring of a 3-span overpass using MEMS accelerometers are reported in the paper. The investigated structure is a steel-concrete bridge including a trapezoidal box girder of variable height and thin-walled cross-beams with curvilinear axis, so that the overall geometry is rather complex. The overpass has a total length of 108 m and spans of 30 + 48 + 30 m length. 22 tri-axial accelerometers were installed along the two sides of the infrastructure, so that 3 cross-sections per span were instrumented (i.e., at quarter spans and midspan); in addition, as the intermediate supports consist of two elastically deformable inclined steel struts, accelerometers were installed also in the cross-sections corresponding to the intermediate supports. After a concise description of the bridge at study, the paper firstly presents the dynamic characteristics of the structure, as evaluated from ambient response data collected in preliminary ambient vibration tests, and the FE model correlation analysis. Subsequently, the paper describes the dynamic monitoring system installed in the bridge and selected results obtained during the first year of monitoring. Several bending and torsion modes are identified, with the temperature significantly affecting all natural frequencies. Moreover, a clear effect of changing environment is detected for torsion modes as well, whereas no remarkable changes of bending mode shapes are detected.

Bridge assessments aim to establish the structural capacity of assets to ensure regulatory standards are met or to eventually design and plan for refurbishment works. However, the techniques used to conduct these assessments are based on potentially overconservative codes of practice, hence resulting in unnecessary repairs. This approach is not cost-efficient nor environmentally friendly. The alternative is to conduct more precise assessments by taking on-site measurements using traditional equipment such as strain gauges; however, equipment costs, disruptions due to road closures or rail possessions, and health and safety risks associated with this alternative approach defeat its very purpose. A mobile phone application, referred to as SMARTdynamics, was developed to take site readings of the dynamic response of a cable-stayed highway bridge and three plate girder railway bridges by making use of built-in accelerometers. The application allows to determine key dynamic properties including natural frequencies, displacements, and damping ratios. SMARTdynamics includes a novel feature for smartphone accelerometers: the ability to control the sensor remotely with automated application feedback, allowing to record the dynamic response at specific times and with precision. The aim of the study was to determine the reliability of smartphone accelerometers, establish the effectiveness of mobile phone usage in the bridges and civil structures industry, and demonstrate that the derivation of more realistic dynamic properties can offer benefits in determining less conservative structural capacities. Acceleration readings were taken for a cable-stayed highway bridge to determine the tension force within the stay cables. The obtained results confirmed the viability of mobile phone accelerometers as a serious asset management tool. Site readings were also taken to determine the dynamic amplification factors of plate girder railway bridges and revealed a mean average reduction of 45% in the real dynamic increment factor when compared with values derived from codes of practice. The study confirmed that moving towards a realistic approach to bridge structural assessments with user-friendly and cost-effective tools can have tangible benefits and should be considered as a viable alternative to traditional methods.