MS3.4 - Bridge Dynamics

The dynamic effects caused by circulating vehicles on bridges are one of the main concerns on the design, monitoring and maintenance of the roadway infrastructure. In the case of modular steel bridges, which are truss structures composed by regular prefabricated units, this regard is critical, as despite presenting numerous advantages, such as a rapid and easy deployment, high adaptability to the terrain and reduced construction costs, they usually face operational restrictions for span lengths larger than 60 m. In the recent years, new efforts are in search of developing long-span modular steel bridges that may be able to overcome these limitations, for which it is essential to fully understand the dynamic effect of vehicles on such bridges. To this aim, this contribution provides a detailed study on two modular steel bridge typologies, considering span lengths from 120 to 140 m. A 3D coupled vehicle-bridge model is used to represent the vehicle-bridge interaction and to evaluate the dynamic load allowance of the structures. The vehicle is represented as a multi-body truck system and the bridges are modelled with the finite-element method. To reduce the computational cost, the modal superposition method is used to calculate the bridge response, assuming that the vehicle does not significantly alter the dynamic behaviour of the structures. The vehicle-bridge coupled system is solved using a fourth-order Runge-Kutta method in the time domain. As part of the analyses, different randomly-generated road surface profiles are considered on the bridge deck, accounting for typical defects and irregularities found on the deck of modular steel bridges. The results reveal the notable influence of defects that involve abrupt vertical displacements that excite bouncing modes of the vehicle. The effect of the velocity of the passing vehicle is also studied, concluding that dynamic load allowance indices tend to decrease as the velocity increases, except when a resonance is produced. Finally, inconsistencies are found when comparing calculated and predicted dynamic load allowance indices with the expressions given in several design codes, as they do not take into account the characteristics of the bridges nor the effect of road irregularities or resonance situations.

Norway is a coastal country that relies heavily on ferry connections in order to provide efficient means of transportation for local communities and industries. Møre and Romsdal county, where the research was conducted has a population of 265.000, an area of 14.300 km2, and over 50 ferry docks in active use that needs to be regularly inspected and maintained. Some of them are part of the E39 coastal route – the main economical artery in Norway. Therefore a potential failure of an important ferry connection would lead to substantial economic losses and massive delays, as well as put a strain on smaller, surrounding road systems that were not designed to accommodate highway traffic. Although ferry docks can be considered to be relatively simple structures, they are often exposed to complex loading conditions caused by heavy traffic, docking operation, and a generally harsh environment. It should also be mentioned that the ferry dock works in two different modes: 1) as a separate structure, and 2) as a combined dock-ship structure when connected with a ferry, making their analysis more difficult. Moreover, there exists little knowledge about the dynamic behavior of those structures. Looking at the available literature it can be deducted that ferry docks, despite fulfilling the same role as bridges, did not receive even nearly a similar amount of attention among researchers in the past. This paper presents lessons learned from measurements on the Rykkjem ferry dock in the Møre and Romsdal county in Norway. Firstly, a description of a new instrumentation system will be given. The system was designed to be flexible and allow for fast deployment since the time spent on installing sensors and preparing for the data acquisition cannot disrupt the ferry schedule. Then, the challenges when creating a Finite Element (FE) model will be described. Here, there exist many parameters that are initially difficult to estimate such as the stiffness of large rubber support that allows absorbing horizontal forces caused by ferry docking, or the stiffness of the vertical supports provided by hydraulic towers. Results show that similarity between mode shapes obtained based on the FE model and measurements can be greatly improved with the use of the finite element model updating technique. Provided results indicate also that the finite element updating can be a useful tool for examining more closely the deterioration process taking place for different structural parts of the ferry bridge decks.

This work addresses the experimental studies developed on a 61m-long stress-ribbon footbridge with fundamental frequencies in the vicinity of regular human-induced excitation, which undergoes significant vibrations when crossed by pedestrians matching a pace around 2 Hz. Past research has shown that the bridge has a complex dynamic behaviour, with a relevant non-linear component, high sensitivity of properties to the construction procedure, and high modal interaction in the frequency range of excitation. Having monitored the dynamic behaviour of the footbridge for several years, this paper summarises the evolution of dynamic properties, namely natural frequencies, vibration modes and damping ratios, and investigates their correlation with temperature and amplitude of vibration.