Skip to main content

Realistic Assessment of Historical Masonry Bridges under Extreme Environmental Actions

Periodic Reporting for period 1 - RAMBEA (Realistic Assessment of Historical Masonry Bridges under Extreme Environmental Actions)

Reporting period: 2019-07-03 to 2021-07-02

Masonry arch bridges form a substantial part of existing bridges and play a critical role within the European transportation system. Moreover, historical masonry bridges belong to the architectural heritage and represent a tangible experience of past construction technologies that should be safeguarded. Many of these structures are located in seismic prone regions and/or in areas subjected to floods and hydrogeological instability aggravated by climate change. Thus besides increasing traffic loading, they can be subjected to extreme environmental actions which may potentially lead to bridge failure causing severe disruptions and major economic and cultural losses. In this regard, realistic structural assessment under extreme loading conditions is a current challenge for the structural engineering community which may prevent future failures by identifying structures in critical conditions, prone of suffering extensive damage from extreme events, and in need of strengthening. In general, the response of masonry bridges is very complex as it is determined by the interaction between different structural and non-structural parts. However, current modelling strategies are formulated following limit analysis principle or simplified 2D finite element descriptions which disregard the masonry anisotropy and the potential activation of spatial, global or local, failure mechanisms of the bridge. Finite element mesoscale approaches enabling an explicit representation of masonry bond, where mortar joints and masonry units are modelled separately, can provide accurate predictions of realistic masonry bridges under different loading conditions representing complex cracking patterns, including transverse cracks due to differential settlements induced by pier scour or earthquakes. However, as this advanced strategy requires superior computational resources and specialist users, it is not suitable for practical assessments.
The RAMBEA project has developed an innovative methodology for realistic analyses of historical masonry bridges under extreme environmental actions. It is based upon a two-scale FE description of masonry components of the bridge combined with contact elements to connect masonry arch and external walls to the continuum backfill domain. The developed modelling strategy allows a reliable description of the anisotropic micro-structure of masonry, considering the cohesive and frictional characteristics of the masonry joints, including the degradation of strength and stiffness under cyclic loading. The model enables realistic predictions of complex, three-dimensional cracking patterns while allowing for computational efficiency. Moreover, the RAMBEA project has developed a practical and robust calibration procedure based upon the mesoscale mechanical properties of bricks and masonry joints which can be easily obtained by non-destructive in-situ tests, suitable to be used for historical constructions and cultural heritage assets.
The developed methodology has been applied to real case studies, where the numerical predictions provided by the novel efficient strategy have been compared against available experimental data or the numerical results obtained employing high-fidelity mesoscale descriptions of the analysed bridge structures. The main outcome of RAMBEA is thus an efficient and accurate numerical approach for the structural assessment of historical masonry bridges subjected to complex loading conditions corresponding to extreme natural events.
The project has developed through the following steps:

• Literature review of the effects of extreme environmental actions on masonry bridges; Selection of representative case studies for which experimental tests are available: two multi-ring arches, one single-bay and one double-bay arch-bridge samples and a large viaduct have been selected; Development of detailed mesoscale models of the selected case studies: the models have been modelled in ADAPTIC1 and calibrated based on the available experimental data.

• Review of the main modelling approaches available in the literature for masonry arch bridges including discrete elements and continuous FE models employing damage-plasticity constitutive laws; Development of a new 3D macroscale model, considering the frictional and cohesive characteristics of masonry and cyclic material degradation; Implementation of the new model in ADAPTIC1; Validation of the model against physical tests and mesoscale virtual tests.

• Extensive review and training on non-destructive testing techniques for the geometrical and mechanical characterisation of historical masonry bridges; Development and validation of a model calibration procedure based on mesoscale parameters of bricks and mortar joints, which can be obtained by non-destructive or low-invasive tests.

• Realistic assessment of the selected case studies under vertical static loading and horizontal seismic actions; Execution of in-situ non-destructive experimental tests on the Quebradas viaduct in Portugal; Identification of the model parameters based on the experiments and an advanced optimisation procedure based on genetic algorithms.

• The study results have been published/submitted for publication in journals and presented at various international conferences and seminars.

____________________________________________________
1 Izzuddin, B.A. 1991. Nonlinear dynamic analysis of framed structures, Imperial College London (University of London).
Current modelling strategies for masonry arch bridges are formulated mostly following the limit analysis principle, which only evaluates the ultimate bridge capacity, or simplified 2D or 3D continuum isotropic FE macroscale models disregarding the anisotropy nature of masonry and the interaction between masonry and backfill, thus providing a crude representation of the 3D response under extreme loading. More detailed FE mesoscale models have been proposed in the literature and applied for the assessment of masonry bridges. These models, represent explicitly masonry bond enabling accurate predictions of realistic masonry bridges under different loading conditions representing complex cracking patterns. However, these advanced strategies require superior computational resources and specialist users, often unsuitable for practical applications.
The RAMBEA project covered a critical knowledge gap between simplistic limit analysis and FE models and mesoscale approaches, providing researchers and practitioners with an efficient but accurate modelling strategy for masonry bridges allowing for the complex 3D interaction between the different components of masonry bridges under extreme loading. Moreover, the project developed a comprehensive model calibration strategy to obtain the model parameters from non-destructive in-situ tests, suitable for historical masonry bridges belonging the cultural and architectural heritage.
In conclusion, the innovative modeling strategy developed in the RAMBEA project drastically limits the computational cost associated with realistic assessments of large bridges, allowing for a model calibration based on practical in-situ non-destructive tests. It is expected to constitute an essential step towards the widespread use of realistic nonlinear analysis in engineering practice to assess existing and historical masonry arch bridges.
Figure 3. Typical brick-masonry multi-ring barrel vault (a) and its description by the FE continuum
Figure 7. Masonry periodic cell (a) and its macroscopic homogenisation (b).
Figure 9. View of the Viaduct (a); accelerometers disposed on the bridge deck (b) and results of the
Figure 10. Failure mechanism of the Quebradas Viaduct predicted by the developed model, due to a ver
Figure 1. Failure mechanisms of the single-span (a) and double-span (b) masonry arch bridges, predic
Figure 4. Numerical simulation of small brick assemblages: capacity curves of the tensile tests (a)
Figure 6. Numerical simulation of the cyclic response of masonry arch bridges interacting with a soi
Figure 5. Experimental in-plane failure mechanisms of a brick-masonry panel (a); failure mechanisms
Figure 8. Simplified normal and shear failure models orthogonal (a) and parallel (b) to the bed join
please, consider the caption reported in Figure 1b
please, consider the caption reported in Figure 10b
Figure 2. Definition of the local layers: (a) masonry bond, (b) macroscale FE discretization; (c) in