Project ASPIRE  aims at assessing the feasibility to construct Integral Abutment Bridges in earthquake-prone areas. In the last years in Europe there was an increasing interest on this particular class of structures, characterized by the absence of bearing supports and expansion joints, elements commonly used in bridges which are subjected to deterioration due to ageing and thermal effects, thus requiring expensive periodic maintenance. Integral Abutment Bridges have very low costs of maintenance and improved durability while their increased redundancy can result to superior behavior during an earthquake. Still, there are high uncertainties and a lack of standard procedures to systematically evaluate their seismic performance. As a consequence, no indication is given in Structural Eurocodes as well as in other major seismic codes worldwide. The project proposes an innovative multi-disciplinary procedure including: (a) the evaluation of structure-specific seismic hazard; (b) experimental investigation on the shaking table of the University of Bristol Earthquake Lab (c) the calibration of numerical models including their complex Soil-Structure Interaction; (d) the development of novel fragility curves tailored to the salient features of integral bridges, and (e) the development of Risk and Resilience metrics for the estimation of the improved lifecycle associated with the better durability and lower damage of integral bridges due to earthquakes. The project will be implemented at the University of Bristol, integrating and extending the experimental and numerical capabilities of the Earthquake Laboratory on Integral Abutment Bridges, improving at the same time the skills and competences of the researcher. The results of the project will provide a robust scientific basis to justify the use of Integral Abutment Bridges in seismic areas, with a special attention to European countries and Road Networks. 

The site-specific hazard was defined both by selection of natural ground motions using data from strong ground motion databases and by defining an up-to-date method for the generation of simulated ground motions. The method is based on the time-frequency decomposition of earthquake ground motion records and allows to generate a number N of simulated ground motions starting from a few input parameters (Magnitude, Distance, soil type and fault type). In addition to the method, a novel software tool was developed allowing to obtain apply the method and also to generate spectrum-compatible ground motions. The definition of seismic hazard is documented in peer-reviewed and conference papers. 

A numerical model was developed using the OpenSees platform. First, the numerical model of the shaking table test of EU/H2020 SERA/SERENA project was defined, with the aim of validating the numerical results with experimental ones, and with the final objectives of 1) evaluate results that were not directly measured during experimental testing, such as earth pressures, and 2) obtain indications to extend the results obtained at the small scale to real bridges. The numerical model, including the model of the bridge and foundations with the piles and the soil domain, takes into account the soil-structure interaction by appropriate interfaces between structural and soil elements. Static and Dynamic analyses were performed, allowing to obtain results in terms of acceleration, frequencies, displacements, bending strains and earth pressures. 

 The numerical results were compared with the experimental ones. The comparison allowed to validate the numerical strategy, and the earth pressures obtained were compared with the new provisions contained in the new version of Eurocode 8 Part 2 on Bridges. In particular, this is the first project that compares the European Code with numerical results, allowing to draw suggestions and guidelines for the application of Eurocodes. The influence of different parameters was investigated, such as the increasing mass of the bridge deck, and the study of the areas of the backfill soil which are more likely to be damaged and therefore need mitigation strategies. 

After having defined the earthquake ground motion and the numerical model of real integral bridges, a number of sets of earthquake-bridge samples was defined. A non-linear time history analysis is performed for each of the cases, allowing to obtain a certain damage state for each of the component of the bridge. While generally the damage of a conventional bridge is described by the damage of piers, in the case of integral bridges different damage indicators have to be defined, involving damage to the abutments-backfill system.

 The life-cycle performance of integral bridges was investigated by taking part to the experimental campaign of project PLEXUS PLUS  funded by UKCRIC and composed by two tests:

A small-scale pseudo-static test was conducted to simulate the effects of repeated thermal variations on the bridge throughout its lifespan. The focus of the test was to replicate the soil-structure interaction resulting from seasonal expansion and contraction of the bridge deck, and to assess the effectiveness of various monitoring techniques. The measurement of earth pressures behind the abutment wall was performed using pressure cells, while Digital Image Correlation (DIC) was employed to monitor lateral stresses and settlements of the backfill soil. Additionally, a Ground Penetration Radar was utilized to estimate changes in soil density during the test. Displacements and deformations of the wall were monitored using LVDTs and strain gauges. By combining data from these different instruments, a preliminary evaluation of the soil-structure interaction behaviour of the system was conducted.

A large-scale pseudo-static experimental campaign (100 cycles) to reproduce the life-cycle thermal variations of integral bridges. It was the first experimental campaign to be carried out in the new National Facility for Soil-Foundation-Structure-Interaction Laboratory (SoFSI) at the University of Bristol.

The 6m x 5m x 4m Soil Pit was used, and settlements, strains and earth pressures were measured on the abutment and in the backfill soil. The setup of Plexus Plus included a reinforced concrete wall hinged at the base surrounded by two towers of “Lego” blocks and connected to a 1MN static actuator, while the area behind the wall was filled with 50 tons of silica sand (see Figure 7a). An example of results in terms of earth pressures behind the abutment wall is given in Figure 7b, with the simulated earth pressures obtained after 10, 60 or 120 years of the bridge life-cycle.

The large-scale test of Plexus Plus, besides being among the first comprehensive tests of Integral bridges at large scale, featured a number of innovative monitoring techniques including optical fibres, digital image correlation (DIC) and Ground Penetration Radar (GPR) to monitor the density of the soil during the tests. In addition, arrays of accelerometers were installed to measure the ambient noise vibrations induced by the shaking table of SoFSI Lab.