Work packages

Objectives, starting hypothesis and methodology

Recent work of members of the research and work teams, provides an excellent mean and opportunity to classify buildings and structures in a systematic and relatively easy way, according to their capacity and degradation patterns. Each parameter of the parametric model is linked to an important property of the building/structure. The objective is "to parameterize capacity spectra of selected buildings included in the Building Typology Matrix (BTM) most frequently used in Europe". The starting hypothesis is that the main strength properties of the buildings, that is, initial stiffness, stiffness degradation and ultimate capacity, are directly linked to the five parameters of the model. Therefore, the expected results are a number table allowing for an expedite classification of buildings and a software package allowing to con- struct, in an automatic way, capacity, fragility and damage index curves. Thus, the idea and a relevant advantage of this result is that, once the specific parameters are assigned to a specific building, capacity, fragility and damage indices are constructed in a straightforward and easy way. Concerning to the methodology, the BTM, as proposed in the framework of the Risk-UE project (see for instance Lantada et al. 2009) will be selected as starting point. Afterwards, a search in the literature of existing capacity curves/spectra will be done. Capacity curves developed in previous research projects by the research group will be compiled too. Then, synthetic capacity curves will be developed. Existing software will be used to model specific buildings of interest (see also WP1.2). The TREMURI program (Lagomarsino et al. 2013) will be used for masonry buildings. Reinforced concrete buildings, steel buildings and mixed steel and concrete structures will be modelled and analyzed by using the commercial codes SAP2000 and ETABS (CSI, 2016) and the Ruaumoko software (Carr, 2000). Noticeably, the probabilistic approach, as described below in the WP1.2 explanation, will be adopted too, so that the parameters of the structural models will be defined in a probabilistic way. MatLab platform will be used to carry out the statistical analyses. Thus, central values and dispersions of the parameters will be obtained. A comparative study of results would allow a discussion about a parametric reclassification of buildings.

Work plan, tasks and Gantt charts

The main tasks foreseen to fulfill this objective are related to the update of the state of the art concerning BTMs and papers published concerning capacity curves of the main types of buildings in Spain and in Europe. All this information/documentation will be used to set up a target BTM. Other tasks are devoted to the software preparation and to obtain the family of capacity curves to which the parametric model will be applied. The final comparison of the results and other dissemination and communication activities are also important tasks of this WP. These tasks are described and scheduled in the Gantt chart.

Participants

Personnel involved in this work package are:

• Pujades, L.G. (researcher in charge of WP)
• González-Drigo, R.
• Lagomarsino, S.
• Schmidt, V.
• Silva, V.
• Hidalgo, D.

Objectives, starting hypothesis and methodology

It is well known that buildings and seismic actions have significant uncertainties, so that great uncertainties are also expected in the seismic damage prediction. The objective is "to deal with uncertainties in the expected damage". The starting hypothesis is that uncertainties in the buildings' response can be quantified starting from the uncertainties in the mechanical and geometrical properties of the buildings and those existing in the expected seismic actions. Concerning the methodology, we should have in mind that the building's response to static and dynamic actions, is strongly nonlinear. Moreover, in nonlinear systems, to know the uncertainties in the inputs does not guarantees to know the uncertainties in the outputs. So that sophisticated nonlinear static and dynamic analyses are needed to deal with uncertaintiesin the expected performance. However, to be realistic, the statistical distributions of the main parameters involved in the building behavior must be known, and these statistical distributions must be set up carefully. Similar words can be said about uncertainties in seismic actions. Thus, an important issue to be properly addressed is the sampling procedure and the minimum number of samples needed to represent the expected uncertainties adequately. A number of samples, greater than needed, would waste a lot of computing time, thus making expensive these analyses. A number smaller than required would lead to biased results. In this proposal, a special care will be devoted to sampling procedures and to uncertainties. The Latin hypercube technique will be used and statistical tests will be used to dimension the number of samples correctly. The nonlinear static and dynamic analysis will be performed by using the software described in WP1.1. Expected results are fully probabilistic damage functions for selected building types. Predefined percentiles, at least 5, 50 and 95 percentile capacity, fragility and damage functions will be delivered. Damage curves will be obtained based on the Park and Ang damage index and based on other damage pointers, which have been defined successfully in our research group (Cabrera, 2018). Indeed, on the basis of incremental nonlinear dynamic analyses on SAP2000 (CSI, 2016) our research group proposed a new index related to the degradation of the strength of the building. This new index will be validated by using the damage index based on capacity curves and, probabilistic damage functions will be developed for these cases too.

Work plan, tasks and Gantt charts

As a starting task, a review of the main parameters involved in the building behavior and a study of the corresponding statistical distributions should be performed. This task will be structured regarding the building’s typologies included in the European representative BTM. A similar procedure will be undertaken concerning the uncertainties in the seismic actions. Once the statistical distributions are proposed for each building typology, a chief task is the sampling procedure leading to the minimum number of samples needed to represent the expected uncertainties adequately. The sample dimension should avoid managing excessive sets of data, leading to time consuming analyses, but also insufficient data which would produce biased, incomplete and wrong results. Nonlinear static and dynamic analyses will be done by using the software described above in the general description of WP1.1. Predefined percentiles of capacity, fragility and damage functions will be delivered. Damage curves will be obtained based on the Park and Ang damage index and other damage pointers proposed by our research group. The main tasks related to this work package are:

Participants

Personnel involved in this work package are:

• González-Drigo, R. (researcher in charge of WP)
• Pujades, L.G.
• Lagomarsino, S.
• López-Almansa, F.
• Silva, V.
• Hidalgo, D.

Objectives, starting hypothesis and methodology

It is well known that power spectral density of time histories of the vibration of buildings subjected to microtremors is clearly related to the modal properties and that they may be nice indicators of the conservation/deterioration of its initial state. The objective is to investigate damage diagnosis criteria based on the power spectral densities of the vibrations of structures subjected to weak microtremors −environmental noise−. To this end, a scaled steel structure will be mounted on top of a shaking table and will be monitored with accelerometers. Noticeably, displacement time histories measurements will be collected by means of a last generation remote sensing device: RAR. The starting hypothesis is that structural damage is linked to changes in the modal properties −natural periods −. In turn, displacement time histories can be recorded with remote-sensing interferometric devices. The advantage of these modern devices is that measurements do not require entering the buildings and that do not compromise safety in field jobs during post-earthquake surveys. That is, RAR based techniques avoid the need for inspections and measurements inside buildings. In general, it is a question of delving into non-invasive and contact-free procedures for the evaluation of structural damage. Special emphasis will be placed on the processes of degradation and structural damage and an attempt will be made to establish its relationship with the variation of the vibration periods. Concerning the methodology, a scaled steel framed structure has been designed and built for dynamic tests in the Geophysics lab (UPC). Beam-column connections are adjustable, allowing simulating different degrees of damage. A complete instrumental and numerical study of the kinematic response of the healthy model (free of damage) and some specific damaged configurations will be performed. The study of the correlation between the measurements of inter-story drifts, damage, peak acceleration and the duration of the dynamic inputs, among other parameters, will allow us to obtain new knowledge about structural damage and its diagnosis via non-invasive and contactless measurements. The contrast of the results of the RAR-based technology with classical acceleration measurements (accelerometers) will help us to see the real advantages and limitations of the new method. It is worth noting that we are dealing with very small displacements (several microns) and that RAR device performs better at long periods, that is for mid- and high-rise buildings. In spite that accelerometers and RAR devices allow preliminary analyses of recorded signals, specific software will be developed in order to get fine and supplementary information from the acceleration and displacement time histories. Expected results will be protocols and guidelines for this kind of measurements, including criteria for buildings' instrumentation. Specific documents will be written as deliverables and the obtained knowledge and results will be disseminated in international meetings and workshops and they will be published in high impact journals. Moreover, specific meetings will be scheduled with institutions involved in civil protection related issues, some of them, as the Civil Protection Department of the municipality of Barcelona and the one of the Generalitat of Catalonia and the Cartographic and Geological Institute of Catalonia, have shown explicit interest and are giving support to this proposal.

Work plan, tasks and Gantt charts

A preliminary task is related to the construction details of the scale model of the framed steel structure. The design of beam-column connections must ensure the control of end rotations with the purpose of simulating damage in different orientations. Furthermore, the mechanical device designed to support the structure should allow different degrees of embedment. Next tasks are related to decisions on the location of conventional measurement devices (mainly accelerometers and desplacimeters) and on the protocols and procedures to undertake during the RAR campaigns. The data analysis and the comparison between the results of the contactless campaigns (RAR) and the conventional measurements would complete the research tasks in the laboratory. Specific ad-hoc software will be developed for time-histories analyses. Additional communication and publication tasks close this WP. The tasks, deliverables and milestones foreseen are summarized below:

Participants

Personnel involved in this work package are:

• González-Drigo, R. (researcher in charge of WP)
• Pujades, L.G.
• Lagomarsino, S.
• López-Almansa, F.
• Schmidt, V.
• Silva, V.

Objectives, starting hypothesis and methodology

A district of the city of Barcelona, will be used as a case study to develop a probabilistic risk scenario. The objective is to go beyond previous studies based on deterministic approaches to assess the expected damage and risk in a fully probabilistic way. Updated information about the building stock will be used for this purpose. The starting hypothesis is that risk scenarios are subjected to great uncertainties and that stakeholders and civil protection services must know about this vagueness and their risk man- agement plans must consider them in the estimates of the expected effects of a likely seismic event. The Methodology will be based on the use of the probabilistic damage curves developed in WP1.2. Updated data will be used for this analysis. Expected results are probabilistic risk scenarios, which will be delivered to civil protection services of the municipality of Barcelona. Although the district of the Eixample seems to be the best candidate for this WP1.4, other options could be considered according to agreements with the municipality. Selected risk maps concerning physical damage, homeless and injured people and economic cost will be developed and will be delivered also to the civil protec- tion of Barcelona and of Cataluña. Maps at 5, 50 and 95% confidence levels will be obtained.

Work plan, tasks and Gantt charts

The initial task will be reviewing and updating previous risk scenarios performed in the representative district selected for this study. Data related to the residential building stock will be collected from public databases. Our group has got experience, in the past, working with the databases administrated by the city council and the architecture professional college. Next task is devoted to set up probabilistic damage curves, developed in WP1.2, and adapted to the typologies included in the building stock selection. GIS tools will be used to arrange the data and to generate risk maps, with predefined percentile of confidence, and including information about physical damage, damage to population and economic costs. A comparison between the proposed probabilistic approach and previous deterministic maps, will be performed. According to agreements with the municipality, a task related to draw up reports for its civil protection services will be completed aiming to help improving present civil protection plans. A task summary of this work package, including additional communication and publication of the results, is included below:

Participants

Personnel involved in this work package are:

• González-Drigo, R. (researcher in charge of WP)
• Pujades, L.G.
• Caselles, J.O.
• Foti, D.
• Murià, D.

Objectives, starting hypothesis and methodology

Based on previous experimental works on characterization of soil samples in laboratory, the objective is "to obtain quantitative relations between selected soil properties and GPR scans". The starting hypothesis is that two GPR properties have a greater relation with the soil's properties:

1. the frequency content of the GPR waves and
2. the noise present in the GPR registers.

The frequency content is related to the water content of the medium, which would act as a low-pass filter. Noise is related to the properties of the sedimentary formation, mainly to the granulometry, which would affect the reflectivity. Lab assays can help to establish the correlation between frequency content and noise in the recorded signals and the soil's properties responsible for amplification effects. Concerning methodology, two main procedures are proposed:

1. lab assays and
2. numerical simulation.

Concerning lab assays, the relationship between the frequency-content and the water-content will be analyzed for several different materials/soils. Mid-(500MHz) and High-(2.3GHz) frequency antennas will be used. Existing correlations in the literature between the frequency content and soil types will be analyzed; also, specific software for field applications will be developed.

Concerning numerical simulations, specific target cases, linked to the lab assays and field measurement will be simulated. Existing available software, will be used for the simulations. In order to be available in field surveys, specific patterns found in radar-images will be collected. In addition to the scientific and technical contributions, which will be disseminated in conferences and workshops and published in scientific and technical journal, expected results are quantitative relations between soil properties and GPR images.

Work plan, tasks and Gantt charts

On the basis of previous laboratory and field assays, new experiments will be performed aiming to go forward with new data and new results concerning the characterization of soil samples by analyzing GPR scans. Available equipment allows us to measure resistivity and conductivity in lab assays and in field surveys. The expected tasks related to this work package are detailed below:

Participants

Personnel involved in this work package are:

• Pérez-Gracia, M.V. (researcher in charge of WP)
• Caselles, J.O.
• González-Drigo, R.
• Dérobert, X.
• Fontul, S.

Objectives, starting hypothesis and methodology

Based on previous works on micro- and nano-zonation in Barcelona, and GPR applications the objective is "to develop expeditious methods for soil's characterization based on GPR response with new applications to Barcelona". The starting hypothesis is that it is possible to detect and characterize the subsoil anomalies responsible for the amplification phenomenon, by means of expeditious techniques based on GPR (Salinas et al., 2014). Observed noise in the radargrams can vary spatially (horizontal variations) and in depth (vertical variations). It is also assumed that the water content and the granulometry are related to the dynamic response of soft soils. Concerning methodology, new fieldwork campaigns are proposed in Barcelona. Two specific target areas, with known geology, will be selected to develop the fieldwork in the city, one in an urban zone (street), the other one in an open area (square or park). Correlations among results of the fieldworks and the lab assays must help to validate, and if needed, to improve, the patterns found in lab assays and in numerical simulations. In addition to the scientific and technical contributions, which will be disseminated in conferences and workshops and published in scientific and technical journal, expected results are detailed information on the properties of the soils prospected in Barcelona.

Work plan, tasks and Gantt charts

The plan is to take advantage of previous works of the group on micro- and nano-zonation to go forward with new data and new results concerning the soil’s characterization based on GPR response with new applications to Barcelona. The first task is selecting two specific target areas, with known geology, to develop the fieldwork in the city. The Lab, model and field results will be compared. The following tasks and deliverables are foreseen in this work package:

Participants

Personnel involved in this work package are:

• Pérez-Gracia, M.V. (researcher in charge of WP)
• Caselles, J.O.
• González-Drigo, R.
• Dérobert, X.
• Fontul, S.