Learning Damage Representations with Sequence-to-Sequence Models

Natural hazards have caused damages to structures and economic losses worldwide. Post-hazard responses require accurate and fast damage detection and assessment. In many studies, the development of data-driven damage detection within the research community of structural health monitoring has emerged due to the advances in deep learning models. Most data-driven models for damage detection focus on classifying different damage states and hence damage states cannot be effectively quantified. To address such a deficiency in data-driven damage detection, we propose a sequence-to-sequence (Seq2Seq) model to quantify a probability of damage. The model was trained to learn damage representations with only undamaged signals and then quantify the probability of damage by feeding damaged signals into models. We tested the validity of our proposed Seq2Seq model with a signal dataset which was collected from a two-story timber building subjected to shake table tests. Our results show that our Seq2Seq model has a strong capability of distinguishing damage representations and quantifying the probability of damage in terms of highlighting the regions of interest.

Cyclic performance of exterior R.C beam-column strengthened with different thicknesses of CFRP sheets

The recent ground motion results indicated that the RC buildings are required to be retrofitted by different strengthening techniques. Nowadays, the external strengthening gain interest since its easy, cost effective and not required redesign of buildings. The CFRP sheets are suitable solution and utilized by a number of researchers. However, the numerical cyclic performance of connection strengthened with different thicknesses of CFRP need to be well investigated. This study assessed the performance of RC exterior beam column connection strengthened with CFRP sheets First, two grades of concrete are utilized to be control specimens, normal concrete compressive strength (C20) and high concrete compressive strength (C50) then, the specimens are retrofitted with different thicknesses (1.2, 2.4, 3.6mm) of CFRP sheets. The stresses and damage states showed the importance of connection retrofitting. The CFRP shift the plastic hinge zone away from the panel zone. Furthermore, the results demonstrated that by increase of CFRP thickness the connection resistance will be improved. The comparison between the hysteresis curves demonstrated that the yield and ultimate loading were enhanced for strengthened connection for both concrete grades and the incremental in thicknesses also increase them. The outputs also exhibited that the stiffness and ductility has increased for retrofitted specimens indicating that the CFRP comprehensively overcome the applied cyclic loading and the beam column connection is able to resist such type of loading.

Evaluation method on seismic risk of substation in strong earthquake area

Frequent earthquakes in strong earthquake areas pose a great threat to the safety operation of electric power facilities. There exists a pressing research need to develop an assessment method for the seismic risk of substations, i.e., the hubs of power system networks. In this study, based on Incremental Dynamic Analysis (IDA), Probabilistic Seismic Demand Model (PSDM) and reliability theory, a vulnerability model for a substation is obtained, based on considering the relationships between Peak Ground Acceleration (PGA) and four seismic damage states (complete, extensive, moderate, and slight.) via a probabilistic approach. After an earthquake, the scope of influence and PGA distribution are evaluated using information recorded by the seismic observation stations, based on using interpolation or an empirical formula for the PGA attenuation. Therefore, the seismic risk can be evaluated by combining ground motion evaluation and the pre-built vulnerability model. The Wuqia- Kashgar area of Xinjiang was selected as the study area; it is an Earthquake-prone area, and one of the starting points for new energy transmission projects in China. Under a hypothetical earthquake (MS 7.9), the seismic risk of the substations was evaluated. The results show that: this method is able to give the probabilities of the four damage states of the substations, four substations close to the epicenter only have a probability of slight damage (45%-88%) and other substations are safer.

Theoretical and experimental evaluation of timber-framed partitions under lateral drift

This paper identifies the inherent strengths/weaknesses of rigid timber-framed partitions and quantifies the onset drifts for different damage thresholds under bi-directional seismic actions. It reports construction and quasi-static lateral cyclic testing of a multi-winged timber-framed partition wall specimen with details typical of New Zealand construction practice. Furthermore, the cyclic performance of the tested rigid timber-framed partition wall is also compared with that of similar partition walls incorporating ‘partly-sliding’ connectiondetails, and ‘seismic gaps’, previously tested under the same test setup. Based on the experimentally recorded cyclic performance measures, theoretical equations proposed/derived in the literature to predict the ultimate strength, initial stiffness, and drift capacity of different damage states are scrutinized, and some equations are updated in order to alleviate identified possible shortcomings. These theoretical estimates are then validated with the experimental results. It is found that the equations can reasonably predict the initial stiffness and ultimate shear strength of the partitions, as well as the onset-driftscorresponding to the screw damage and diagonal buckling failure mode of the plasterboard. The predicted bi-linear curve is also found to approximate the backbone curve of the tested partition wall sensibly.

A Screening Methodology for the Identification of Critical Units in Major-Hazard Facilities Under Seismic Loading

The complexity of process industry and the consequences that Na-Tech events could produce in terms of damage to equipment, release of dangerous substances (flammable, toxic, or explosive), and environmental consequences have prompted the scientific community to focus on the development of efficient methodologies for Quantitative Seismic Risk Analysis (QsRA) of process plants. Several analytical and numerical methods have been proposed and validated through representative case studies. Nevertheless, the complexity of this matter makes their applicability difficult, especially when a rapid identification of the critical components of a plant is required, which may induce hazardous material release and thus severe consequences for the environment and the community. Accordingly, in this paper, a screening methodology is proposed for rapid identification of the most critical components of a major-hazard plant under seismic loading. It is based on a closed-form assessment of the probability of damage for all components, derived by using analytical representations of the seismic hazard curve and the fragility functions of the equipment involved. For this purpose, fragility curves currently available in the literature or derived by using low-fidelity models could be used for simplicity, whereas the parameters of the seismic hazard curve are estimated based on the regional seismicity. The representative damage states (DS) for each equipment typology are selected based on specific damage states/loss of containment (DS/LOC) matrices, which are used to individuate the most probable LOC events. The risk is then assessed based on the potential consequences of a LOC event, using a classical consequence analysis, typically adopted in risk analysis of hazardous plants. For this purpose, specific probability classes will be used. Finally, by associating the Probability Class Index (PI) with Consequence Index (CI), a Global Risk Index (GRI) is derived, which provides the severity of the scenario. This allows us to build a ranking of the most hazardous components of a process plant by using a proper risk matrix. The applicability of the method is shown through a representative case study.

Development and Calibration of a 3D Micromodel for Evaluation of Masonry Infilled RC Frame Structural Vulnerability to Earthquakes

Within the scope of literature, the influence of openings within the infill walls that are bounded by a reinforced concrete frame and excited by seismic drift forces in both in- and out-of-plane direction is still uncharted. Therefore, a 3D micromodel was developed and calibrated thereafter, to gain more insight in the topic. The micromodels were calibrated against their equivalent physical test specimens of in-plane, out-of-plane drift driven tests on frames with and without infill walls and openings, as well as out-of-plane bend test of masonry walls. Micromodels were rectified based on their behavior and damage states. As a result of the calibration process, it was found that micromodels were sensitive and insensitive to various parameters, regarding the model’s behavior and computational stability. It was found that, even within the same material model, some parameters had more effects when attributed to concrete rather than on masonry. Generally, the in-plane behavior of infilled frames was found to be largely governed by the interface material model. The out-of-plane masonry wall simulations were governed by the tensile strength of both the interface and masonry material model. Yet, the out-of-plane drift driven test was governed by the concrete material properties.

Correlated Damage Probabilities of Bridges In Seismic Risk Assessment of Transportation Networks

This paper aims to identify and use a logistic regression approach to model the spatial correlation of damage probabilities in expanded transportation networks. This paper uses Bayesian theory and the multinomial logistic model to analyze the different damage states and damage probabilities of bridges by considering the damage correlation. The correlation of the damage probabilities is considered both in different bridges of a network and in the different damage states of each bridge. The correlation model of the damage probabilities is considered in the seismic assessment of a part of the Tehran transportation network with 26 bridges. Moreover, the extra daily traffic time (EDTT) is selected as an operational parameter of the transportation network, and the shortest path algorithm is considered to select the path between two nodes. The results show that including the correlation of the damage probabilities decreases the travel time of the selected network. The average decreasing in the correlated case compared to the uncorrelated case, using two selected EDTT models are 53% and 71%, respectively.