Abstract

     This paper presents a rapid machine learning-based damage detection framework for identifying the damage extent of concrete shear wall buildings. For this purpose, a parametric study was carried out to determine the most efficient machine learning algorithm in classifying the damage states of the building. According to this parametric study, the K-Nearest Neighbor (KNN) learner was selected as the reference prediction model because of the higher accuracy achieved by this algorithm. Bayesian Optimization (BO) algorithm was used to tune the hyperparameters affecting the accuracy of the model. The most efficient attributes were selected from the set of damage indicators through the BO algorithm to train the model. Three different benchmark buildings, including 7-,9-, and 13-story concrete shear wall buildings, were used to evaluate the robustness of the proposed framework. A suite of 111 pair motions, originally developed for the SAC project, were employed to create a generalized dataset. These motions were uniformly scaled from 0.05 g to 1.5 g to expand the intensity range of the events. All the acceleration signals were polluted to 10% noise using white Gaussian signals to simulate the field condition. Results reveal the efficiency of the proposed framework in identifying the extent of damage in concrete shear wall elements of the building. In addition, a parametric study was conducted to illustrate the reliability of two commonly used features, called Cumulative Absolute Velocity (CAV) and the energy ratio between the acceleration response and the input excitation, in determining the damage states of the shear walls under seismic motions.

Introduction

     Assessment of structural safety is essential for post-earthquake restoration. Generally, to evaluate the post-earthquake vitality of the exposed structures, a complete visual inspection is required [1]. Coordination and implementation of the manual visual inspection needs several dedicated teams and monetary resources. In this regard, considerable efforts have been carried out to automate the visual inspection process, e.g., image-based visual inspection [2]. However, such an engineering visual inspection is only able to detect the visible defects that occurred in the structures [3]. This means that some serious invisible damages may be left latent during the visual inspection. The process of identifying and tracking the structural damage is known as the Structural Health Monitoring (SHM) [4,5]. In SHM, damage detection is related to the methods developed for identifying the probable existence, severity, and location of the structural damage. Model-based and data-driven methods are two of the most commonly used strategies proposed for damage detection. Model-based methods generally involve a system identification algorithm paired with a finite element analysis to update the structural model [6]. The performance of the model-based approaches directly depends on the accuracy of the information about the physical properties of the under-study structure. In addition, updating the finite element model based on the physical properties of the structure is computationally expensive for large-scale structures, and rapid condition monitoring could be challenging in this condition [7,8]. On the other hand, data-driven methods apply statistical learning algorithms to the vibration data captured from the structure. This method uses learning algorithms to construct a classification or regression learner for predicting structural damage [9, 10]. Tsou and Shen [11] proposed the use of Neural Networks (NNs) for predicting the severity and location of the structural damage. They used the variations in the modal properties of the structure as the damage feature to identify the damage. Worden et al.

Conclusion

     This paper presented a rapid algorithm for identifying the severity of local damage in the concrete shear wall buildings. A total number of 1884 nonlinear response history analyses were conducted for each building using the SAC motions. A suite of damage indicators was extracted from the acceleration signals to construct the prediction models. A parametric study was carried out to determine the most efficient learner for classifying the damage states of the buildings. The KNN classifier was selected to construct the predictive models because of its maximum accuracy compared to the other algorithms. The Bayesian optimization algorithm implemented to tune the hyperparameters of the classification learners. The main conclusions derived from the study are as below:

Abstract

     This paper proposes an innovative seismic design approach for shallow rectangular cut-and-cover underground subway or railway stations. The traditional approach is to design rigid frame-like structures by connecting rigidly the main horizontal and vertical structural elements (side walls, top, bottom and intermediate slabs, and central columns); on the contrary, the proposed strategy consists of joining them by means of hinged and sliding connections, in order to obtain structures whose lateral stiffness is almost zero. The objective of this approach is to be able to adapt to the transverse racking motion imposed by the seismic ground motion without significantly increasing the internal forces in the structural members. The aforementioned flexibility of the joints is achieved by interposing rubber bearings between the connected structural elements. As a case study, an existing 2-story 3-bay subway station located in Southwest China is redesigned with the proposed technology; its seismic performance is numerically investigated by performing nonlinear dynamic analyses for a number of horizontal transverse input ground motions (accelerograms) representing the site seismicity. Such inputs are scaled to fit PGAs ranging from 0.1 to 0.6 g. As expected, the results of the time-history analyses reveal that the seismic damage to the structural members is significantly alleviated in the sliding-hinged alternative solution. This conclusion can be understood as a preliminary confirmation of the satisfactory seismic performance of the proposed technology.

Introduction

     The nowadays rapid development of buried structures, such as tunnels, underground parking lots, subway and railway stations, etc. provides alternatives to the insufficient ground space. Taking subway stations as an example, according to the International Association of Public Transport [1], there were 11,084 stations spreading over 13,903 km of subway operating lines at the end of 2017 in the world, and in the next few years another 1700 km were to be built. However, recent and past investigation reveals that earthquakes can damage underground stations [[2], [3], [4], [5], [6], [7], [8]]; therefore, further research is required on their seismic design.

     The traditional seismic design of underground stations is based on providing sufficient stiffness and strength to absorb the strains imposed by the seismic ground motion without excessive damage. This approach is basically correct and leads to safe constructions, but omits that earthquakes are indirect actions (i.e. imposed displacements): designing stiffer structures generates an increase of the soil-structure interaction force, thus providing little benefit (if any). Another smarter strategy is to try to reduce that force, either by lowering the lateral stiffness of the soil (locally) or the station. The first objective is achieved with seismic isolation [9,10], and the second with structural flexibilization measures [9,10]. Both approaches are briefed in the next two paragraphs, respectively.

Conclusions

     This paper proposes a new alternative seismic design solution for rectangular cut-and-cover underground stations; it consists of replacing the traditional rigid connections between the structural elements with hinged and sliding joints, as to obtain near-zero lateral stiffness. The required flexibility of the joints is achieved through the use of rubber bearings. The ensuing structures are expected to be able to accommodate the imposed seismic racking without relevant strains in the structure. As a case study, an existing 2-story 3-bay subway station located in Southwest China is redesigned with the proposed technology; its seismic performance is investigated numerically by time-history analyses for a set of input seismic accelerograms. These inputs are scaled to fit PGAs ranging from 0.1 to 0.6 g.

      The main output of this study is that the damage to the structural elements is considerably lessened in the sliding-hinged alternative solution; this trend is particularly intense for not only central columns but also structural connections. On the other hand, the acceleration and drift do not alter much; therefore, no higher damage to the acceleration and drift-sensitive nonstructural components is to be expected in the proposed alternative solution. These conclusions hold for all the considered levels of PGA. The remarks obtained can be read as a preliminary validation of the proposed seismic design approach.

Abstract

     The paper studies the performance of Offshore Wind Turbines (OWTs) founded on Suction Bucket Jackets (SBJs) in clay under combined wind and seismic loading. A detailed 3D FE model of the soil–foundation–structure (SFS) system is developed and used as a benchmark to assess the efficiency of an enhanced Winkler-based “Caisson-on-Winkler-Soil” (CWS) model, where the soil is replaced by nonlinear hysteretic elements. The proposed CWS model captures residual deformations and hysteresis and offers physical coupling between vertical and moment loading. It allows excellent prediction of the  failure envelope in the most relevant first quadrant of  space, where the loads act in the same direction. Despite its successful application for the simulation of inertial loading, the CWS model fails to reproduce the dual shearing mechanism that develops at the caisson shaft during shaking, stemming from the combination of kinematic soil shearing due to the vertically propagating shear waves, and shearing due to the superstructure inertial response, thus underpredicting the co-seismic caisson settlements. For the prediction of the latter, the research utilizes spectrum compatible input motions and 3D FE models of varying geometric and material properties to derive linear regression equations that correlate the co-seismic dimensionless settlement of caissons ) with characteristic dimensional variables of the problem under investigation and the Arias Intensity  of the surface ground motion. As a final step, the paper proposes a hybrid method for performance-based assessment of SBJ OWTs. The proposed method employs the simplified CWS model to calculate the VHM loads and approximately estimate horizontal displacements and rotations at the jacket legs, followed by a preliminary assessment of caisson settlements using the correlations of  with  on the basis of spectrum-compatible input motions.

Introduction

     Following the ambitious energy targets set by countries worldwide, the offshore wind sector has seen impressive growth over the last decade, transforming from niche technology to a global industry. As of 2019, Europe alone has 22 GW of installed offshore wind capacity – enough to cover 2.3% of its electricity consumption [1], while similar or even more potent growth is observed in many countries around the world, such as China, the USA, and India. Part of this energy transition is led by technology-driven innovation, with offshore wind turbines reaching new heights and larger capacities, while wind farms are moving deeper into the sea to harness the increased energy potential. Giventhis new “deep-water” environment, traditional OWT support structures are gradually being replaced by more cost-effective and agile foundation systems in a quest for reduction of investment costs.

Conclusions

     The paper has developed a simplified performance-based assessment technique for SBJ OWTs founded in clay under the combined action of wind () and earthquake () loading. Using an example of an 8 MW jacket-supported OWT, installed at 60 m depth in the Adriatic Sea, system performance was assessed employing a detailed 3D FE model of the soil–foundation–structure (SFS) system (global 3D FE model). After deriving insights on system performance, the global 3D FE model was used as a benchmark to assess the efficiency of an enhanced Winkler-based “Caisson-on-Winkler-Soil” (CWS) model. Soil–suction caisson interaction is represented by nonlinear hysteretic elements, capturing residual deformations and hysteresis. The proposed CWS model offers physical coupling between vertical and moment loading by introducing distributed vertical hysteretic elements along the caisson shaft, simultaneously contributing to vertical and moment shaft resistance.

Abstract

     The performance of structural timber connections is of utmost importance since they control the global response of the building. A ductile failure mechanism on the global scale is desirable, especially in the design of structures in seismic areas, where dissipative components in which ductile failure modes need to be ensured are considered. Therefore, the knowledge of possible brittle failure modes of connections is crucial. The paper investigates the brittle failures of laterally loaded dowel-type connections in cross-laminated timber subjected to tensile load in a lap joint configuration through experimental investigations and analytical estimations. A set of 13 different test series has been performed with fully threaded self-tapping screws of 8 mm diameter and different lengths (40 to 100 mm) in cross-laminated timber composed of 3 or 5 layers (layer thickness range from 20 to 40 mm), giving rise to the activation of different brittle failure modes at different depths. Plug shear was among the most typically observed failure modes. A previously proposed model for the brittle capacity was applied to the tested connections at the characteristic level. As shown by the performed statistical analysis, the existing model is not reliable and mainly unconservative.

Introduction

     Nowadays, one of the best solutions for decarbonisation of the construction sector is to introduce more wood into building practices, which means that this material should not be used only for single-family housing, as was common practice in the 20th century. Consequently, there is a high demand for wood to be used in various types of structures, especially mid-rise (up to 10 storeys) and high-rise buildings (more than 10 storeys) that are the predominant type of buildings in dense urban environments. With the development of high-performance engineered wood products (EWPs) such as cross-laminated timber (CLT), building with wood has enabled larger and taller buildings to be erected. However, due to the relatively short period since emerging and implementation of advanced EWPs and their new structural systems have been present on the market, the knowledge and experience using these materials for high-rise buildings is still limited.

Conclusions

     The paper analyses the possibility of brittle failure modes for tensile loaded CLT with laterally-loaded screw connections, considering loading orientations both parallel and perpendicular to the direction of the laminations in the outer CLT layer. High-load bearing connections in CLT structures as analysed in this work would mainly be needed for hold-downs, where in most cases the connection is oriented parallel to the grain of the outer layer (α = 0°), and mainly designed for tensile loads.

     The experimental results suggest that the load-bearing capacity of the connection increases with the width of the CLT specimen; e.g. a 45% increase was found between series CD2 (w = 250 mm) and series CD4 (w = 750 mm). However, increasing specimen width does not appear to affect the resulting elastic stiffness. The length of the screws was found to have a large impact on the elastic stiffness; e.g. an increase of more than 70% was found between 40 mm (series CD1) and 100 mm (series EF1) fastener lengths. The CLT layup does not seem to influence the elastic stiffness notably, while it influences the resulting load-bearing capacity, mainly concerning the penetration of the fastener into the different layers.

Abstract

     The purpose of this paper is to study the behavior of pile-supported structures in liquefiable soils, specifically when the soil surrounding the pile transits from no-liquefaction to full-liquefaction. A series of shaking table tests were performed on four pile-supported structures subjected to different input motions and, as a result, with different times to reach full-liquefaction. The bending moment of the piles during the transient phase is compared with those predicted in pre- and post-liquefaction stages. The experimental results showed that the maximum bending moment may occur during the transient phase (i.e. during the development of excess pore pressure before the soil is fully liquefied). Arguably, the observed amplification in bending moment is caused by the the tuning effect between the predominant frequency of the input motion and the frequency of the pile-supported structure, which is progressively decreasing during the liquefaction process. Results are presented using a non-dimensional framework whose parameters are derived from the governing mechanics. A new parameter TAF (Transient Amplification Factor) is defined to predict the design bending moment during the transient phase. It is shown that the transient bending moment can be obtained from the newly introduced parameter TAF, and the values of maximum bending moments in the pre- and post-liquefaction stages. It is found that TAF is a function of two easily obtainable parameters: (a) time taken to reach full liquefaction, this can be obtained through site response analysis; (b) elongation of natural period of vibration, expressed as ratio of the time period of the structure at full liquefaction to the time period at zero-liquefaction. Finally, practical implications of the main findings are discussed.

Introduction

     The dynamic behaviour of pile-supported structures founded in liquefiable deposits is still an area of active research [[1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30]] due to the poor performance of piled foundations as observed in most of the recent earthquakes [9,31]. The current methods of pile design – see for Example codes of practices such as JRA [32,33] NEHRP [34], IS-1893 [35], focus on avoiding bending, buckling and settlement failures. The effect of dynamic soil-structure-interaction (SSI) effects is taken into account by means of empirical correlations.

     Fig. 1 shows the two main stages of loading in a typical pile-supported structure during an earthquake. Fig. 1 (a) shows the stage at the start of the shaking, before any build-up of the excess pore water pressure takes place. In this stage, piles are mostly subjected to inertia loads induced by the ground shaking and oscillation of the superstructure. It is known that inertia loads tend to generate high bending moments at relatively shallow depths along the pile. Fig. 2 (a) shows the “Beam on Non-Linear Winkler Foundation” model, in which the effect due to soil-structure interaction is modelled by a set of springs distributed at discrete locations along the length of the pile. Each spring is defined by a relationship between the soil pressure (p) and pile deflection (y) referred to as p-y curve. This method is conventionally used to compute the internal forces (i.e. bending moment and shear force) and pile’s deflection. The typical shape of p-y curves for sand in non-liquefied condition is shown in Fig. 2 (b), and further details can be found in API [36]. However, in saturated loose to medium dense sand,the ground shaking induce a gradual increase in pore water pressure, resulting in the soil to progressively lose its strength and stiffness. When the excess pore pressure equalises the overburden pressure, the soil is said to be in a full-liquefaction condition; at full liquefaction the foundation loses the support from its sorrounding soil and the piles tend to act as an unsupported column over the liquefied layer, see Refs. [37,38]. Fig. 1c shows the time history of a real earthquake together with excess pore water pressure profile, which can be either measured in experiments or computed by means of site response analyses. From the figure it can be noted that the onset of liquefaction occurs in a finite amount of time, whose duration depends mostly on the input motion characteristics and the density of the soil; the latter is typically expressed in terms of relative density. Typical values of time to reach liquefaction range between 6 and 15s, however, the actual time can be estimated by means of nonlinear finite element analysis.

Experimental results

3.1. Frequency domain analysis

     To quantify the change in vibration characteristics of the models, the experimental data was first analysed in frequency domain. Before the start of each test, the natural frequencies of the models were estimated from free vibration tests performed by exciting the structure with an impact hammer. The frequency response functions FRFs obtained from the free vibration tests are depicted in darker lines (denoted by before shaking) in Fig. 4, Fig. 5, Fig. 6. In the same figures, the power spectral density (PSD) of the input motion and FRF estimated after liquefaction is shown for comparison. It may be observed that after liquefaction the natural frequency of the models reduced significantly, which may be attributed to the development of excess pore pressure, and in some cases formation of plastic hinges in the piles. From Table 3 it is interesting to note that the four models failed in tests CH1 to CH4. (see Fig. 4 for frequency response). Other tests listed in Table 3 were carried out to investigate the sole effect of liquefaction on the variation in natural frequency of the models and transient bending moments. In these tests, formation of plastic hinges was deliberately avoided with lower pile-head mass. The results shown in Fig. 5, Fig. 6 indicate that the natural frequency of the models reduced to about half due to subsurface liquefaction. As the model transited from higher to lower frequencies it was likely that the seismic response amplified due to a temporarily matching between natural frequency of the models and predominant frequency of the input motion. This aspect is further investigated in the next section by considering the bending moment time histories along the piles.

Abstract

     The geometry of three-dimensional subsurface structures plays an important role in determining local seismic site effects as in the case of alpine valleys. Detailed knowledge of these structures is fundamental in seismic hazard and risk studies. In this study we investigate an area in the upper Rhone valley around Visp, in the southwestern part of Switzerland. A large dataset of geological and geophysical data, consisting of borehole logs, microtremor horizontal to vertical spectral ratios and shear-wave velocity measurements, was compiled to build a detailed 3D model of the subsurface. By combining fundamental frequency information from noise recordings and shear-wave velocity profiles, three main geophysical discontinuities were identified and their physical properties constrained through a stepwise process. First, the bedrock depth was estimated; in a second step a generic velocity model was defined and finally, combining all the available geological and geophysical information, we developed a 3D geophysical model. The model was compared with a local 3D geological model and a model derived from gravimetric data. The study area is a complex alpine valley where 2D/3D wave propagation phenomena occur. In such case a purely 1D response assumption is considered to be invalid. In order to test the 3D model, we modelled different ambient-vibration wave fields and compared observed and synthetic H/V spectral ratios. We slightly modified our 3D geophysical model in some areas based on this comparison. Finally, a good match between simulated and empirical spectral ratios corroborated the model. The results suggest that the use of ambient vibration techniques are a powerful and cost-effective tools to reconstruct three-dimensional models of the subsurface. Finally, we used the 3D model to predict amplification of earthquake ground motion in the basin.

Introduction

     The Rhone valley in the South-Western part of Switzerland (see inset in Fig. 1) is located in the most active seismic zones of the country [1]. This area was struck by a series of strong seismic events such as the 1755 Brig (Mw 5.7), the 1855 Visp (Mw 6.2) and the 1946 Sierre (Mw 5.8) earthquakes ([[2], [3]]. The shape of the valley and the high velocity contrast between sediments and bedrock make this area susceptible to 2D/3D seismic site amplification effects [[4], [5], [6], [7], [8], [9]]. Due to the ongoing urban development, the Rhone valley is an area of particular interest to study site amplification effects. For this purpose, a detailed knowledge of the geometry, thickness and velocity of the main sedimentary layers in the valley is required. Previous studies have shown that such 3D structures play an important role in seismic wave propagation and amplification (e.g. Refs. [[10], [11], [12]]. In particular, several authors have successfully modelled, with numerical simulations, the amplification in a 3D basin setting [[13], [14], [15]]. Among these for the Rhone valley area Roten et al. [6] developed a 3D geophysical model for the city of Sion in the central Swiss Rhone valley, and successfully compared simulated earthquake ground motion with observations. They quantified the effects of 2D/3D resonances and edge-generated surface waves on the ground motion amplification.

Conclusions

     We characterized the main geophysical discontinuities and their properties in the alpine basin in the Visp area based on ambient vibration data and earthquake recordings integrated with borehole information.

     The ambient vibrations recorded with small aperture seismic arrays were processed by means of the f-k technique to retrieve dispersion curves and to determine 1D VS profile at several locations, whereas HVSR were used to determine f0 related to the sediment thickness. The VS profiles allowed us to highlight the presence of three main layers for which we derived VS values. The two shallowest layers were considered homogenous and with constant velocity, whereas for the thick third layer the effect of sediment compaction was taken into account by introducing an exponential functional form. The VP values were determined based on earthquake P-wave arrival times recorded at the SVISP vertical array. For VS at the bedrock we adopted an approach similar to that proposed by Poggi et al. [55] by extracting from the empirical amplification function for station SVIO a VQWL profile subsequently inverted using a global optimization process.

Highlights

Abstract

Keywords

1. Introduction

2. Numerical analysis models for the existing members and a column retrofitted by wing walls

3. Optimal seismic retrofit method using wing walls

4. Applications to seismic retrofit

5. Conclusion

Acknowledgments

Appendix A.

Appendix B.

References

Abstract

Seismic retrofit of reinforced concrete (RC) columns using wing walls can be used to improve the shear and flexural strength of the column through a relatively simple process. However, the feasibility and efficiency of the seismic retrofitting of RC frames with wing walls heavily depends on the selection of number of columns to be retrofitted, the cross-sectional dimensions of wing walls, and the quantity of re-bars of the wing wall. In this study, an optimal seismic retrofit design method is proposed to minimize not only the initial retrofit cost but also the earthquake-induced damage expected during the life cycle of the building. The seismic performance of structures before and after the application of the retrofit has been verified with the comparison of four response parameters: pushover curves, the inter-storey drift ratios, the energy dissipation capacities, and failure modes. The proposed retrofit method is applied to seismic retrofit of a six-storey RC building example and an actual RC building structure in use. For the retrofit of actual building structure, with an initial retrofit weight of 70.85 kN, which corresponds to 1.85% of the weight of the non-retrofitted building, the energy dissipation capacity was increase by 3.02 times and the life cycle cost (LCC) of the retrofit was reduced to 69.47% of the required LCC for the non-retrofitted building. In addition, it has been confirmed that no storey collapse occurred in collapse prevention level, which indicates the most severe failure mechanism of a structure due to an earthquake.

1. Introduction

During and prior to the 1980 s, when the concept of seismic design had not yet been established in developing countries, buildings were designed without consideration of seismic performance. Accordingly, those buildings tended to suddenly collapse under an external force, such from an earthquake, due to displacement capacity not being considered [1]. In particular, the sudden failure of columns, which are structural members bearing vertical loads, leads to failure to support building weight. This causes the collapse of the entire building and enormous human and structural damages [2–4]. To prevent such damage, various seismic retrofit studies have been conducted on old buildings that have not been seismically retrofitted, and the number of seismically retrofitted buildings is therefore on the rise [5–10].

Seismic retrofit methods, which are practically applicable to the improvement of the seismic performance of existing buildings, can be classified as either strength and stiffness retrofit methods or ductility retrofit methods. Seismic retrofitting of building structures with steel bracings and bearing walls is a representative strength and stiffness retrofit method. Although this method can sufficiently improve seismic performance, a building may still have brittle behavior, which can result in sudden failure [11–13]. On the other hand, ductility retrofit using carbon fiber jackets or steel jackets has a confinement effect on structural members, which partially improves strength and significantly enhances ductility capacity. However, the number of structural members to be retrofitted must be increased, which increases both the working duration and retrofit cost [14–17]. For this reason, many studies have attempted to find the most appropriate retrofit method that can increase strength and ductility capacity satisfactorily

1. Introduction

2. Case study

3. Proposed optimization framework

4. Results of the optimization for the case study

5. Conclusions

References

Steel jacketing (SJ) of beams and columns is widely employed as retrofitting technique to provide additional deformation and strength capacity to existing reinforced concrete (RC) frame structures. The latter are many times designed without considering seismic loads, or present inadequate seismic detailing. The use of SJ is generally associated with non-negligible costs depending on the amount of structural work and non-structural manufacturing and materials. Moreover, this kind of intervention results in noticeable downtime for the building. This paper presents a new optimization framework which is aimed at obtaining minimization of retrofitting costs by optimizing the position and the amount of steel jacketing retrofitting. The proposed methodology is applied to the case study of a 3D RC frame realized in OpenSEES and handled within the framework of a genetic algorithm. The algorithm iterates geometric and mechanical parameters configurations, based on the outcomes of static pushover analysis, in order to match the optimal retrofitting solution, intended as the one minimizing the costs and, at the same time, maintaining a specified safety level. Results of the proposed framework will provide optimized location and amount of steeljacketing reinforcement. It is finally shown that the use of engineering optimization methods can be effectively used to limit retrofitting costs without a substantial modification of structural safety.

Introduction

Steel jacketing is a widely employed technique to improve strength and deformation capacity reinforced concrete (RC) elements presenting critical conditions with respect to seismic and gravity loads condition Steel jacketing of columns is arranged in two possible ways. The first provides a full connection between the steel cage and the slab. In this case, besides the confinement effect, additional flexural strength is provided. For the cases in which full connections are not simple to be realized, steel jacketing is arranged by simply providing the cages. Also in this case a some additional flexural resistance is provided because of friction [1], but the most significant effect is related to the increase of column deformation capacity as consequence of confinement. Experimental and numerical investigations have been carried out by researches for both for the first typology of arrangement (e.g. [2-4]) and for the second one (e.g. [1,6-9]). Despite the effectiveness, it must be also said that steel jacketing is an invasive strengthening technique. In fact, its application to columns provides demolition and reconstitution of portions of masonry infills and plaster. Structural optimization is widely recognized as a valuable computational tool allowing engineers identifying cost-effective design solutions controlling design parameters, and several seismic design optimization applications for steel and RC structures are available in the literature (e.g. [10-17]). At the same time, the issue of the optimization of retrofitting interventions for RC structures has not been investigated many times in the past.

1- Introduction

2- Case studies

3- Global seismic performance-based assessment

4- Fragility and loss assessment

5- Cost-benefit analysis

6- Final remarks

References

Abstract

This study discusses the cost-benefit analysis resulting from the application of traditional seismic retrofitting strategies on four case studies considered representative of both rural and urban stone masonry building stock of Faial Island, in Azores (Portugal). The seismic performance-based assessment was carried out by applying the N2 Method procedure, and the global seismic capacity of each case study estimated by using a three-dimensional model based on the macroelement approach, which combines both the in-plane and out-of-plane response of masonry buildings, to perform nonlinear static analyses. Fragility and loss estimation was evaluated according to the HAZUS methodology. In general terms, the cost-benefit analysis has demonstrated that the studied retrofitting strategies are indeed capable of reducing substantially the seismic vulnerability of the considered case studies, and also that they represent, in fact, a limited amount of the total replacement cost, not compromising, therefore, their economic viability.

Introduction

The recent upsurge on the revitalisation of historical centres in Portugal, is contributing to the mischaracterisation phenomenon of urban cultural heritage (UCH) assets, as traditional construction techniques and materials are massively being replaced by modern solutions, totally disrupting the aesthetic value of existing assets, and often jeopardising the seismic vulnerability of such assets [1]. One of the preferred arguments used to underpin the phenomenon of both architectural and constructive mischaracterisation is that, allegedly, structural renovation works compliant with existing materials and traditional building techniques are not viable from the economic viewpoint. Another often cited argument is that such traditional materials and building techniques do not comply with the regulatory requirements in force in terms of seismic performance. Hence, and as a follow-up of the research published by Maio et al. [1], the current study aims not only at investigating the authenticity of the arguments mentioned above but also to demystify the generalised idea that traditional strategies for the seismic retrofitting of UCH assets have a significant impact over the total renovation cost. In literature, the term “traditional techniques” usually refers to enhancing measures of the buildings’ structural integrity/stability by using compatible and local materials such as earth and wood, and solutions as ring beams, wooden ties interconnecting parallel walls, corner keys, or the addition of buttresses, for example. However, “traditional” has in this study an additional meaning, in the sense that these strategies are quite common and have been widely applied in the framework of the reconstruction process of Faial island (in Azores, Portugal), after the 1998 Azores earthquake [2]. The validation of the former hypotheses is going to be investigated by assessing the Cost-Benefit Analysis (CBA) of two of these strategies. One of the first cost-benefit models for the seismic retrofitting of buildings was issued by the Federal Emergency Management Agency in 1994 [3,4], which encouraged the development of several studies in this topic ever since [5,6]. In the past few years, several compelling studies within this particular research field have been published, focusing either on residential [7,8] or public-school buildings [9,10]. It is worth noting that despite being widely acknowledged worldwide as a significant decision-supporting tool commonly used for evaluating the efficiency of projects, CBA does not provide an absolute answer about whether or not to undertake the seismic retrofitting of a given asset. This because the decision-making process usually depends on many factors beyond the boundaries of benefit-cost analysis, such as the definition of life safety and post-earthquake performance levels [3].

1- Introduction

2- Bridge description

3- Bridge instrumentation

4- Test procedure

5- Field test results

6- Finite element analysis

7- Summary and conclusions

References

Abstract

Many steel bridges in the United States designed before the mid-1980s are highly susceptible to distortion-induced fatigue cracking. This vulnerability is substantially increased if the out-of-plane driving force caused by differential girder displacement is increased for any reason. This research examines one such case where a double-deck bridge complex, originally built in the 1960s, was retrofitted to improve seismic performance. As part of the retrofit, single angle K-type diaphragms were replaced with stiffer double-angle cross-type diaphragms. This seismic retrofit led to an increase in web-gap stresses, and within approximately one year following the retrofit, inspectors identified numerous fatigue cracks in the web of longitudinal girders where connection plates terminate near top flanges. A repair measure was implemented to provide a positive connection between the connection plate and the girder flange. The purpose was to reduce the high stress concentrations in the web-gap region by restricting the out-of-plane distortion in the web-gap region. Field tests were carried out in addition to developing finite element (FE) models to investigate the efficacy of the repair technique. The results confirmed a significant decrease in the web-gap stress after implementation of the repair, and a subsequent FE analysis showed that the new load path through the repair angle section did not introduce a new fatigue sensitive area. In fact, the repair resulted in stresses well below the constant amplitude fatigue threshold (CAFT) for this type of detail.

Introduction

Most continuous-span double-deck viaducts built in the San Francisco Bay Area during the 1950s and 1960s were damaged during the 1989 Loma Prieta earthquake [1]. The deficiencies in these viaducts led to an immediate review of all double-deck bridge structures in the United States. One particular seismic retrofit project (and the focus of this work) involved a Midwestern double-deck bridge complex with substructure and superstructure elements without adequate capacity based on current seismic criteria [2,3]. A seismic retrofit strategy was adopted based on the criteria described in the FHWA Seismic Retrofitting Manual for Highway Bridges [4]. Almost immediately following completion of the seismic retrofit, horseshoe-shaped cracks were identified in the unstiffened regions (frequently called the web gap) of the longitudinal girders where the original K-type diaphragms were replaced with new, stiffer cross-diaphragms. This problem mostly occurs in bridges built in the United States prior to the mid-1980s when the design specifications required connection plates to have a tight-fit detail at the girder flanges, but not welded [5–7]. Normally, part of the connection plate, attached to plate girders, must be clipped to clear the web-to-flange weld resulting in a soft and short section that is more flexible in the transverse direction than the rest of the girder’s height. This area in the web is known as the web-gap region. Cracks originating in this region are likely the result of secondary stresses resulting from out-of-plane distortion in the unstiffened web-gap region and account for approximately 90% of all fatigue cracking [23]. As a result, a significant number of older bridges have been affected by distortion-induced fatigue cracking in the web-gap region.

1. Introduction

2. Mathematical Formulation of the Problem

3. Grid-Characteristic Method

4. Sequence of Nested Hierarchical Grids

5. Dependence of Destructions on the Hypocenter Depth

6. Conclusions

Acknowledgements

References

The purpose of this work was the numerical simulation of dynamic wave processes in heterogeneous media with the subsequent evaluation of their influence on various structures. The effect of earthquakes on ground buildings and underground structures was studied. In the work, we used the grid-characteristic method, which allows taking into account the physics of arising wave processes. The use of hierarchical grid systems allows the calculation of seismic waves directly from the earthquake source to ground-based buildings and underground structures. The work investigated the destruction of high-rise buildings under the influence of seismic loads. The emerging areas of fractures in various structures were studied. The effect of the earthquake on the underground structure was considered. The influence of the depth of the earthquake source was analyzed. Images of wave processes in structures and places of destruction were obtained. Note that we do not calculate the damage caused by falling of some collapsed parts of the building to others, but we calculate only damage caused directly by seismic activity. This type of damage is of the greatest interest, since it is necessary to avoid precisely this primary destruction. Also we investigate the nature of this type of damage, e.g, internal, near the surface, causing collapse, asymmetric. The novelty of the approach used also lies in the fact that we solve a linear elastic wave equation with constant coefficients in various subdomains of the integration domain, between which contact conditions are established. And the size and shape of these subdomains vary over time and depend on the solution. That is, the complete boundary problem of the elastic wave equation is nonlinear.

Introduction

The earthquake is shocks of the ground and earth surface oscillations. Earthquakes reflect the process of geological transformation of the planet. The main cause of earthquakes are global geological and tectonic forces, at present their nature is not entirely clear. Temperature inhomogeneity in bowels of the earth causes the appearance of geological and tectonic forces. Most earthquakes occur on contacts of tectonic plates.

1- Introduction

2- Problem formulation

3- Optimization procedure

4- Numerical study

5- Conclusion

References

Abstract

In this paper an efficient first-order multi-objective optimization scheme is adopted for the design of linear viscous dampers for the seismic retrofitting of frame buildings. A retrofitting cost function serves as one objective while the expected losses serve as the other objective. These two objectives are well understood by decision makers that may not be engineers. Furthermore, with the Pareto front for these two objectives at hand, the decision maker can make his decisions with the whole picture at hand. To allow achieving the Pareto front with a reasonable computational effort, a first-order multi-objective optimization approach is adopted. The gradients of the expected loss function, required for the optimization, are analytically derived using the very efficient Adjoint Variable method. This considerably improves the computational efficiency of the methodology. The efficacy of the framework is illustrated with a 2D four storey frame and an eight-storey 3D asymmetric building.

Introduction

Recent seismic events have shown that, when it comes to modern countries with modern seismic codes, loss of human lives due to collapse of buildings is considerably smaller compared to that in the past. For example, the 1994 Northridge and the 2011 Christchurch earthquakes caused a total of 57 and 185 deaths, respectively. As saving human lives has been the major task of earthquake engineering, this is indeed an achievement for the structural engineering community. However, with the relatively small number of deaths due to these events, came huge monetary losses. The 1994 Northridge and 2011 Christchurch led to losses estimated at ∼ US $44 billion and NZ $40 billion (corresponds to approximately 20% of the GDP), respectively (direct and indirect). This motivates a design considering expected losses in parallel to reducing loss of human lives [2]. Adopting such a design approach is necessary for new buildings, where new technologies to limit damage could be easily included. One such loss based design approach is the Loss Optimization Seismic Design, commonly referred as LOSD [3,4]. This may highly affect the losses to be expected in the far future. Nonetheless, in the foreseen future, most buildings to experience earthquakes are ones that have already been designed and built based on the sole goal of saving human lives. Thus, new technologies and design approaches for their retrofitting are required to reduce expected losses.

A simple formula is proposed that provides interstorey velocities of steel framed structures at specific interstorey drift levels. The proposed equation is statistically derived from the variation of floor relative velocity results along the height of the structure. These results are obtained by non-linear inelastic time history analyses of a number of plane steel moment resisting frames. The proposed equation can then be used for seismic retrofit purposes, i.e., in dimensioning the linear or non-linear viscous dampers to be inserted in a steel frame. A detailed numerical example is provided and conclusions regarding the accuracy and the potential use of interstorey velocity are drawn.

Introduction

The variation (distribution) of maximum interstorey velocities have been recognized as key parameter for the evaluation of the along-theheight effectiveness of and demand for viscous dampers [1]. This variation (distribution) seems to depend mainly on the modes (number of stories) of the structure under consideration. In building codes, e.g., ASCE 7–10 [2], peak (design) interstorey velocity (IV) is approximately calculated in the fundamental and higher modes using the design (maximum) interstorey drift (IDR). This approximation permits the calculation of the maximum damping force in terms of the design IDR. Moreover, building codes, e.g., ASCE 7–10 [2], seem to accept that the most effective way of allocation of viscous dampers is to place them where large IDRs are exhibited. Therefore, the maximum IDR is used to estimate the maximum IV, without, however, providing an insight about the real relationship of maximum IDR and IV values. In literature so far, the relationship between IDR and IV is formulated either by using simplified building models and the first mode shape [3,4] or by employing single-degree-of-freedom systems and their maximum displacement [5,6]. The importance of having reliable estimates of the true IV is initially stressed by Pekcan et al. [7] in view of the operating velocities of non-linear viscous dampers and very recently by Favvata [8] in the context of seismic pounding of adjacent structures. It is the purpose of this paper to establish a relationship between IDR and IV values for steel moment resisting frames (MRFs). More specifically, the seismic responses in terms of floor relative displacements and relative velocities are obtained by non-linear inelastic seismic analyses of 20 plane steel MRFs under 22 real and as recorded seismic motions (accelerograms). Then, for each frame-accelerogram pair, height-wise variations of IDR and IV are constructed. Considering a specific value for IDR along-height, i.e., 1.5%, the corresponding IV values are maintained. Statistical processing of these IV values permits the derivation of an equation that provides IV along-the-height of a structure for this specific IDR. The proposed equation is then used in a seismic retrofit scheme, where dimensioning of viscous dampers is based on IV and targets a specific IDR range. An example that involves the retrofit of a 12-storey with 4-bays, steel MRF is presented. This steel MRF is retrofitted either with linear viscous dampers or with non-linear ones. For each kind of dampers, two cases of target IDR are considered. The retrofitted MRFs are then subjected to non-linear inelastic time history analyses and IDR, IV and damper forces are computed. On the basis of the numerical results of the present work, it can be concluded that the proposed IV equation is quite effective in satisfying target IDR. It is also demonstrated that for a specific level of seismic demand (in terms of mean acceleration spectra) the proposed IV equation offers controlled IV and IDR values as well as damper forces under the condition that plastic hinge formations to columns due to additional axial forces induced by the damper forces are within acceptable limits.