Abstract

     Concrete is one of the most widely used materials in various civil engineering applications. Its global production rate is increasing to meet demand. Mechanical properties of concrete are among important parameters in designing and evaluating its performance. Over the past few decades, machine learning has been used to model real-world problems. Machine learning, as a branch of artificial intelligence, is gaining popularity in many scientific fields such as robotics, statistics, bioinformatics, computer science, and construction materials. Machine learning has many advantages over statistical and experimental models, such as optimal accuracy, high-performance speed, responsiveness in complex environments, and economic cost-effectiveness. Recently, more researchers are looking into deep learning, which is a group of machine learning algorithms, as a powerful method in matters of diagnosis and classification. Hence, this paper provides a review of successful ML and DL model applications to predict concrete mechanical properties. Several modeling algorithms were reviewed highlighting their applications, performance, current knowledge gaps, and suggestions for future research. This paper will assist construction material engineers and researchers in selecting suitable and accurate techniques that fit their applications.

Introduction

     Concrete is the most widely used building material worldwide. With population growth and urbanization, the demand for concrete is expected to reach 18 billion by 2050 [[1], [2], [3]]. In order to improve the design of concrete structures, it is necessary to gain a better understanding of concrete performance, relying on accurate evaluation of its mechanical properties. Among the various properties of concrete, compressive strength has been considered a direct indicator of performance. It is directly related to the safety and performance of the structure throughout its life cycle [1,4,5].

     Concrete is a complex system of combinations of different components (coarse and fine aggregates, water, cement, and additional mixtures) that are randomly distributed throughout the concrete matrix [[6], [7], [8], [9]]. This heterogeneous feature makes it difficult to accurately predict certain mechanical properties, especially compressive strength [[10], [11], [12]]. The most direct way to evaluate the compressive strength of concrete is through physical tests performed on specimens cured to the desired age [13,14]. Such a method for evaluating the compressive strength needs time while being affected by other factors related to specimen fabrications and test operators. Moreover, the test tends to damage the specimens. Empirical regression methods were therefore proposed to predict compressive strength [[15], [16], [17]], but the disadvantage of this method is the non-linear relationship between the concrete mixture and the concrete’s compressive strength. This prevents an accurate regression expression. Numerical simulation is another method that can predict the behavior of concrete. However, a good prediction of concrete behavior is not easy to achieve due to the non-linearity and randomness [[18], [19], [20]].

Conclusions and recommendations

     The present study investigated the performance of different ML/DL models to predict the mechanical properties of concrete. Among the wide range of ML/DL models, support vector machines, decision trees, evolutionary algorithms, and artificial neural networks were examined due to the popularity of these models as well as their frequent use in the field of civil engineering. Therefore, the performance of these models in various studies to evaluate the compressive strength, tensile strength, shear strength, flexural strength, and elastic modulus was evaluated. A comparison of experimental models with ML/DL methods shows that ML/DL models with better updating capability and the ability to analyze large datasets perform better. On the other hand, statistical models are not good for predicting complex structures due to their high cost and time-consuming nature. According to ML/DL methods, selecting the appropriate model for predicting the target strength of concrete should be made by considering different criteria. A review of various studies shows that the relationship between concrete components and mechanical strength is influential in choosing the forecasting model. Therefore, models that can respond in nonlinear space should be used if the relationship is nonlinear. In these cases, SVM and ANN models can be used for their acceptable performance in a non-linear environment with fewer errors. To achieve more accurate results, optimization of these models with metaheuristic algorithms can also be used. But if we need the transparency of the model and the explicit mathematical formula between input and output, we can use decision tree models and evolutionary algorithms. On the other hand, the use of ensemble models to optimize these models, although it results in higher accuracy, increases computation time and model complexity. The combined SVM and ANN models, although they increase the computation time, have accurate results in the face of extensive data. Therefore, based on the study, the best option for predicting the strength of concrete in terms of model accuracy and model implementation is the combined use of SVM and ANN combined models.

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

     As a key component in composite beams, the studs are prone to fatigue failure under reciprocating load. Fatigue problem is more prominent especially under circumstances of concrete cracking, stud corrosion or with initial damage. Therefore, static and fatigue tests are carried out to study the fatigue properties of push-out specimens with initial defects, including stud damage and concrete cracking. The test results show that the damage of stud has a great influence on fatigue life, while the crack of concrete has a little influence on fatigue life. The damage of stud has a greater influence on fatigue performances compared to static properties. In addition, based on the dissipative energy theory and experimental phenomenon, the roles of stud and concrete in the process of stud fatigue failure and their interaction are summarized. The dissipative energy mainly includes the shear plastic strain energy of the stud, the plastic strain energy of concrete under compression after the stud, and the heat energy generated by the friction between the stud and concrete.

Introduction

     Steel-concrete composite beams have become popular in recent years for use in bridge engineering [1,2]. Under the complicated environment and load, there may be some damage to the composite beam. There might be some damage, such as corrosion, fractures of weld leg and longitudinal or transverse cracks in the concrete deck. The static and fatigue behavior of stud connection is sophisticated and affects the design resistance and stiffness of composite beams [3,4]. The damage of the stud and concrete may have a great influence on the shear fatigue performance of the interface.

      Welding defects significantly lower the shear capacity of studs, as is widely known. There are many types of welding defects that might appear, such as inadequate penetration, absence of fusion, and gas inclusion [5]. To determine how the damage location and degree affected the static behavior and shear capacity of stud shear connections, six specimens with identical geometrical dimensions were investigated by Qi et al [6]. The test findings showed that, in comparison to the shear capacity of normal specimen, a loss of up to 36.6 % and 62.9 % of the shank area might cause a falling shear capacity of 7.9 % and 57.2 %, respectively. The push-out test was simulated using mathematics, and the results of the test were used to validate the study. It was shown that even though the area of the stud had significantly decreased, the shear capacity was not affected by the severity of the damage when the damaged part was placed 0.5d (where d is the shank diameter) from the stud root, based on the numerical model.

Conclusions

     Static and fatigue tests were carried out on studs in steel–concrete composite study the effects of initial damage in stud and concrete on fatigue performance. The following conclusions are drawn:

     1.  Compared with nondestructive specimens, the fatigue life of transverse crack and longitudinal crack specimens were reduced by 22 % and 4.9 %, respectively, which is determined by the stress characteristics of the stud. Transverse crack makes the stud easily in a state of bending and shear composite, while longitudinal crack does not

     2.  Compared with the preset cracks of concrete, the damage of stud and weld collar had more influence on fatigue performance, in which the loss of stud section of 62.9 % leaded to only about 10 % of fatigue life. The initial damage of stud accelerates the damage accumulation in the stable fatigue development stage, while the concrete cracking leads to the increase of residual slip in the early stage.

     3. The existing calculation methods of residual slip are not completely applicable to the test results of different stresses, and most of the calculation results are too small, which requires some modification of their coefficients.

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

     An increasing number of offshore wind farms are being constructed in seismic regions over liquefaction susceptible soils. This paper presents a methodology for the analysis and design of monopiles in seismically liquefiable soils by extending the established “10-step methodology” with an additional 7 steps. These additional steps include assimilation of seismic data, site response analysis, stability check of the structure (ULS check through the concept of load-utilization ratio), input motion selection, prediction of permanent tilt/rotation, and ground settlement post liquefaction. A flow chart, which shows the interdependence of the different disciplines, is presented and can be extended to routine design. This proposed method is validated using the observed performance of an offshore and nearshore turbine from the Kamisu wind farm during the 2011 Great East Japan earthquake. Predicted results based on the proposed methodology compare well with the field observation and demarcate the (i) good overall performance of the offshore turbines and (ii) limit state exceedance of the nearshore turbine. It is envisaged that the proposed method will be useful towards the design of monopiles-supported wind turbines in seismic areas.

Introduction

     Towards the end of 2020, the global wind energy generation capacity amounted to 733 GW, with new installations accounting for 111 GW, almost doubling that of 2019 [1]. Due to higher efficiencies and more stable wind conditions offshore than onshore sites, an increasing proportion of wind power is produced through large offshore wind farms. This demand for cost-efficient wind energy production has facilitated the industry to develop larger turbines with higher capacities. Fig. 1 presents the evolution of turbine capacity over the last 16 years, color-coded with rotor diameter. However, due to the high capital investments involved, profitability of these wind farms requires continuous operation and immediate functionality post-natural hazards. Therefore, the seismic resilience of OWTs is an important consideration in their engineering design and is, therefore, an area of active research. Readers are referred to recent publications [[2], [3], [4], [5], [6], [7], [8], [9]] on various aspects of seismic design.

Discussion and conclusions

     An existing framework for the foundation design of monopile-supported wind turbines was extended to include seismic load considerations in liquefiable and non-liquefiable soil. The framework was validated using a case study from Wind Power Kamisu, a near-shore farm that performed well during the 2011 Tohoku earthquake and subsequent tsunami.

The discussion and conclusions are presented succinctly below:

     1. A quantitative appraisal of appropriate damping, as recommended by Lombardi and Bhattacharya [58] and Adhikari and Bhattacharya [57], could vary the structural demands and therefore needs proper consideration. Back analysis of a case study from the Kamisu wind farm further indicated that structural demands could vary if suitable damping models are not selected.

      2. Definition of mechanism-specific p-y spring models for liquefiable and non-liquefiable soil is necessary as their selection could underpredict or overpredict structural response (deformation, tilt, RNA acceleration). As near-level ground conditions prevailed in this study, the authors employed hyperelastic (strain hardening) p-y curves for liquefiable soil to account for the increased resistance attributed to shear strain-induced soil dilation. During shear strain excursions, such dilative tendencies can transfer high acceleration pulses to the RNA and therefore need to be accounted for in the design stage.

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.

Abstract

     Determining the relationship between the cause of damage and the subsequent structural behavior of infrastructure systems requires an accurate characterization of the propagation of cracks, which represents the evolution of the damage state. When no information about the cause of damage is available, kinematic approaches can be used to describe the motion of crack contours. Current image-based approaches to derive crack kinematics use digital image correlation (DIC) on a set of sequential images as the crack propagates. However, DIC is invasive in that the structure surfaces must be painted with random speckle patterns, limiting its use primarily to controlled experiments. In this paper, we propose a novel image-based methodology for computing crack opening in Mode I or Mode II. As an input, this method takes a binary image from a semantic segmentation of an image of a crack pattern. This binary image is used to detect the opposite edges along the crack, which are then registered using an optimization algorithm based on the Euclidean transformation model and non-linear least squares. As a final output, this method produces displacement maps in the tangential and normal directions to the crack skeleton. To demonstrate its performance, we validate our methodology first with synthetic crack patterns and then with real crack patterns. Because this methodology for determining crack openings requires only simple data (just a binary crack pattern image), it is straightforward, robust, and adaptable, thus contributing to the development of structural image-based damage assessments. The computational codes and datasets are available to the public for future research and benchmarking on https://github.com/eesd-epfl/crack_kinematics and https://doi.org/10.5281/zenodo.6632071.

Introduction

     Structural inspections of existing buildings and infrastructure comprise localizing and classifying damage features in the structure. These engineering inspections are conducted on a regular basis, as is common for infrastructure, as well as after extraordinary events, such as earthquakes. Among the various manifestations of damage that engineers evaluate, cracks are the most common feature in quasi-brittle structures, such as concrete and masonry [1].

     The likely causes of damage and the mechanical properties of the damaged structural element (e.g., stiffness, strength) can be estimated from the crack pattern and kinematics of a crack. Crack kinematics describes how two crack surfaces move in relation to each other, and provide important information about the stress state that caused the damage. A Mode I crack opens normal to the crack surface and is typically associated with normal stresses. A Mode II crack opens parallel to the crack surface and is associated with shear stresses in the plane of the element [2]. Shear stresses acting in the crack plane cause Mode III deformation as well, but in this case the deformation is out-of-plane [2]. Concrete and masonry elements that fail in shear (Mode II) have a lower deformation capacity than elements that fail in flexure (Mode I). Therefore, identifying cracks in structures and classifying their opening mode is important for preventing potential failures through timely interventions designed to minimize economic losses, future deterioration, and even loss of human life [3].

Conclusions

     In this paper, we proposed a novel methodology for determining the propagation of cracks in structural elements caused by unknown forces (crack kinematics). This procedure takes as inputs binary images obtained by semantically segmenting a crack from an image of a crack pattern. To the best of our knowledge, this is the first method to solve the crack kinematics determination problem based on single images, significantly advancing the state-of-the-art in the field.

     The first step of our developed technique consists of detecting the contours of a crack pattern as a 2D set of points that are divided into opposite edges along the crack length. Next, the edges are registered using Euclidean transformations encoding both normal and tangential displacements, which are used to determine whether the cracks propagated in Mode I or Mode II. Additional normal features are included in our loss function to improve the results of the registration problem. We validated our method with synthetic data based on lines and real crack contours, where pre-defined rotation and translation were used to generate crack patterns, obtaining an absolute error of less than a pixel. Monte Carlo simulations were used to demonstrate the stability of our method as well as to guide the selection of hyper-parameters. Then, we put our methodology to the test with challenging crack patterns that contained multiple cracks and branches, including two real structural applications (damaged wall during laboratory testing, damaged beam and damaged building due to an earthquake event). Extra validation of the performance, accuracy and robustness of our methodology is made comparing results with a DIC-based method and manual measurements with metric gauge. The results of both validation and test experiments showed that the developed approach has the potential to deduce the crack kinematics simply from spatial features (2D edges coordinates) extracted from a binary input image.

Abstract

     Frost and thawing damage asphalt in cold climates. Water that enters the pores of asphalt at low temperatures may accelerate its degradation. Water can expand into pores, altering the void content and shape, thus affecting the asphalt’s internal structure and fracture properties. We can develop more durable asphalt types if we understand how void topology changes with freeze–thaw cycles. The purpose of this study is to establish a correlation between the mechanical properties of asphalts and the topological properties of voids. To determine their internal structure, various asphalt types representing dense asphalt and asphalt with voids were made and X-rayed before and after each freeze–thaw cycle. We also obtained the mechanical properties of asphalt and correlated them with the void properties. It was found that dense asphalt has the lowest degradation rate in wet conditions characterised by non-connected gaps, which was approximately constant with freeze–thaw cycles; however, dense asphalt was least durable under dry conditions compared with asphalts with more voids. Due to its high water retention rate, asphalt with a 10% void content degraded at an accelerated rate during the initial cycles. As a result of bigger voids, asphalt with a higher pore content plateaued in later cycles due to reduced water retention.

Introduction

     Asphalt mixture is a heterogeneous material made of aggregates, mineral filler, and a bituminous binder. It is compacted at temperatures ranging between 135 °C and 155 °C; asphalt mixtures may also include voids, up to more than 20 % of the total asphalt’s volume, depending on the gradation and compaction methodology. Asphalt is the most used material to build pavement surfaces. Asphalt undergoes different mechanical loads and environmental conditions during its lifetime, leading to crack and damage and reducing its lifespan [1], [2]. In particular, the extreme environmental condition in cold regions with large temperature fluctuations [3] and the water permeation are some of the most harmful deterioration mechanisms that may affect asphalt, leading to thermal stress [4], loss of adhesion at the interface between aggregate and bitumen [5], and loss of cohesion of the binder [6].

Conclusions

     Several types of asphalt with acceptable air void contents ranging from 3% to 14% were examined for their influence during freeze–thaw cycles, where a correlation between differences in air void topology and changes in mechanical properties was established. According to the results in this study, the topological properties evolution under freeze–thaw cycles was influenced by three different phenomena: (1) volume increase of the existing voids and formation of small cracks, (2) voids coalescence, and (3) cracking and new voids formation. The voids were analysed using X-ray CT scans after every freeze–thaw cycle, and the mechanical properties were measured from 3-point bending tests. The following conclusions were obtained:

     The three different asphalts had different behaviours when exposed to freeze–thaw cycles, attributed to the different aggregate size distributions and air void topologies. Hence, the durability of asphalt to freeze–thaw cycles could be controlled by carefully selecting the aggregate gradation, aggregate morphology and air void topologies.

Abstract

Keywords

1. Introduction

2. Site characterization

3. Methods

4. Results

5. Conclusions

Declaration of competing interest

Acknowledgments

Nomenclature

References

Vitae

Abstract

Characterizing the permeability variation in fractured rocks is important in various subsurface applications, but how the permeability evolves in the foundation rocks of high dams during operation remains poorly understood. This permeability change is commonly evidenced by a continuous decrease in the amount of discharge (especially for dams on sediment-laden rivers), and can be attributed to fracture clogging and/or hydromechanical coupling. In this study, the permeability evolution of fractured rocks at a high arch dam foundation during operation was evaluated by inverse modeling based on the field time-series data of both pore pressure and discharge. A procedure combining orthogonal design, transient flow modeling, artificial neural network, and genetic algorithm was adopted to efficiently estimate the hydraulic conductivity values in each annual cycle after initial reservoir filling. The inverse results show that the permeability of the dam foundation rocks follows an exponential decay annually during operation (i.e. K/K0 = 0.97e−0.59t + 0.03), with good agreement between field observations and numerical simulations. The significance of the obtained permeability decay function was manifested by an assessment of the long-term seepage control performance and groundwater flow behaviors at the dam site. The proposed formula is also of merit for characterizing the permeability change in riverbed rocks induced by sediment transport and deposition.

1. Introduction

Characterizing the permeability of fractured rock formations is crucial to various subsurface applications, such as groundwater modeling (Chen et al., 2016a), oil and gas production (Mohamadi-Baghmolaei et al., 2016), thermal energy extraction (Sun et al., 2017), contaminant transport simulation (Carrera, 1993), hyporheic exchange characterization (Rozemeijer et al., 2010; Barlow and Coupe, 2012; Kiel and Cardenas, 2014), and optimization design of impervious barriers in dam engineering (Li et al., 2014, 2017; Chen et al., 2015). It has been well understood that the permeability of rocks is determined by the geometries (e.g. size, roughness and interconnectivity) of the void space consisting of pores and fractures through which the fluid transmits. The permeability could hence be highly anisotropic, heterogeneous, and scale-dependent (Snow, 1969), and varies with time as the flow geometries change. This could be induced by various processes such as rock deformation (Chen et al., 2007), damage (Chen et al., 2014), erosion (Sadhukhan et al., 2007), and particle clogging (Nowinski et al., 2011).

A comprehensive review of the statistical experimental optimization problem concerning the mixture design of various cement-based materials is presented herein. This review summarizes and discusses over 80 applications of optimum design regarding the basic test information under response surface method (RSM), including influence factor and corresponding response, statistical method, and coefficient of determination. The statistical experimental design reported in previous studies has shown that RSM is a sequential procedure to provide a suitable approximation for the mixture optimization. Then, linear, quadratic and interactive relationships of the statistical model can be evaluated available. Especially, the multi-objective optimization issues with multiple or competing performance requirements for various cement-based materials have also been reported, by considering fluidity, strength development, environmental impact, cost and durability. Overall, the results from existing publications have demonstrated that statistical inference and analysis of variance (ANOVA) are suitable for mix proportion design and process optimization of cement-based materials. The W/B ratio and mixture components are the prevalent factors in experimental design optimization, and then the fluidity and strength as the most popularly used response. Thus, theoretical optimum mixture proportioning can be used to predict valuable fresh and hardened properties. Finally, a critical discussion of the selection of design strategy, independent factors and their responses, and the experimental region involved in statistical experimental design, is provided. Based on this review, we conclude that the multi-objective optimization approaches need a further systematic study, and further studies of sustainable concrete optimization are needed by comparing the different chemical composition and particle characteristics.

1. Introduction

The cement-based materials are prepared by using various types and quantities of individual constituents. These mixture proportions play an important role in fresh- and hardened-state performance, such as fluidity, rheological properties, strength development and durability. Therefore, many research studies have been dedicated to experimental optimization of cement and concrete mixtures.

1- Introduction

2- Prototype (code-designed) RC building

3- Benchmark for computational modeling of RC structures under fire with heating and cooling phases

4- Behavior of the prototype building under fire

5- Computational model for a column part of a structure

6- Parametric study of the residual deformations of RC columns

7- Conclusion

References

Abstract

Reinforced concrete (RC) structures often remain stable under fire, but exhibit damage and residual deformations which require repairs. While repair operations and building downtime are expensive, current fire design approaches do not consider post-event resilience. The first step to enable predicting the resilience of RC structures under fire is to develop capabilities to model the damage of these structures after various fire exposures. This paper focuses on the prediction of the residual (post-fire) deformations of RC columns within a code-designed five-story RC frame building. Computational modeling approaches to capture the fire behavior of the columns are investigated. The models range from isolated columns with linear springs at the boundaries to full building model coupling beam and shell elements, with intermediate approaches. The analyses highlight the critical nonlinear role of the thermal expansion-contraction of the surrounding beams and slabs on the column deformations. Large transversal residual deformations develop particularly in perimeter columns, combined with residual shortening. This invalidates models based on isolated column or 2D frame. A parametric study of the residual deformations of RC columns is then conducted, with due consideration of the 3D restraints and interactions, to investigate the effects of different design parameters and fire scenarios on the residual deformations after a fire event. The results of the parametric study indicate that fire load density and opening factor significantly influence the residual deformations of RC columns, compared to the thermal conductivity of concrete and live loads. This research improves the understanding and provides recommendations for numerical modeling of the effect of fire on the residual capacity and deformations in RC structures.

Introduction

Reinforced concrete (RC) buildings generally exhibit a good structural performance under accidental fire events, as seen in, for instance, the 2005 Windsor Tower fire or the 2017 Grenfell Tower fire. In the two latter events, no global structural collapse of the concrete structure occurred, despite fires raging for hours. Yet, while fire does often not result in global collapse of RC structures, the potential loss due to downtime and repairs may be significant. In many instances, the fire accident is not as severe as in the two aforementioned cases, and a rehabilitation is possible [1,2]. The question of post-fire damage and downtime cost has gained increasing attention due to the requirements for resilience of structures under hazards. To develop optimum provisions to design fire resilient structures, engineers need the ability to accurately estimate the potential economic loss for different design alternatives due to fire damage. This in turn requires the ability to predict the behavior of RC structures under fire, including the residual deformations and residual load-bearing capacity of a structure after fire. Structural members are often tested as independent elements under fire, disregarding global behavior. However, the proper inclusion of structural continuity of fire induced effects is crucial for an accurate evaluation of RC building’s response. The heating of structural members leads to thermal expansion, which may cause the surrounding structure to impose high restraint forces. The significant impact of boundary conditions on the fire behavior of RC structures under fire has been observed in many historical accidents and previous research. In the Katrantzos Department Store fire (an eight-story RC building) in Greece in 1980, the restraints from differential thermal expansion in the structure led to the collapse of a major part of the 5th to 8th floors and the failures of various other floors and columns throughout the building [3]. Extensive research works have demonstrated the considerable effect of restraints on the behavior of RC structural members under fire, based on numerical or experimental studies of RC columns, beams, slabs and walls [4–11].