Journal of Structural Engineering

Relative displacement, which is displacement of a point on a structure with respect to its original location or an adjacent point on the structure that has also undergone movement, can be an effective indicator of postevent structural damage. Although quantifying relative deformations for postevent damage quantification is currently technically feasible, and of great potential benefit, the field remains undeveloped because of lack of knowledge among structural engineers of (1) the enabling technologies and (2) the performance limits of these technologies. The primary objectives of this paper are to review available techniques for measuring relative deformations, identify their limitations, and propose areas where further research is needed. Current methodologies are divided into remote and in situ methods. The latter techniques are deemed most practically feasible and are surveyed in detail and critiqued. Suggestions for current challenges and research opportunities are proposed with emphasis on accuracy considerations, the need for creating a national database of structure information, and methods for large-scale automated assessment.

The debonding between the steel tube and the confined concrete core of concrete-filled steel tubes (CFSTs) can weaken the confinement effect of the steel tube on the concrete core and may induce a decrease in the load-carrying capacity and in the ductility. The debonding detection technique for CFSTs is still a challenging problem resulting from the inaccessibility and invisibility of the interface. In this paper, by embedding piezoelectric lead zirconate titanate (PZT) based functional smart aggregates (SAs) in concrete core as actuators and pasting PZT patches on the predetermined locations of the outer surfaces of the specimen as sensors, a novel active interface condition monitoring approach for rectangular CFST members is proposed and is experimentally validated with a scale rectangular CFST specimen with artificial debonding defects. Analysis of the wavelet packet energy spectrum (WPES) of the measurements of PZT patches is carried out, and a damage index, called the weighted variation of WPES (WVWPES) is defined to detect the artificial debonding regions. The results show that the proposed WVWPES index is sensitive to the debonding defect. Finally, the proposed approach is applied to evaluate the interface condition of a large rectangular CFST column of a super high-rise building.

Operational modal analysis is a challenging task to deal with output-only vibration measurements contaminated by noise. This paper proposes a new method for operational modal identification of a linear system using continuous-wavelet transmissibility (CWTR) to make full use of the advantages of operational transmissibility measurements and wavelet transform. A new theorem on limits is mathematically proved, for which the proposed CWTR at different scales are independent of stationary excitations acting on the structure at the system poles. With such a unique feature, the operational modal frequencies and mode shapes can be extracted by combing different CWTRs at numerous wavelet scales with different transferring outputs. The applicability of the method is numerically verified by a 4-story frame subjected to random forces. The effects of wavelet functions and scale discretion step have been investigated. The real case application of a concrete-filled steel tubular arch bridge tested in the field under operational conditions further illustrates that the operational modal parameters, identified by the present technique, agree well with those obtained from the existing identification methods and calculated by finite-element analysis. It is demonstrated that the proposed CWTRs are capable of identifying the operational modal parameters of full-sized structures.

A low- and high-velocity based controller (LH controller) is developed to model the force-velocity relationship of a magnetorheological (MR) fluid damper that enables the controlling current to be calculated as a function of required control force. The proposed bilinear force-velocity model predicts the high-velocity region as well as the often neglected low-velocity region that corresponds to the damper responses in the postyield and preyield regions, respectively. The calculated current may then be supplied to an experimentally parameterized prototype small-capacity MR damper (50 kN) to control seismic responses at a lower cost. Numerical examples are used to demonstrate the ability of the LH controller to reduce structural displacements and accelerations in comparison with three other state-of-the-art controllers, namely, the clipped optimal controller, modified clipped optimal controller, and inverse model, which neglect the low-velocity region of the damper. After including this region, the LH controller significantly lowers peak accelerations by 57.7, 33.4, and 13.0%, respectively, over the three other common controllers and also lowers mean displacements by 85.9, 81.7, and 85.4%, respectively. Lastly, the LH controller is used to reduce the peak drifts and accelerations in comparison with those of a 20-story benchmark building.

Connections with glued-in rods are very efficient, high-strength joints and can be successfully used, for example, for the anchorage of vertical timber elements to a foundation. Although extensive research was done on the pullout strength of single rods, few references can be found on the behavior of multiple glued-in rods. This paper presents the results of an experimental program carried out on timber-to-steel connections with multiple glued-in rods. Two series of 10 specimens with four large-diameter, glued-in rods made from mild steel (Series GB) and 10 specimens with 12 small-diameter, high-grade glued-in rods (Series GS) were tested to failure under monotonic tensile load. Both types of connections ultimately failed in a brittle manner; however, the GB series developed some plastic deformations prior to failure which was caused by the pullout of two rods and longitudinal splitting of the timber member. By comparing the experimental results with the analytical predictions, no significant group effect could be detected in the GB test series, and an overall acceptable approximation was found. In the GS test series, no final conclusion on the group effect could be drawn as several rods failed by nut-stripping. A parametric study using analytical formulas was carried out to investigate the effect of rod strength, rod diameter, glued length, and timber density on the connection performance. Based on the results of the parametric study and experimental tests, it can be concluded that the use of mild steel as well as more rods of larger diameter, properly spaced from each other, and from the edges are all effective measures to increase the connection ductility. Furthermore, by leaving a suitable length of rod unbonded between the steel plate and the glued length, it is possible to control ductility, providing all brittle failure modes have a higher strength capacity than the rod-yielding force. Special care should be taken during the manufacturing process to ensure good quality of gluing (required glued length and amount of adhesive) and rod alignment.

This paper presents probabilistic models for the structural performance of rounded dovetail joints. The models are developed with a Bayesian technique, which implies that the model uncertainty is explicitly characterized by random variables. The Bayesian approach also promotes model updating when new test results become available in the future. Practical insight is gained from the modeling process, which includes a novel search for influential parameters, and from the subsequent probabilistic analysis with the models. The models are based on 80 tests of single and double dovetail joints with varying geometric parameters, specifically the flange angle and the dovetail height. A significant effort was made to record variables that conceivably influence the performance of this type of joint, including a series of material parameters: tension strength perpendicular to grain, shear strength parallel to grain, moisture content, density, growth ring density, and growth ring orientation. This paper explores the significance of each parameter and proposes models that include the most significant parameters while retaining a measure of the model uncertainty. In contrast to most models used in structural design, the probabilistic models presented herein are unbiased and suitable for future reliability-based calibration of code equations.

Under extreme earthquake loading, light-frame wood building collapse is often caused by excessive interstory drifts at one or more story levels, leading to catastrophic $P\text{-}\mathrm{\Delta }$ failure once the shear walls, and subsequently the entire structure, become unstable. This soft-story collapse mechanism has been observed in numerous earthquakes. Current performance-based seismic design methods for light-frame wood buildings typically uses very conservative drift levels to represent the near collapse deformation of wood frame building systems, such as the 3% drift limit used in ACSE Standard 41 corresponding to a collapse prevention performance target. A series of full-scale collapse loading tests on wood shear walls was conducted in this study to identify the ultimate drift level at which $P\text{-}\mathrm{\Delta }$ collapse will occur and is described in this paper. It was concluded that laterally braced wood shear walls can remain stable up to $\sim 7–10\%$ interstory drift, depending on the magnitude of the vertical loading. If one considers the typical design range of gravity load commonly seen in multistory light-frame wood construction, assigning a single ultimate drift level of 7% for light-frame wood building collapse limits appears to be justifiable based on the experimental results presented herein.

Several structural collapse incidents indicate that failure usually starts from beam-column joints exposed to abnormal loads, especially for steel and composite structures. If the connections are sufficiently robust and there is adequate axial restraint from adjoining structures, catenary action forms in the beams and slabs, causing alternate load paths when affected columns are severely damaged, and resulting in large deformations in the beams and slabs. This paper presents experimental results of bolted-angle beam-column joints under a middle column–removal scenario. Three types of connections—including (1) web cleat, (2) top and seat angle, and (3) top and seat with web angle connections—were investigated, and three angle thicknesses (8, 10, and 12 mm) were tested. The results of the nine experimental tests conducted demonstrate the ductility and load-carrying capacities of these three connection types with different angle thicknesses in catenary action mode. When the angle thickness increases, the failure mode changes from angle fracture to bolt fracture. A component-based model is also developed to predict the behavior of bolted-angle beam-column joints up to total failure. The validation study indicates that the proposed models can represent the key responses of bolted-angle beam-column joints under a middle column–removal scenario, including the formation of flexural action at small deformation stage, the development of catenary action at large deformation stage, and fractures of the connection components at the last stage. Based on the proposed joint model, frame analyses are conducted. The analytical results of the frame models demonstrate that the conducted joint tests could represent the behavior of prototype steel frames against progressive collapse. The effect of horizontal restraint stiffness, which is critical to the development of catenary action, is also investigated.

Probabilistic seismic demand models are developed for bridges elevated with steel pedestals by adding correction and error terms to commonly used models. Separate probabilistic demand models are developed for the force demand on steel pedestals and the shear and deformation demands on concrete columns. Nonlinear time history analyses on detailed, three-dimensional finite-element models are used to generate virtual experimental data. By applying a Bayesian updating method to the generated data, parameters of the probabilistic models and their correlations are estimated. Comparisons between the demands from the developed probabilistic demand models and the demands from their corresponding demand models without correction and error terms reveal that the developed probabilistic models provide more accurate and unbiased predictions of the demands of interest. As an illustration of the developed framework, fragilities are estimated for a two-span bridge. The results show that pedestals are more vulnerable in the longitudinal direction, and columns are more vulnerable in the transverse direction. A sensitivity analysis on the studied bridges shows that decreasing the pedestal height, increasing the length of the pedestal anchor bolts within the concrete bent, and increasing the concrete cover on the anchor bolts are the most effective ways to decrease the probability of failure.

A general methodology for probabilistic corrosion analysis of reinforcing bar in RC bridges is proposed in this paper. Uncertainties due to limited number of experimental data, incomplete inspection information, as well as the intrinsic randomness of random variables affect the prognostics of corrosion damage. The proposed study includes both fuzziness and randomness to consider different types of uncertainties. First, the chloride-induced corrosion initiation and propagation model are developed for reinforcement in RC bridge. The relationship between the area corrosion rate and the yield strength degradation is proposed on the basis of experimental investigation from accelerated corrosion testing. Following this, the randomness and fuzziness are included using fuzzy random variables to consider the uncertainties of the degradation under corrosive environments. The probabilistic modeling of the mean and the SD of reinforcement yield strength is discussed in detail. Finally, the proposed methodology is illustrated and validated with field measurements of corroded reinforcement in RC beams from a 34-year-old decommissioned RC bridge. The predicted results agree well with experimental values.

Systems of parallel beams connected with springs appear often in engineering. Typical examples are bundles of wires, such as the ones appearing in suspension bridge cables. These connections result in coupling of the differential equations describing the problem. In this paper, two systems of parallel beams connected with springs perpendicular to the beams are examined: multiple beams connected with springs, and multiple beams of common stiffness connected with springs and subjected to a common tensile axial load. A method is proposed to decouple the system of coupled equations into uncoupled differential equations for which exact shape functions can be constructed, so that the computational effort associated with systems involving a very large number of beams can become manageable. Numerical examples are presented demonstrating the capabilities of the proposed method.

When designed according to current special moment frame requirements, conventional RC beam-column joints in frame-braced buildings subject to design level seismic loads suffer severe damage. Construction of these conventional beam-column joints is also costly and labor intensive. In this paper, the full-scale experimental behavior of two alternative frame-braced subassemblies incorporating ductile embeds is summarized. Observations from the experiments indicate that both specimens exhibit stable hysteretic behavior with no strength degradation in excess of 7% drift ratio. At design plastic rotations expected of conventional beam-column joints, no joint spalling and only minor damage along beam members was observed. A design-oriented lumped element model is described, which reasonably captures the local behavior of the specimens.

The distribution of bond on the lateral surface of steel reinforcement embedded in concrete is explored through systematic solution of the governing field equations of the associated mechanical problem. By separating the variables, the state of stress in the concrete surrounding the bar is represented by coupling two independent solutions, each describing attenuation of bond stresses either in the longitudinal direction or in the radial direction of the cover, respectively. Kinematic considerations are used to couple longitudinal slip with the radial translation of the cover, whereas the corresponding stress components developing along the lateral bar surface over the anchorage (radial pressure and bond stress) are related through a frictional relationship. Using the derived solution, various experimentally documented trends and phenomena pertaining to the study of bond, as represented by the international experimental databank, are reproduced analytically and interpreted. These include the processes of debonding and yield penetration and the sensitivity of the development capacity to important design variables such as the related rib area, cover to bar diameter ratio, concrete strength, bar yield strength, bar hardening characteristics, and confinement. A model representing the envelope curve of the local bond-slip relationship suitable for analysis and simulation is presented and calibrated in the paper.

Experimental research is presented in this paper to investigate the hydrostatic behavior of concrete under dynamic compression. By using the MTS servohydraulic testing system, strain rates from ${10}^{-5}$ to $3.5\times {10}^{-2}\hspace{0.17em}{\mathrm{s}}^{-1}$ were achieved. A hydrostatic pressure up to the uniaxial compressive strength of concrete was applied to the specimen with the help of the triaxial loading cell. A series of complete stress-strain curves was obtained for the specimens subjected to different combinations of strain rates and confining pressures, and significant enhancements of the material strength were observed. In particular, the experimental results suggest a clear coupling effect between the enhancements induced by the strain rate and the confining pressures. Finally, a set of empirical formulas is proposed to describe the enhancement of the compressive strength of concrete under different strain rates and relatively low confinement levels. The calculated results agree well with the data of the low confining pressure test and could meet the accuracy requirement in engineering design and applications.

Many models to determine the shear capacity of shear reinforced beams are based on the truss analogy. Various proposals have been formulated in recent years, all of which differ with regard to the limits of strut inclination. Remarkably, those limits do not depend on the type of concrete, which could be expected to be critical for the shear friction capacity of the cracks, which is supposed to be a major influencing factor with regard to the limit of strut rotation. Tests on beams with I-shaped cross sections have been carried out on beams made of normal, lightweight, and high-strength concrete. The experimental results are compared, in Part I of this paper, with those obtained by analytical models and, in Part II, with those obtained by nonlinear finite element programs tailored to this specific application. The result is that the type of concrete does not lead to significant changes in strut rotation capacity, so that the strut rotation limit values have general validity. The level of approximation approach, as presented in the Fédération Internationale du Béton/International Federation for Structural Concrete (fib) Model Code of Concrete Structures 2010 is justified: more sophisticated calculation models lead indeed to more accuracy in the determination of the shear capacity.

Nonlinear finite element analysis (NLFEA) is a useful tool for the assessment and design of structures if the reliability of the prediction is known in advance. Therefore, it makes sense to tailor finite element programs to specific types of structural members by calibrating them against test results obtained on similar types of specimens. By increasing the accuracy of prediction for such specified types of structural members, assessment and design are possible with higher reliability than can be obtained with analytical models. This strategy is illustrated for the case of shear-reinforced flanged beams using the constitutive model PARC_CL implemented into ABAQUS code. A comparison with analytical models shows the power of tailored FEM calculations in combination with reliability considerations.

This paper proposes a new practical and efficient procedure to investigate the seismic response of gravity dams that (1) avoids the finite or boundary element discretization of the impounded reservoir; (2) can be applied using standard finite-element software not necessarily including fluid-structure interaction capabilities; and (3) accounts for dam and foundation flexibility, water compressibility, and reservoir bottom wave absorption. The proposed technique consists of modifying the original input ground acceleration to obtain a new accelerogram that directly accounts for the complex effects of fluid-structure interaction. This new accelerogram can then be applied to a dam or dam-foundation system without the impounded reservoir. The exact and simplified formulations of the proposed method are developed, and its efficiency is validated through examples of dam-reservoir systems with different geometries. Very satisfactory agreement is obtained when comparing the results to more advanced finite-element solutions including fluid-structure interaction capabilities. The new procedure enhances the efficiency of seismic assessment of gravity dams by reducing the modeling and computational burden associated with reservoir discretization while keeping the main advantages of conventional solid finite-element software in conducting dynamic analyses and providing the resulting stresses, strains, force distributions, and other dam response indicators.

Double-layer tensegrity grids (DLTGs) are spatial reticulated systems based on tensegrity principles, which have been studied in detail over recent years. The most important investigations have been carried out focusing on a short list of tensegrity grids. This paper explains, using real examples, how to use Rot-Umbela manipulations, a unique technique developed for generating innovative typologies of tensegrity structures. It is applied to two already existing tensegrity grids to obtain two new DLTGs. Their analysis permits us to identify, inside these novel grids, the modules that compose them, which were unknown until now. A brief description of these components and some information about their static analysis, for example states of self-stress and internal mechanisms, is provided. These novel modules belong to a family, all of them with similar characteristics in terms of geometry and topology, and can be used to generate a wide catalog of DLTGs. Some examples of new grids are presented, describing the methodology on how to obtain more models for other designers interested in creating and studying innovative DLTGs.