Steel Frame Structure Research Articles

Vibration data‐driven machine learning architecture for structural health monitoring of steel frame structures

AbstractA vibration data‐based machine learning architecture is designed for structural health monitoring (SHM) of a steel plane frame structure. This architecture uses a Bag‐of‐Features algorithm that extracts the speeded‐up robust features (SURF) from the time‐frequency scalogram images of the registered vibration data. The discriminative image features are then quantised to a visual vocabulary using K‐means clustering. Finally, a support vector machine (SVM) is trained to distinguish the undamaged and multiple damage cases of the frame structure based on the discriminative features. The potential of the machine learning architecture is tested for an unseen dataset that was not used in training as well as with some datasets from entirely new damages close to existing (i.e., trained) damage classes. The results are then compared with those obtained using three other combinations of features and learning algorithms—(i) histogram of oriented gradients (HOG) feature with SVM, (ii) SURF feature with k‐nearest neighbours (KNN) and (iii) HOG feature with KNN. In order to examine the robustness of the approach, the study is further extended by considering environmental variabilities along with the localisation and quantification of damage. The experimental results show that the machine learning architecture can effectively classify the undamaged and different joint damage classes with high testing accuracy that indicates its SHM potential for such frame structures.

Life-cycle cost analysis of steel frames with shape-memory alloy based dampers

The shape-memory alloy (SMA) based dampers are passive dampers, with their most important feature being their memory and super elastic effect. They significantly reduce the residual response displacements of the structure subjected to earthquakes. Although the initial cost of these dampers is high, they can lower the structural expected life-cycle costs subjected to earthquakes. This study dealt with the life cycle cost analysis to calculate the expected costs of the structures equipped with shape-memory alloy-based dampers. Subsequently, two 4- and 12-story steel frame structures were investigated in controlled and uncontrolled cases in terms of life cycle costs. The results revealed that these dampers can reduce the inter-story drift ratios and the residual displacements of the 4-story structure subjected to the design level seismic acceleration by 31% and 63%, respectively. This reduction for the 12-story structure was 39% and 83%, respectively. Also, these dampers would decrease the total life-cycle cost of the 4-story and 12-story by 53%, and 90%, respectively. Finally, the result indicated that the total life-cycle cost was reduced upon increasing the capacity of the SMA-based dampers up to a specific capacity, beyond which the dampers would not play a role in reducing total life-cycle costs.

Optimum design of steel frames with sizing and orientation design variables using a reformulated evolution strategy

Sizing optimization of steel frame structures has been traditionally conducted based on fixed orientations of structural members. However, it seems more expedient to search for an optimum design where orientations of column members are considered as design variables together with sizing of all members for the minimum weight design of a steel frame. In order to solve the resulting optimization problem effectively, a reformulated evolution strategy (ES) algorithm is proposed and employed in the present study. The numerical investigations are carried out using three test problems regarding the optimum design of steel space frames. In each test problem, a frame is first designed for the minimum weight considering sizing design variables only, where initial orientations of the column members are kept unchanged. Next, sizing and orientation design variables are considered simultaneously for minimizing the weight of the same frame. The results obtained in the foregoing two cases are compared and the associated optimal layouts for the column orientations are depicted. The study reveals that the optimal selection of column orientations has a significant influence on the minimum weight design of steel frames.

A novel data-driven analysis for sequentially formulated plastic hinges of steel frames

Although many different nonlinear analysis techniques for steel frame structures, they causes the increase in total computational cost. A data-driven approach can then be used to substitute the classic finite element nonlinear analysis including second order methods and inelastic modelling. In this paper, a novel data-driven method using ensemble learning models for geometric and material nonlinear analyses of frame structures is proposed. For this approach, the data acquisition process and machine learning-based models are needed to train a large amount of dataset, which is collected from the nonlinear analysis and a yield surface equation. Ensemble learning algorithms are then used to build the data-driven models. After training, the sequence of plastic hinge formation in steel frame structures can be predicted without any nonlinear analysis process. In that situation, the data-driven model is more effective, especially in practical cases with the lack of experimental information due to limited sensors. The validity of the proposed method has been demonstrated by several numerical examples.

Strengthening of autoclaved aerated concrete (AAC) masonry wallettes with fabric reinforced cementitious matrix for in-plane shear and out-of-plane loads

Autoclaved aerated concrete (AAC) masonry is popularly used for the construction of infill and partition walls in the RC and steel-framed structures. However, past research highlighted that AAC masonry may exhibit poor performance during seismic events due to its low tensile capacity. In this study, the flexural and shear performance of AAC masonry was enhanced by using the fabric-reinforced cementitious matrix (FRCM). This strengthening scheme was applied in two modes: direct mode using anchor and sandwich method using adhesive mortar. These strengthening schemes were different from each other in the method of placing of the fabric. The experimental results showed that the proposed strengthening methods could improve the strength and ductility of the masonry under shear and flexural loads. The test results showed comparable strengths for both types of strengthening methodologies and either of two methods of fabric application may be chosen for the strengthening of AAC masonry, depending on the availability of materials and ease in construction. Moreover, the outcome of the conducted tests were compared with the respective theoretical predictions estimated using available analytical models and equations.

Quantification and parametric investigation for plan asymmetric low-rise steel frame structure considering soil-structure interaction under vertical ground motion

Quantification and parametric investigation for plan asymmetric low-rise steel frame structure considering soil-structure interaction under vertical ground motion

Numerical simulation of RC-masonry infill wall system strengthened with textile reinforced concrete

Masonry walls are generally considered non-load-bearing elements in most concrete or steel-framed structures. Being a heterogeneous structure, the masonry system effectively enhances the strength and lateral stiffness of the overall panel when subjected to horizontal forces. But the recent unsatisfactory performance of unreinforced Masonry Infill Walls (MIW) during in-plane seismic loading has caught the attention of many researchers for the last few years. To overcome these detrimental effects, strengthening the MIW frame is considered the best procedure to enhance the horizontal load-carrying capability of the system. Numerous studies evaluating various parameters affecting the behaviour of MIW were conducted using Fiber Reinforced Cementitious Material (FRCM) as a strengthening material. This paper conducted a validation study that includes numerical simulation of the experimental research carried out at Wellington Institute of Technology in which nine 2:3 scaled single bay and single story Reinforced Concrete Frame with Masonry Infills (RCFMI) specimens strengthened with different fibers (basalt, carbon, and glass) of FRCM was tested under in-plane seismic/cyclic loads. The study went into detail on creating a numerical model replicating the non-linear structural cyclic behaviour of an infill wall bound by a reinforced Concrete frame and subjected to displacement-controlled in-plane lateral stress. The numerical analysis was performed in the Finite Element Method (FEM) programme ABAQUS using a streamlined micro-model approach. To simulate the non-linear behaviour of the masonry blocks and concrete, the Concrete Damage Plasticity (CDP) model was employed. The tested specimen was retrofitted with glass fiber grid diagonal bands, each with a width equal to 1/6 of the infill diagonal length. This analysis was used to create the numerical model. The validation demonstrated that the numerical model could faithfully simulate the behaviour and forecast the strength of masonry infill walls with RC frames. The findings of the observations are explained in the form of load–displacement hysteresis loops and excursion curves. The numerical results demonstrate excellent agreement with the experimental data, a significant impact of FRCM on the system's dynamic behaviour, and an improvement in MIW effectiveness with FRCM under cyclic loading is observed.

Experimental study on the seismic performance of a cold-formed thin-walled steel–concrete composite column-H steel beam frame

For lightweight steel frame structures consisting of steel H-beams and cold-formed steel columns filled with concrete, seismic performance comparison tests and numerical simulation analyses were performed for bare and infilled frames. The effects of the lightweight wall panels, the axial compression ratio and the wall thickness of the steel sections of the columns on the seismic properties of the structure were investigated. The failure of the bare frame was concentrated in the weld fractures at the beam-column joints. When the wall panels were embedded in the frame, the damage was concentrated at the corners and edges of the wall panels and the connectors. The wall panels significantly improved the initial stiffness of the frame, early energy dissipation and resistance, and the wall panel energy dissipation rate was initially as great as 91%. As the axial compression ratio increased, the resistance of the structure significantly decreased. Under monotonic loading, the resistance on the structure with an axial compression ratio of 0.4 was reduced by nearly 44% compared to the structure without axial compression. Increasing the wall thickness of the steel sections of the columns increased the load-bearing capacity of the structure, but the increase diminished with increasing wall thickness.

Progressive collapse behavior of beam-to-column connections involving flange openings

Improving the ductility of the structure to withstand local damage and prevent progressive collapse is critical for steel-frame structures. In this study, a novel principle of beam-to-column connections was developed, which arranged openings at the beam flange to form a flange-opening section (FO). First, experimental investigations were conducted on two FO substructures, considering their deformation capacity, failure modes, internal force evolution, and resistance mechanism development. Second, numerical simulations verified by experiments were conducted to further investigate the number n, distance l, and diameter d of the flange openings, and the results were compared with those of the case that had no weakening of mechanical properties. Third, because some flange openings weakened the mechanical properties of the structure, specimens with V-shaped reinforcing plates (FOV) were studied. The results indicated that the mechanical properties of the FO specimens increased with n, decreased with an increase in d, and increased with l. The mechanical behavior of the new beam-to-column connections using flange openings and reinforcing plates was studied to relocate the plastic hinge, realize a second path, and successfully improve the progressive collapse resistance of the steel frame structure.

Modal identification of building structures under unknown input conditions using extended Kalman filter and long-short term memory

Various system identification (SI) techniques have been developed to ensure the sufficient structural performance of buildings. Recently, attempts have been made to solve the problem of the excessive computational time required for operational modal analysis (OMA), which is involved in SI, by using the deep learning (DL) algorithm and to overcome the limited applicability to structural problems of extended Kalman filter (EKF)-based SI technology through the development of a method enabling SI under unknown input conditions by adding a term for the input load to the algorithm. Although DL-based OMA methods and EKF-based SI techniques under unknown input conditions are being developed in various forms, they still produce incomplete identification processes when extracting the identification parameters. The neural network of the developed DL-based OMA method fails to extract all modal parameters perfectly, and EKF-based SI techniques has the limitations of a heavy algorithm and an increased computational burden with an input load term added to the algorithm. Therefore, this study proposes an EKF-based long short-term memory (EKF-LSTM) method that can identify modal parameters. The proposed EKF-LSTM method applies modal-expanded dynamic governing equations to the EKF to identify the modal parameters, where the input load used in the EKF algorithm is estimated using the LSTM method. The EKF-LSTM method can identify all modal parameters using the EKF, which is highly applicable to structural problems. Because the proposed method estimates the input load through an already trained LSTM network, there is no problem with computational burden when estimating the input load. The proposed EKF-LSTM method was verified using a numerical model with three degrees of freedom, and its effectiveness was confirmed by utilizing a steel frame structure model with three floors.

An efficient differential evolution-based method for optimization of steel frame structures using direct analysis

Sizing optimization of steel frames using direct analysis is a highly complex optimization problem with discrete design variables and highly nonlinear constraints of structural safety and serviceability conditions. To tackle this problem, this study develops an improved differential evolution (DE), namely 2EpDE, with a new mutation technique using two p-best individuals. The results of twenty benchmark functions and four real-world constrained optimization problems prove that 2EpDE effectively saves computational efforts and searches for better global optimal solutions. A framework for the optimization of steel frames is then proposed using a combination of 2EpDE, Multi-Comparison Technique (MCT), and Promising Individual Method (PIM). The strength and serviceability constraints of the optimization are evaluated by direct analysis that can accurately perform the nonlinear inelastic behaviors of the structure. Two examples including two-story and six-story space frames are optimized to evaluate the efficiency of the proposed method. Compared to Jaya and Particle Swarm Optimization (PSO), the proposed method can save more than 35% of computational efforts and yields better optimum results.

Comparative analysis of the whole life carbon of three construction methods of a UK-based supermarket

The built environment has been a significant contributor to global carbon emissions. It, therefore, has a vital role to play in the reduction efforts of future climate change. While the design of buildings may determine future energy use for cooling, heating, and lighting during the operational stage of the building, this study aims to observe the effect of the building design on the operational as well as the whole-life carbon emissions. Past studies have focused on either the operational carbon or the embodied carbon of a building. Using a cradle-to-grave assessment of a typical UK supermarket, this study explores the relationship between embodied carbon and operational carbon. Additionally, it examines the effects of the variables between three approved construction methods of the same design on the whole life of carbon. These methods are a steel structural frame and cladding panel external wall, steel frame and poroton walls, precast concrete and glulam frame and precast concrete walls. The findings of this research will contribute to mitigation strategies for the environmental impacts of supermarket building construction whilst providing a framework for future assessment of the whole-life carbon of supermarket buildings. Employing the life cycle assessment methodology, this paper examines the potential of minimising both embodied and operational carbon by observing the whole life carbon. Highlighting the influence of the GHG emission contributing factors in each stage on each other. Additionally, the recommended methodology for the supermarket building types of this case study, could be adapted for other types of buildings. The findings could also augment carbon emission research and guide the development of supermarket buildings to low carbon intensive. Furthermore, collaboration with the industry in carrying out this research aids in adopting the findings as practical and theoretical guides for engineers and designers in reducing the building sector’s harmful environmental impact.

Seismic behaviour of Concrete-filled steel tubular column to H-Shape steel beam connection with side plate

To ensure the quality of filled concrete and improve the indoor utilization rate of steel frame structure residential buildings, a novel concrete-filled steel tubular (CFST) column to H-shape steel beam connection with side plate was proposed in this study. Six specimens of 1/3 scale were tested under reversed cyclic loading. To evaluate the hysteretic performance, failure mode, ductility, stiffness degradation, strength degradation and energy dissipation capacity. The evaluation parameters include the axial load ratio, and the side plate height and thickness. The results indicated that the failure mode mainly exhibited the steel beam buckling failure, making all hysteretic curves exhibit typically spindle-shaped. The equivalent viscous damping coefficient is 0.305–0.394. The average story drift angle at the ultimate stage is 1/31 ∼ 1/37. It shows that all connections possess sufficient energy dissipation capacity and excellent ductility. The peak and ultimate load are significantly reduced with the increase of the axial load ratio. And the secant stiffness increases in the elastic stage with the rise of the thickness and height of the side plate. Finally, the failure mode, hysteresis curve, and envelope curve achieved from the finite element method(FEM) established in this paper are consistent with the test results. And the FEM disclosed that the beam flange stress is transferred to the side plate through the fillet weld, and the side plate stress is transferred to the column panel through the fillet weld.

Seismic performance analysis of steel frame structures with separated three‐dimensional isolation

AbstractThree‐dimensional (3D) base isolation can significantly improve the seismic performance of structures. However, significant 3D movement coupling and rocking behavior of the superstructure due to the high gravity center of the structure can be observed, which will lead to a complex design of the isolation system and the isolators. In addition, as the 3D isolators are installed beneath the structures, they need to support the heavy weight of the superstructure with low vertical stiffness. It is difficult to meet both the vertical load and stiffness design demands of the isolators using traditional isolation materials and mechanisms, especially for large‐size structures. A separated 3D isolation scheme is proposed by installing the horizontal isolation at the base of the superstructure and the vertical isolation under the floor slabs. Simulations are carried out to examine the dynamic responses including the acceleration, inter‐storey drift, and stress of structural members of a typical steel frame structure with the separated 3D isolation scheme. The isolation effect was also studied by shaking table tests of a two‐storey steel frame structure. The simulation and test results show that the proposed separated 3D isolation scheme has high performance in mitigating the structural accelerations in both horizontal and vertical directions, and the vertical load on isolators. The rocking behavior of the structure is effectively restrained for the separated 3D isolation scheme due to a low mass center configuration of the vertically isolated slabs. Although the isolation of the floor slabs from the beams in the structure with separated 3D isolation weakens the constraint of the floor slabs to the beam and column members as in a traditional frame structure, the horizontal isolation at the base of the superstructure reduces the adverse effect and the stress of structural members is similar to that in a horizontal base‐isolated structure and significantly lower than that in a base‐fixed structure. In addition, as the vertical isolation is installed under the slabs, the vertical load on the vertical isolators is small and it is easy to design the isolators.

Seismic Performance Evaluation of Carbon Steel Frame Structure Using Rubber Friction Bearing and HSS Bracing

Seismic Performance Evaluation of Carbon Steel Frame Structure Using Rubber Friction Bearing and HSS Bracing

Stochastic analysis of steel frames considering the material, geometrical and loading uncertainties

This paper develops a Stochastic Practical Advanced Analysis Program for stochastic analysis of structural steel frames. The second-order refined plastic-hinge analysis method combined with the technical simulation of Latin Hypercube Sampling is developed to predict the actual ultimate load-carrying capacity of steel frames and investigate the sensitivity of the uncertain input parameters. The input parameters of material properties, geometrical characteristics, and load combinations are considered as independent random variables that may occur in simultaneous randomness. A proposed parallel analytical technique integrates the modified Newton-Raphson and Generalized Displacement Control algorithms to solve the nonlinear inelastic problems to estimate the critical displacement-based system reliability index. The results of the statistical analysis in terms of coefficients of variation and Pearson correlation index show that the yield strength of material is the most sensitive with respect to the behavior of steel frames. The Bayesian Model Averaging is employed to find the most influential structural components on the ultimate structural resistance. The useful results of this research may be used in steel structure design and maintenance in practice.

Dynamic Characteristics of a Novel Right-Angle Viscoelastic Damper (RVD) Using Polyurethane Damping Materials

Excessive plastic deformation may occur at the beam-column joints under seismic action and lead to connection failure, increasing the possibility of collapse of the entire frame structure. In order to improve the seismic performance of assembled steel structure joints, a right-angle viscoelastic joint damper with polyurethane as the core energy dissipation material is proposed in this paper. First, the temperature scanning tests of polyurethane materials were carried out based on the dynamic mechanical analysis method. Second, the dynamic mechanical test and numerical simulation analysis of the designed right-angle viscoelastic damper were performed to reveal the dynamic energy dissipation characteristics of the right-angle damper. Finally, the dynamic time-history damping analysis was performed on the steel frame structure equipped with RVDs. The results show that the TA value of polyurethane material reached its peak at 2.0 Hz, which is the ideal frequency for the material’s damping ability. The right-angle damper made of polyurethane material has a softening nonlinear characteristic, and the peak value of the loss factor was obtained at 2.0 Hz, which is consistent with the results of the dynamic performance of the polyurethane material. The numerical simulation results demonstrate that stress on the steel plates and viscoelastic layers is reasonable. When the excitation level did not exceed 9 mm displacement amplitude, the energy input to the damper was dissipated by polyurethane, and the steel plates never showed plastic deformation. The time-history analysis of the steel structure shows that the dampers designed in this paper have a good control effect on the interstory displacement, acceleration, and interstory shear force of the structure. The research results lay the necessary foundation for the engineering application of polyurethane materials in the field of beam-column joints vibration damping.

Numerical prediction method for vibration characteristics of steel-framed ALC floor structure

The prediction of floor vibration is of great importance from the viewpoint of accurate prediction of sound environment. In this paper, applicability of the finite-element analysis to the vibration simulation of autoclaved lightweight aerated concrete (ALC) floor structure on a steel-framed structure was verified. The results of a numerical case study has confirmed that the numerical coupling scheme of the ALC floor panels and the steel-framed structure had a significant influence on the simulated vibration characteristics of the ALC floor panels. Then, the validity of the proposed method was finally confirmed by comparison with the measurement results.

Damping evaluation of an eight-story steel building with nonlinear oil damper under strong earthquakes

Fluid viscous damper has been applied widely for energy dissipation and shock absorption in dynamic structural systems, specifically the oil damper. Nevertheless, there's an issue of precisely incorporating shaft bearings and pressure sealings for oil dampers, which restricts its wide application in new or existing structures. A type of nonlinear oil damper with viscoelastic polymer packing soft links is posed and investigated to allow for easy production. To explore the effectivity and actual performance of the nonlinear oil dampers, vibration monitoring system has been instrumented both on the dampers and the building since it was built in 2003, which accumulated huge amount of precious data. A Bayesian based order selection of auto-regressive exogenous model is exploited to the recorded earthquake responses to extract dynamic properties of the passively-controlled 8-story steel building. Thus, damping characteristics of the whole structure can be identified and analyzed. Besides, an analytical model for the nonlinear oil damper is investigated and estimated by particle filter improved with Metropolis-Hastings algorithm, which can be incorporated into the whole structure model for model updating, reliability as well as response prediction. Furthermore, the calculation formulas of the additional effective damping ratio provided by oil damper are deduced based on above analysis using modal strain energy, and damping ratio of the main steel frame structure are also estimated. Taking a set of actual ground motion monitoring data as an example, the proposed method is then exploited to estimate the effective damping ratio of oil damper for each floor.

Rapid evaluation of structural soundness of steel frames using a coupling coefficient (CC)‐based method

AbstractA method is proposed for evaluating the post‐earthquake condition of steel frame structures based on the coupling coefficient (CC). The CC is an index that reflects the internal force distribution of each storey and damage mechanism in frames. It is computed using the drift distribution of the maximum structural response estimated from limited monitoring data and the design drawings of the target structure. Preliminary finite element analyses showed that the CC contains some important information that is not obvious in displacement responses, such as the global and local working modes of the structure. Useful information of this type includes whether the damage to each storey is mainly distributed on the beams or columns. An incremental dynamic analysis with an 18‐storey steel frame tested at the E‐Defense shaking table facility confirmed that the structural condition identified by the CC‐based method corresponds to the damage information reported by visual inspection. The analysis results verify the theory of the CC‐based method and provide limit values for the damage grades of steel frames.