Reinforced concrete (RC) flat-plate structures are vulnerable to punching shear failure at their slab-column connections, potentially leading to a catastrophic progressive collapse. In practice, the slab-column connection above an interior column, removed due to abnormal loads, may be subjected to a concentrated downward force because of the absence of the supporting column and further being pushed as a result of different live load intensities on individual stories. This force is different from the full design load that the column withstands in normal situations and, combined with the gravity load acting on the slab, may cause punching shear failure at the interior slab-column connection. This will further trigger failure propagation to the surrounding slab-column connections. This paper presents the experimental tests performed on two identical large-scale $2\times 2$ -bay RC flat-plate specimens under an interior column removal scenario. A 5-kPa uniformly distributed load was applied first to the slab followed by an incremental concentrated force imposed on the slab-column connection above the removed interior column. The complete collapse-resistant behavior and load redistribution pattern of the specimens were investigated and are reported herein. Results show that more than 90% of the applied concentrated force is solely distributed to the four nearest adjacent columns. Three load-carrying mechanism phases, in the form of flexural, tensile membrane, and a combination of one-way catenary and dowel actions can be distinguished in resisting the applied concentrated load.

This study introduces a multiple-input convolutional neural network (MI-CNN) model for the seismic damage assessment of regional buildings. First, ground motion sequences together with building attribute data are adopted as inputs of the proposed MI-CNN model. Second, the prediction accuracy of MI-CNN model is discussed comprehensively for different scenarios. The overall prediction accuracy is 79.7%, and the prediction accuracies for all scenarios are above 77%, indicating a good prediction performance of the proposed method. The computation efficiency of the proposed method is 340 times faster than that of the nonlinear multi-degree-of-freedom shear model using time history analysis. Third, a case study is conducted for reinforced concrete (RC) frame buildings in Shenzhen city, and two seismic scenarios (i.e., M6.5 and M7.5) are studied for the area. The simulation results of the area indicate a good agreement between the MI-CNN model and the benchmark model. The outcomes of this study are expected to provide a useful reference for timely emergency response and disaster relief after earthquakes.

A structural progressive collapse is usually a local failure, in which the damage is concentrated at beams that bridge the removal column and the column itself. In many cases, retrofitting the damaged structure is more economical and more sustainable than reconstructing the entire structure. A progressive collapse test of a 1/3 scale, four-bay by two-story reinforced concrete (RC) frame was conducted, after which the structure was retrofitted with carbon fiber reinforced polymer (CFRP) wraps and retested. The center column in the first story was removed and the frame was pushed down quasistatically under displacement control to investigate the progressive collapse performances of the retrofitted RC frame. The test results were represented systematically at different areas in terms of the resistance forces, crack developments, and local and global failure modes. Numerical models were built to verify the test frame before and after the retrofitting. A design method was proposed to retrofit an RC frame using CFRP wraps after a progressive collapse. The test frame was redesigned to improve the retrofitting and used as an example to demonstrate the rationality of the proposed retrofit design method. The results indicated that the proposed retrofitting technology rapidly restored the frame structure to its original capacity before the progressive collapse occurred, whilst consistently satisfying the priorities of being economical and sustainable.

This study presents a machine learning-based method for the destructive power assessment of earthquake to structures. First, the analysis procedure of the method is presented, and the backpropagation neural network (BPNN) and convolutional neural network (CNN) are used as the machine learning algorithms. Second, the optimized BPNN architecture is obtained by discussing the influence of a different number of hidden layers and nodes. Third, the CNN architecture is proposed based on several classical deep learning networks. To build the machine learning models, 50,570 time-history analysis results of a structural system subjected to different ground motions are used as training, validation, and test samples. The results of the BPNN indicate that the features extraction method based on the short-time Fourier transform (STFT) can well reflect the frequency-/time-domain characteristics of ground motions. The results of the CNN indicate that the CNN exhibits better accuracy (R² = 0.8737) compared with that of the BPNN (R² = 0.6784). Furthermore, the CNN model exhibits remarkable computational efficiency, the prediction of 1000 structures based on the CNN model takes 0.762 s, while 507.81 s are required for the conventional time-history analysis (THA)-based simulation. Feature visualization of different layers of the CNN reveals that the shallow to deep layers of the CNN can extract the high to low-frequency features of ground motions. The proposed method can assist in the fast prediction of engineering demand parameters of large-number structures, which facilitates the damage or loss assessments of regional structures for timely emergency response and disaster relief after earthquake.

In flat plate structures, the slab-column joints are prone to punching shear failure, which would trigger progressive collapse of the entire structural systems. The joints subjected upward punching shear (UPS) and downward punching shear (DPS) actions exhibit different post-punching behavior, which will affect the progressive collapse of flat plate structures. A static collapse test was conducted on two slab-column joint specimens (UPS-1 and DPS-1). The structural resistance and deformation were analyzed to explore the collapse-resistant mechanism of the joints after punching shear. The test results illustrate that the post-punching bearing and deformation capacities are governed by the through column longitudinal reinforcement (including both integrity reinforcement (IR) and flexural reinforcement (FR)) in the slab. The longitudinal bars ruptured in reverse orders under UPS and DPS failure scenarios. In DPS-1, the peak resistance under the suspension mechanism was reached when the first FR bar ruptured while the IR bars remained intact. The peak resistance of UPS-1, on the other hand, was achieved at the onset of the IR bars being ruptured and the FR bars being detached from the concrete.

In reinforced concrete (RC) frames, horizontal progressive collapse (H-PC) may be triggered due to insufficient lateral stiffness of the structural systems. Current research studies have been mainly focused on vertical progressive collapse (V-PC) rather than H-PC. To analyze the effect of lateral stiffness and capacity on the behavior of progressive collapse (PC) of RC frames, a static collapse test of two asymmetric one-story beam-column substructure specimens with flange slabs was conducted. The two specimens S1 and S4 were subjected to an interior column removal scenario, in which S1 has one edge column and S4 has an edge and a penultimate column. The test results show that under the compressive arch action (CAA), S1 developed a smaller peak resistance and a larger vertical displacement as compared to S4 due to insufficient horizontal restraint of the columns. In addition, the contribution of the CAA within the beams and slabs in S1 was significantly reduced. Under the catenary action (CA), the collapse resistance of S4 developed rapidly and eventually exceeded the peak resistance of CAA; while the resistance of S1 remained almost constant, caused by the continuously weaken horizontal constraint provided by the edge column. Additionally, the final failure mode for both specimens was found to be in form of the compression-bending failure of the edge and penultimate columns.