A machine learning-based fast seismic risk assessment framework is proposed to ease the computational burden in estimating the potential earthquake-induced loss of a building during its intended life. Both the hazard parameters of sites and structural parameters of buildings were incorporated as inputs. The continuous risk values and discrete risk levels were used as outputs for the regression and classification tasks of supervised learning, respectively. A proof-of-concept study of steel frames proved the feasibility of the proposed approach. Artificial neural networks achieved the lowest root mean square error of 0.0051 for regression and the highest accuracy of 96.8% for classification.

Coupling beams in shear wall and frame–shear wall systems are structural fuses that undergo significant inelastic deformation and absorb earthquake input energy. However, severe damage to coupling beams can disrupt building functions and have substantial repair costs. This study proposes frictional steel truss coupling beams (FTCBs) that aim to resolve these issues. The truss configuration is advantageous because it decouples shear and bending demands and facilitates pipeline layout. Shear-critical FTCBs (SFTCBs) adopt a smaller span-to-height ratio and place friction dampers in the diagonal webs, while bending-critical FTCBs (BFTCBs) adopt a larger span-to-height ratio and place friction dampers in the bottom chords. Quasi-static tests were conducted to validate the seismic performance of the FTCBs. A traditional reinforced-concrete coupling beam (RCCB) specimen was also tested for comparison. Results showed that FTCBs can realize damage control by concentrating inelastic deformation in friction dampers while keeping the main body of the steel truss and wall piers elastic. The FTCBs exhibited full and stable hysteretic behavior and enhanced energy dissipation capacity and replicability, which is difficult to achieve in RCCBs. Therefore, the FTCBs provide resilient alternatives to coupling beams over a range of span-to-height ratios.

Gusset plates connect the brace and frame in concentrically braced frames (CBFs). Failure at the interface of gusset plates with the beam or column will result in an inability to achieve energy dissipation through brace buckling and yielding of the brace in tension and failure of the connection before the brace. Traditional design methods for the gusset plate beam or column interfaces in CBFs often consider only forces from brace action. Although studies have shown that forces from frame action can be significant, the complex stress distribution in the gusset plate makes it difficult to effectively evaluate their effects. Neglecting or inaccurately evaluating the frame action may result in unreasonable estimation of the forces in gusset plates, leading to failure at the interface connections. In this paper, the brace action and the frame action forces on gusset plates are systematically explored through an experimental test of CBFs. The proposed design method combines two models for brace action and frame action. Finally, the proposed design method is validated against results from CBF experiments. The proposed design method estimates the magnitude and orientation of the gusset plate interface forces with different configurations of CBFs well.