This paper proposes a framework for critical element identification and demolition planning of frame structures.Innovative quantitative indices considering the severity of the ultimate collapse scenario are proposed u...This paper proposes a framework for critical element identification and demolition planning of frame structures.Innovative quantitative indices considering the severity of the ultimate collapse scenario are proposed using reinforcement learning and graph embedding.The action is defined as removing an element,and the state is described by integrating the joint and element features into a comprehensive feature vector for each element.By establishing the policy network,the agent outputs the Q value for each action after observing the state.Through numerical examples,it is confirmed that the trained agent can provide an accurate estimation of the Q values,and handle problems with different action spaces owing to utilization of graph embedding.Besides,different behaviors can be learned by varying hyperparameters in the reward function.By comparing the proposed method and the conventional sensitivity index-based methods,it is demonstrated that the computational cost is considerably reduced because the reinforcement learning model is trained offline.Besides,it is proved that the Q values produced by the reinforcement learning agent can make up for the deficiencies of existing indices,and can be directly used as the quantitative index for the decision-making for determining the most expected collapse scenario,i.e.,the sequence of element removals.展开更多
Progressive collapse is a relatively rare event which happens due to unusual loading on a structure that lacks adequate continuity,ductility and indeterminacy which causes local collapse in that structure and then ext...Progressive collapse is a relatively rare event which happens due to unusual loading on a structure that lacks adequate continuity,ductility and indeterminacy which causes local collapse in that structure and then extends it to other structural parts.The US department of defense published UFC4-023-03 regulation regarding the building design against progressive collapse.This regulation,based on the ASCE 7-05 standard,introduces two general approaches to building design against progressive collapse,including direct design and indirect design approaches.In this study,a variety of structural design methods for progressive collapse have been investigated.Moreover,their strengths and weaknesses have been mentioned.In general,the results of this study show that design based on alternative path(AP)method is more economical than other methods.Moreover,application of AP method is much more commonly accepted by researchers and designers.展开更多
This study presents the investigation of the approach which was presented by Thaer M.Saeed Alrudaini to provide the alternate load path to redistribute residual loads and preventing from the potential progressive coll...This study presents the investigation of the approach which was presented by Thaer M.Saeed Alrudaini to provide the alternate load path to redistribute residual loads and preventing from the potential progressive collapse of RC buildings.It was proposed to transfer the residual loads upwards above the failed column of RC buildings by vertical cables hanged at the top to a hat steel braced frame seated on top of the building which in turn redistributes the residual loads to the adjacent columns.In this study a ten-storey regular structural building has been considered to investigate progressive collapse potential.Structural design is based on ACI 318-08 concrete building code for special RC frames and the nonlinear dynamic analysis is carried out using SAP2000 software,following UFC4-023-03 document.Nine independent failure scenarios are adopted in the investigation,including six external removal cases in different floors and three removal cases in the first floor.A new detail is proposed by using barrel and wedge to improve residual forces transfer to the cables after removal of the columns.Simulation results show that progressive collapse of building that resulted from potential failure of columns located in floors can be efficiently resisted by using this method.展开更多
基金The authors gratefully acknowledge the financial support provided by the China Scholarship Council(CSC)during a visit of Shaojun Zhu to Kyoto University(No.201906260152)The second author acknowledges the support of JSPS KAKENHI(Grant No.JP20H04467)The third author acknowledges the support of Grant-in-Aid for Young Scientists(Start-up)(Grant No.JP21K20461).
文摘This paper proposes a framework for critical element identification and demolition planning of frame structures.Innovative quantitative indices considering the severity of the ultimate collapse scenario are proposed using reinforcement learning and graph embedding.The action is defined as removing an element,and the state is described by integrating the joint and element features into a comprehensive feature vector for each element.By establishing the policy network,the agent outputs the Q value for each action after observing the state.Through numerical examples,it is confirmed that the trained agent can provide an accurate estimation of the Q values,and handle problems with different action spaces owing to utilization of graph embedding.Besides,different behaviors can be learned by varying hyperparameters in the reward function.By comparing the proposed method and the conventional sensitivity index-based methods,it is demonstrated that the computational cost is considerably reduced because the reinforcement learning model is trained offline.Besides,it is proved that the Q values produced by the reinforcement learning agent can make up for the deficiencies of existing indices,and can be directly used as the quantitative index for the decision-making for determining the most expected collapse scenario,i.e.,the sequence of element removals.
文摘Progressive collapse is a relatively rare event which happens due to unusual loading on a structure that lacks adequate continuity,ductility and indeterminacy which causes local collapse in that structure and then extends it to other structural parts.The US department of defense published UFC4-023-03 regulation regarding the building design against progressive collapse.This regulation,based on the ASCE 7-05 standard,introduces two general approaches to building design against progressive collapse,including direct design and indirect design approaches.In this study,a variety of structural design methods for progressive collapse have been investigated.Moreover,their strengths and weaknesses have been mentioned.In general,the results of this study show that design based on alternative path(AP)method is more economical than other methods.Moreover,application of AP method is much more commonly accepted by researchers and designers.
文摘This study presents the investigation of the approach which was presented by Thaer M.Saeed Alrudaini to provide the alternate load path to redistribute residual loads and preventing from the potential progressive collapse of RC buildings.It was proposed to transfer the residual loads upwards above the failed column of RC buildings by vertical cables hanged at the top to a hat steel braced frame seated on top of the building which in turn redistributes the residual loads to the adjacent columns.In this study a ten-storey regular structural building has been considered to investigate progressive collapse potential.Structural design is based on ACI 318-08 concrete building code for special RC frames and the nonlinear dynamic analysis is carried out using SAP2000 software,following UFC4-023-03 document.Nine independent failure scenarios are adopted in the investigation,including six external removal cases in different floors and three removal cases in the first floor.A new detail is proposed by using barrel and wedge to improve residual forces transfer to the cables after removal of the columns.Simulation results show that progressive collapse of building that resulted from potential failure of columns located in floors can be efficiently resisted by using this method.
