Aiming at the limitations of the existing railway foreign object detection methods based on two-dimensional(2D)images,such as short detection distance,strong influence of environment and lack of distance information,w...Aiming at the limitations of the existing railway foreign object detection methods based on two-dimensional(2D)images,such as short detection distance,strong influence of environment and lack of distance information,we propose Rail-PillarNet,a three-dimensional(3D)LIDAR(Light Detection and Ranging)railway foreign object detection method based on the improvement of PointPillars.Firstly,the parallel attention pillar encoder(PAPE)is designed to fully extract the features of the pillars and alleviate the problem of local fine-grained information loss in PointPillars pillars encoder.Secondly,a fine backbone network is designed to improve the feature extraction capability of the network by combining the coding characteristics of LIDAR point cloud feature and residual structure.Finally,the initial weight parameters of the model were optimised by the transfer learning training method to further improve accuracy.The experimental results on the OSDaR23 dataset show that the average accuracy of Rail-PillarNet reaches 58.51%,which is higher than most mainstream models,and the number of parameters is 5.49 M.Compared with PointPillars,the accuracy of each target is improved by 10.94%,3.53%,16.96%and 19.90%,respectively,and the number of parameters only increases by 0.64M,which achieves a balance between the number of parameters and accuracy.展开更多
The seismic analysis of a rigid-framed prestressed concrete bridge in Tianjin Light Railway is performed. A 3-D dynamic finite element model of the bridge is established considering the weakening effect caused by the ...The seismic analysis of a rigid-framed prestressed concrete bridge in Tianjin Light Railway is performed. A 3-D dynamic finite element model of the bridge is established considering the weakening effect caused by the soft soil foundation. After the dynamic characteristics are calculated in terms of natural frequencies and modes, the seismic analysis is carried out using the modal response spectrum method and the time-history method, respectively. Based on the calculated results, the reasonable design values are finally suggested as the basis of the seismic design of the bridge, and meanwhile the problems encountered were also analyzed. Finally, some conclusions are drawn as: 1) Despite the superiority of rigid-framed prestressed concrete bridge, the upper and lower ends of the piers of the bridge are proved to be the crucial parts of the bridge, which are easily destroyed under designed earthquake excitations and should be carefully analyzed and designed; 2) The soft soil foundation can possibly result in rather weakening of the lateral rigidity of the rigid-framed bridge, and should be paid considerable attention; 3) The modal response spectrum method, combined with time-history method, is suggested for the seismic analysis in engineering design of the rigid-framed prestressed concrete bridge.展开更多
基金supported by a grant from the National Key Research and Development Project(2023YFB4302100)Key Research and Development Project of Jiangxi Province(No.20232ACE01011)Independent Deployment Project of Ganjiang Innovation Research Institute,Chinese Academy of Sciences(E255J001).
文摘Aiming at the limitations of the existing railway foreign object detection methods based on two-dimensional(2D)images,such as short detection distance,strong influence of environment and lack of distance information,we propose Rail-PillarNet,a three-dimensional(3D)LIDAR(Light Detection and Ranging)railway foreign object detection method based on the improvement of PointPillars.Firstly,the parallel attention pillar encoder(PAPE)is designed to fully extract the features of the pillars and alleviate the problem of local fine-grained information loss in PointPillars pillars encoder.Secondly,a fine backbone network is designed to improve the feature extraction capability of the network by combining the coding characteristics of LIDAR point cloud feature and residual structure.Finally,the initial weight parameters of the model were optimised by the transfer learning training method to further improve accuracy.The experimental results on the OSDaR23 dataset show that the average accuracy of Rail-PillarNet reaches 58.51%,which is higher than most mainstream models,and the number of parameters is 5.49 M.Compared with PointPillars,the accuracy of each target is improved by 10.94%,3.53%,16.96%and 19.90%,respectively,and the number of parameters only increases by 0.64M,which achieves a balance between the number of parameters and accuracy.
文摘The seismic analysis of a rigid-framed prestressed concrete bridge in Tianjin Light Railway is performed. A 3-D dynamic finite element model of the bridge is established considering the weakening effect caused by the soft soil foundation. After the dynamic characteristics are calculated in terms of natural frequencies and modes, the seismic analysis is carried out using the modal response spectrum method and the time-history method, respectively. Based on the calculated results, the reasonable design values are finally suggested as the basis of the seismic design of the bridge, and meanwhile the problems encountered were also analyzed. Finally, some conclusions are drawn as: 1) Despite the superiority of rigid-framed prestressed concrete bridge, the upper and lower ends of the piers of the bridge are proved to be the crucial parts of the bridge, which are easily destroyed under designed earthquake excitations and should be carefully analyzed and designed; 2) The soft soil foundation can possibly result in rather weakening of the lateral rigidity of the rigid-framed bridge, and should be paid considerable attention; 3) The modal response spectrum method, combined with time-history method, is suggested for the seismic analysis in engineering design of the rigid-framed prestressed concrete bridge.
