Aiming at the problem of dynamic instability of hard-brittle jointed rock surrounding in deep tunnel/roadway engineering,combining with the support concepts of"coupling rigidity with flexibility"and"ove...Aiming at the problem of dynamic instability of hard-brittle jointed rock surrounding in deep tunnel/roadway engineering,combining with the support concepts of"coupling rigidity with flexibility"and"overcoming rigidity by flexibility",the prevention and control method with"rigid-flexible coupling(R-F-C)"was put forward.Through numerical simulation calculation,the impact damage process,acoustic emission(AE)evolution characteristics,and element stress/displacement evolution characteristics of unsupported surrounding rock structure model,rigid supporting surrounding rock structure model,and"R-F-C"supporting surrounding rock structure model under horizontal bidirectional impact loading were compared and analyzed.Based on the theory of stress wave propagation,the dynamic instability catastrophe mechanism of three kinds of supporting structure models induced by horizontal bidirectional impact loading was revealed.Based on the Mohr-Coulomb strength theory,the stress discrimination methods of dynamic catastrophe of surrounding rock induced by horizontal bidirectional impact loading under three kinds of supporting structures were proposed.Combined with the above numerical simulation study,the explosion impact physical and mechanical test of"R-F-C"surrounding rock supporting plate structure was further designed and carried out.Finally,combined with the"conceptual model of ball-cliff potential energy instability",the energy driving theory and energy transformation mechanism of impact-induced rockburst under three kinds of supporting structures were discussed deeply.The research results provided a scientific basis for further promoting the effective application of"R-F-C"supporting structure in the prevention and control of dynamic instability of deep tunnel/roadway surrounding rock.展开更多
针对当前无人机(UAV)视角下小目标检测性能低以及漏检和误检的问题,提出基于YOLOv8改进的BDSYOLO(BiFPN-Dual-Small target detection-YOLO)模型。首先,使用RepViTBlock(Revisiting mobile CNN from ViT perspective Block)与EMA(Effici...针对当前无人机(UAV)视角下小目标检测性能低以及漏检和误检的问题,提出基于YOLOv8改进的BDSYOLO(BiFPN-Dual-Small target detection-YOLO)模型。首先,使用RepViTBlock(Revisiting mobile CNN from ViT perspective Block)与EMA(Efficient Multi-scale Attention)机制构造C2f-RE(C2f-RepViTBlock Efficient multi-scale attention)从而改进骨干网络中深层的C2f(faster implementation of CSP bottleneck with 2 Convolutions)模块,提升模型对小目标特征的提取能力并降低参数量;其次,使用双向特征金字塔网络(BiFPN)重构颈部网络,从而使不同层级的特征得以相互融合;然后,在改进颈部网络的基础上构造双重小目标检测层,并结合浅层和最浅层特征来提高模型对小目标的检测能力;最后,引入改进损失函数Inner-EIoU(Inner-Efficient-Intersection over Union),该函数使用更合理的宽高比衡量方式并解决交并比(IoU)自身的局限。实验结果表明,改进模型在VisDrone2019数据集上相对原始模型的精确率、召回率、mAP@50、mAP@50:95分别提升了8.5、7.7、9.2和6.3个百分点,而参数量仅为2.23×10~6,模型大小减小了19.1%。可见,所提模型在实现一定轻量化的同时显著提升了性能。展开更多
基金Project(2023AH051167)supported by the Natural Science Research Project of Anhui Educational Committee,ChinaProject(AHBP2024B-04)supported by the Foundation of Anhui Engineering Research Center of New Explosive Materials and Blasting Technology,China+1 种基金Project(GXZDSYS2023103)supported by the Open Fund for Anhui Key Laboratory of Mining Construction Engineering,ChinaProjects(52274071,52404155)supported by the National Natural Science Foundation of China。
文摘Aiming at the problem of dynamic instability of hard-brittle jointed rock surrounding in deep tunnel/roadway engineering,combining with the support concepts of"coupling rigidity with flexibility"and"overcoming rigidity by flexibility",the prevention and control method with"rigid-flexible coupling(R-F-C)"was put forward.Through numerical simulation calculation,the impact damage process,acoustic emission(AE)evolution characteristics,and element stress/displacement evolution characteristics of unsupported surrounding rock structure model,rigid supporting surrounding rock structure model,and"R-F-C"supporting surrounding rock structure model under horizontal bidirectional impact loading were compared and analyzed.Based on the theory of stress wave propagation,the dynamic instability catastrophe mechanism of three kinds of supporting structure models induced by horizontal bidirectional impact loading was revealed.Based on the Mohr-Coulomb strength theory,the stress discrimination methods of dynamic catastrophe of surrounding rock induced by horizontal bidirectional impact loading under three kinds of supporting structures were proposed.Combined with the above numerical simulation study,the explosion impact physical and mechanical test of"R-F-C"surrounding rock supporting plate structure was further designed and carried out.Finally,combined with the"conceptual model of ball-cliff potential energy instability",the energy driving theory and energy transformation mechanism of impact-induced rockburst under three kinds of supporting structures were discussed deeply.The research results provided a scientific basis for further promoting the effective application of"R-F-C"supporting structure in the prevention and control of dynamic instability of deep tunnel/roadway surrounding rock.
文摘针对当前无人机(UAV)视角下小目标检测性能低以及漏检和误检的问题,提出基于YOLOv8改进的BDSYOLO(BiFPN-Dual-Small target detection-YOLO)模型。首先,使用RepViTBlock(Revisiting mobile CNN from ViT perspective Block)与EMA(Efficient Multi-scale Attention)机制构造C2f-RE(C2f-RepViTBlock Efficient multi-scale attention)从而改进骨干网络中深层的C2f(faster implementation of CSP bottleneck with 2 Convolutions)模块,提升模型对小目标特征的提取能力并降低参数量;其次,使用双向特征金字塔网络(BiFPN)重构颈部网络,从而使不同层级的特征得以相互融合;然后,在改进颈部网络的基础上构造双重小目标检测层,并结合浅层和最浅层特征来提高模型对小目标的检测能力;最后,引入改进损失函数Inner-EIoU(Inner-Efficient-Intersection over Union),该函数使用更合理的宽高比衡量方式并解决交并比(IoU)自身的局限。实验结果表明,改进模型在VisDrone2019数据集上相对原始模型的精确率、召回率、mAP@50、mAP@50:95分别提升了8.5、7.7、9.2和6.3个百分点,而参数量仅为2.23×10~6,模型大小减小了19.1%。可见,所提模型在实现一定轻量化的同时显著提升了性能。
