With the goal of low-carbon environmental protection,the application of solid waste steel slag in ultra-high performance concrete(UHPC)was explored to develop green and environmentally friendly steel slag UHPC.The research examined the effect of varying contents of steel slag powder and steel slag aggregate on the tensile properties of steel slag UHPC through axial tensile tests.Specifically,it analyzes the effects of replacing cement with steel slag powder and quartz sand with steel slag aggregate at different contents on the axial tensile curve and mechanical indices of UHPC.The tensile failure mechanisms of steel slag UHPC were also explored.The results indicated that while the axial tensile strength of UHPC decreased with the inclusion of either steel slag powder or steel slag aggregate,it remained within acceptable limits.The material's ductility is impacted by the type and amount of steel slag used;specifically,adding steel slag powder alone enhanced ductility,whereas adding steel slag aggregate alone reduced it.When the combined content of steel slag powder and aggregate was relatively low,the material exhibited good ductility and a multi-crack failure mode.Based on the analysis of the tensile stress and strain characteristics,an optimal steel slag content for steel slag UHPC was recommended.
针对连续复合运动的关键帧对应的空间范围差异较大且存在重复,难以采用固定的空间特征标准提取的问题,提出一种基于多模态分段与聚类的层次化关键帧提取方法.在完整运动层面按照背景音乐节拍与时空信息等多模态信息将运动序列分割为多个片段;对各片段内部的帧进行空间特征聚类与时序分割,得到若干具有代表性的、姿势可能重复的候选关键帧;根据运动的时空特性消除冗余.以广播体操运动为例提取关键帧并与现有方法进行对比试验与分析,本研究方法能够更加准确、充分地提取运动的关键帧.
针对现有多目标跟踪方法易受到遮挡、运动模糊等问题干扰的情况,提出基于轨迹掩膜的在线多目标跟踪方法(online multi-object tracking method based on trajectory mask,OMTMTM).提出轨迹掩膜生成算法,利用前一帧跟踪轨迹结果生成轨迹掩膜,设计轨迹掩膜网络对轨迹掩膜提取多维度特征,包含目标可见区域的估计值、大致位置及形状等信息;将该特征与基础骨干网络提取的原始图像特征融合后进行多目标检测跟踪.OMTMTM的目标跟踪器具备先验判断能力,可实现遮挡情况下的准确跟踪;OMTMTM利用目标跟踪轨迹的时空信息,恢复出部分漏检或低置信待检目标,使轨迹掩膜更加合理,有利于后续跟踪.对OMTMTM的性能进行多维度评估,并结合基线模型进行对比分析.试验结果表明,OMTMTM具有先进的多目标跟踪性能.
为精准、高效改善城市老旧小区居住环境品质,提高其防灾、减灾能力,提出一种基于云模型的老旧小区韧性评价体系.基于韧性理论,对所选老旧小区按照不同年代进行分组,通过实地调研,获得陕西省西安市 30 组具有代表性的老旧小区特征数据,参考相关文献及相关专家建议,确定建筑韧性、设施韧性、环境韧性、人员韧性共计 4 个一级指标、9 个二级指标、30 个三级指标,构成老旧小区的评价指标体系.采用层次分析法(analytic hierarchy process,AHP)确定各指标的主观权重,采用熵权法(entropy weight method,EWN)和准则去除效果法(method based on the removal effects of criteria,MEREC)确定客观权重,采用组合赋权法确定最终权重,运用MATLAB构建了基于云模型的老旧小区韧性评价体系.基于此评价体系,选取西安市一老旧小区——万庆巷小区进行韧性评价,所得结果能够较准确地反映该小区韧性薄弱环节,表明所建评价体系具有一定的适用性与有效性.
基于济南某深基坑案例,建立数值模型,利用前期监测数据,进行位移反分析获得上部土的小应变硬化土(hardening soil-small,HSS)模型刚度参数,证明土、岩分别采用HSS模型和摩尔-库伦(Mohr-Coulomb,MC)模型的深基坑数值方法符合土岩双元基坑实际.对比土体与土岩基坑支护结构变形规律,明确基坑土体开挖深度 18m处存在"深度效应".发现土体开挖深度决定土岩基坑最大变形,岩体开挖的深度对基坑最终变形影响有限,土、岩界面处岩体对土体约束明显.利用土、岩深基坑数值模型,选择支护结构位移控制范围内地下室外墙变形最小情况的锚索回收顺序,可有效缩短工期,保证基坑与主体结构安全.基坑支护采用回收式锚索时,锚索回收存在最佳方案.
