An algorithm for the anchor points of the PPS of the BCC model

Anchor points play an important role in data envelopment analysis theory and applications;they delineate the efficient part of the production possibility set(PPS)frontier.In this paper,we propose an approach for finding the anchor points of the PPS of the Banker,Charnes and Cooper(BCC)model.This approach is based on a variant of super-efficiency models and their duals.The necessary and sufficient conditions for the characterization of the anchor points are also provided.Finally,the applicability of the proposed model is illustrated with some numerical examples.

Performance Evaluation of Mobility Anchor Point with Guard Load Reservation in Hierarchical Mobile IPv6

Hierarchical mobile IPv6 （HMIPv6） introduces a mobility anchor point to reduce the signaling overhead and handoff latency. In this paper, we apply the matrix-analytical approach to explore the performance measures of the ongoing mobile nodes （MNs） drop and new MNs block probabilities of mobility anchor point with a guard bandwidth reservation scheme. We apply the Markovian arrival process （MAP） to model ongoing MNs and new MNs. Five related performance measures are derived, including the long-term new MN block and ongoing MN drop probabilities, and the three short-term measures of average length of a block period and a non-block period, as well as the conditional ongoing MN drop probability during a block period. These performance measures greatly assist the guard bandwidth reservation mechanism in determining a proper threshold guard bandwidth. The results presented in this paper can provide guidelines for designing adaptive algorithms to adjust the threshold in the guard bandwidth reservation scheme.

Motion Planning for a Cable-Driven Lower Limb Rehabilitation Robot with Movable Distal Anchor Points

This article introduces a cable-driven lower limb rehabilitation robot with movable distal anchor points(M-CDLR).The traditional cable-driven parallel robots(CDPRs)control the moving platform by changing the length of cables,M-CDLR can also adjust the position of the distal anchor point when the moving platform moves.The M-CDLR this article proposed has gait and single-leg training modes,which correspond to the plane and space motion of the moving platform,respectively.After introducing the system structure configuration,the generalized kinematics and dynamics of M-CDLR are established.The fully constrained CDPRs can provide more stable rehabilitation training than the under-constrained one but requires more cables.Therefore,a motion planning method for the movable distal anchor point of M-CDLR is proposed to realize the theoretically fully constrained with fewer cables.Then the expected trajectory of the moving platform is obtained from the motion capture experiment,and the motion planning of M-CDLR under two training modes is simulated.The simulation results verify the effectiveness of the proposed motion planning method.This study serves as a basic theoretical study of the structure optimization and control strategy of M-CDLR.

Modified laparoscopic Sugarbaker repair of parastomal hernia with a three-point anchoring technique

AIM To retrospectively evaluate the safety and feasibility of a new modified laparoscopic Sugarbaker repair in patients with parastomal hernias.METHODS A retrospective study was performed to analyze eight patients who underwent parastomal hernia repair between June 2016 and January 2018. All of these patients received modified laparoscopic Sugarbakerhernia repair treatment. This modified technique included an innovative three-point anchoring and complete suturing technique to fix the mesh. All procedures were performed by a skilled hernia surgeon. Demographic data and perioperative outcomes were collected to evaluate the safety and efficacy of this modified technique.RESULTS Of these eight patients, two had concomitant incisional hernias. All the hernias were repaired by the modified laparoscopic Sugarbaker technique with no conversion to laparotomy. Three patients had in-situ reconstruction of intestinal stoma. The median mesh size was300 cm2, and the mean operative time was 205.6 min. The mean postoperative hospitalization time was10.4 d, with a median pain score of 1(visual analog scale method) at postoperative day 1. Two patients developed postoperative complications. One patient had a pocket of effusion surrounding the biologic mesh,and one patient experienced an infection around the reconstructed stoma. Both patients recovered after conservative management. There was no recurrence during the follow-up period(6-22 mo, average 13 mo).CONCLUSION The modified laparoscopic Sugarbaker repair could fix the mesh reliably with mild postoperative pain and a low recurrence rate. The technique is safe and feasible for parastomal hernias.

基于锚点的快速三维手部关键点检测算法

在人机协作任务中,手部关键点检测为机械臂提供目标点坐标,A2J(Anchor-to-Joint)是具有代表性的一种利用锚点进行关键点检测的方法。A2J以深度图为输入,可实现较好的检测效果,但对全局特征获取能力不足。文中设计了全局-局部特征融合模块(Global-Local Feature Fusion,GLFF)对骨干网络浅层和深层的特征进行融合。为了提升检测速度,文中将A2J的骨干网络替换为ShuffleNetv2并对其进行改造,用5×5深度可分离卷积替换3×3深度可分离卷积,增大感受野,有效提升了骨干网络对全局特征的提取能力。文中在锚点权重估计分支引入高效通道注意力模块(Efficient Channel Attention,ECA),提升了网络对重要锚点的关注度。在主流数据集ICVL和NYU上进行的训练和测试结果表明,相比于A2J,文中所提方法的平均误差分别降低了0.09 mm和0.15 mm。在GTX1080Ti显卡上实现了151 frame·s^(-1)的检测速率,满足人机协作任务对于实时性的要求。

三亚旅游流空间结构及旅游者游览特征研究

基于去哪儿网在线游记所获取的数字足迹数据,在锚点理论框架下综合运用流向统计、关联规则挖掘、社会网络分析等方法,对三亚市旅游流的空间结构进行分析,并凝练不同类型旅游者在不同时间尺度下的游览特征。结果表明:1)三亚市旅游流呈现“一心两核”和若干核心路径的流动格局,三亚市旅游流的空间结构受景点受欢迎程度的影响,核心区处于沿海岸线区域,展现滨海旅游城市的特征;2)在三亚市旅游流的空间结构中核心节点分布不均匀且结构洞现象明显;3)三亚市旅游流结构网络的中尺度旅行的旅游者占多数,其次为小尺度旅游者,大尺度旅游者最少,各尺度各游览类型的旅游者都倾向于游览沙滩型旅游地和岛区,即地文景观类旅游地,而不同尺度下不同游览类型的旅游者游览倾向又有所不同;4)三亚市旅游流的旅游吸引系统呈分级分布且有多个锚点吸引物,不同的锚点吸引物除沙滩型旅游地和岛区外,对不同游览类型和游览时间的旅游者吸引力不同。

考虑系泊点深度的单点系泊系统锚桩水平承载性能理论分析

单点系泊系统因其克服恶劣海况的优越性成为我国海洋石油开发的重要系泊形式。然而,水下单点系泊系统锚桩-土体相互作用机制尚不清楚,尤其缺少考虑系泊点深度的水平承载性能计算方法。基于复合地基反力法,建立水平荷载下单点系泊系统锚桩的挠曲控制方程和边界条件,并编制相关计算程序;通过迭代法求解,获得不同系泊点位置下的锚桩荷载-位移曲线以及桩身位移、弯矩沿桩身的分布规律曲线;通过与数值模拟和既有离心模型试验结果的对比分析,验证了其准确性和可靠性;继而考虑系泊点深度对锚桩水平承载力的影响,进行参数分析,给出了最优系泊点深度取值范围。分析结果表明,增加系泊点深度可以显著减小锚桩的水平位移及弯矩最大值,从而提高锚桩的水平极限承载力。

软黏土中吸力锚承载特性离心试验研究

加载倾角和系泊点位置是吸力锚基础最重要的设计要素之一,能够改变锚体的破坏姿态以影响极限承载力。在饱和软黏土地基中选择不同的加载倾角及系泊点位置进行位移控制下的张紧式吸力锚离心模型试验,利用六自由度磁力计装置定量分析了加载倾角和系泊点位置对吸力锚破坏姿态与极限承载力的影响。研究发现,系泊点位置在锚体约2/3高度时,吸力锚发生平动破坏,当加载倾角由35°变化至20°时,在相同的系泊点位置,吸力锚发生后仰破坏,归一化极限承载力稍增大,并且达到极限承载力后仍保持一定的承载力余量;系泊点位置在锚体2/3高度以上时,吸力锚发生前倾破坏,归一化极限承载力降低了25%左右,并且破坏后的承载力余量大大降低。无论破坏模式如何,均未发现锚内土塞与锚有明显分离。

基于数据特征的接触网补偿装置卡滞识别方法研究

对于铁路接触网中心锚结及下锚补偿装置出现异常,根据6C检测系统中高速铁路弓网综合检测装置(1C)采集的波形数据总结出数据特征并提出相应算法。选取弓网接触力和接触线高度作为判断此类缺陷的2个特征变量,经过数据预处理后,计算弓网接触力一跨内标准差,接触线高度通过移动滑动窗,采取相应规则形成上界线,结合波形数据特征确定阈值并进行逻辑判断,得出异常锚段公里标。通过对典型线路进行计算分析,结果表明该方法可以识别出接触网中心锚结及下锚补偿装置缺陷,可为快速寻找此类缺陷及相关分析提供数据支撑。

基于深度模块训练的数据库查询效率预测研究

针对现有数据库查询算法效率低、预测精度不足等问题,提出一种基于深度模块的数据库查询效率预测算法。先基于图结构提取待查询数据集的相关特征,采用模块化思维构建深度模块网络模型,通过调整残差块数量和神经网络层数,应对不同规模的数据集并实现模型性能和代价的匹配,明确目标数据点和锚数据点的空间位置关系,通过距离判断提升对目标数据查询预测效率。实验结果表明,所提出算法在应对LUBM5大规模数据集时查询时间控制在50 s之内,同时数据查询的准确率在95%以上。

水库溢洪道预应力锚索施工要点探讨

为探索预应力锚索加固技术在水工结构物加固中应用的可行性和适用性,以某山坡坡体高差大、溢洪道建造于坡体下方的水库工程为例,针对边坡浆砌石挡墙体系稳定性不足的客观事实,提出预应力锚索加固方案;对方案设计、施工准备、施工工艺展开分析探讨。结果表明,与其余加固技术相比,预应力锚索的应用能显著提升水库溢洪道边坡浆砌石挡墙稳定性及承载力,施工过程简便、安全,加固效果可长期保持。

某抽水蓄能电站地下厂房洞室群三维有限元稳定性分析

为了了解抽水蓄能电站地下厂房洞室群开挖支护后围岩的稳定性,以某抽水蓄能电站地下厂房洞室群为例,采用隐式弹粘塑性本构关系,建立三维有限元模型,对比分析地下洞室群毛洞工况、锚杆+锚索+衬砌联合支护工况及支护参数优化后围岩的变形、应力场分布及D-P点安全系数。结果表明:采用锚杆+锚索+衬砌联合支护的方式可以有效减小地下洞室群围岩的位移量,提高D-P点安全系数,使应力分布更为均匀,围岩的稳定性更好。锚杆长度在6m~8m范围变化及锚索预应力变化对围岩的变形、应力以及D-P点安全系数的影响不显著,但是对围岩局部有一定的影响,施工过程中应动态优化调整。

基于关键点的Anchor Free目标检测模型综述

目标检测是计算机视觉应用的基础,基于锚框的一些目标检测算法已难以满足目标检测中对目标处理的效率、性能等诸多方面的要求,而anchor free方法逐渐广泛应用于目标检测.本文首先重点介绍了CornerNet、CenterNet、FCOS模型的一系列基于关键点的anchor free目标检测方法,综述了算法思路及其优缺点;然后分别对基于锚框和基于关键点的目标检测算法在同一个数据集上作了性能比较和分析;最后对基于关键点的目标检测进行了总结,并展望了目标检测的未来发展方向.

基于混合现实的自行火炮维修指导系统

针对自行火炮现场维修过程中存在的维修知识获取不直观、维修操作不当易造成二次损伤、专家亲临指导不及时等难题,设计了一种基于混合现实技术的自行火炮维修指导系统。分析建立了系统实现的整体框架、运行机制,研究了虚实坐标同一化表达、三维注册融合、环境感知运动追踪和虚实遮挡处理等关键技术,实现了虚实维修空间的几何一致性。最后通过设计搭建的自行火炮混合现实辅助维修指导系统,结合传动装置齿轮箱故障维修实例验证了系统空间锚定、辅助维修、远程指导等关键功能技术可行性,实现了现场维修的实时指导,提升了维修效率。

二分k-means锚点提取的快速谱聚类

光谱聚类(spectral clustering,SC)由于在无监督学习中的有效性而受到越来越多的关注。然而其计算复杂度高,不适用于处理大规模数据。近年来提出了许多基于锚点图方法来加速大规模光谱聚类,然而这些方法选取的锚点通常不能很好地体现原始数据的信息,从而导致聚类性能下降。为克服这些缺陷,提出了一种二分k-means锚点提取的快速谱聚类算法(fast spectral clustering algorithm based on anchor point extraction with bisecting kmeans,FCAPBK)。该方法利用二分k-means从原始数据中选取一些具有代表性的锚点,构建基于锚点的多层无核相似图;然后通过锚点与样本间的相似关系构造层次二部图。最后在5个基准数据集上分别进行实验验证,结果表明FCAPBK方法能够在较短的时间内获得良好的聚类性能。

大型起重船系泊系统时域分析

起重船因其结构的特殊性以及作业工况的复杂性,对其在施工作业和安全方面产生不利的影响。多点锚泊定位系统在控制船位保证起重船正常工作方面起到了非常关键的作用,故有必要对起重船的系泊展开专门研究。基于1 500 t回转式起重船采用Workbench/AQWA来计算不同工况、不同水深、不同浪向角下系泊缆的张力、卧链长度以及船体运动响应。结果表明,起重船运动响应满足设计要求,系泊系统具有良好定位能力;在尾吊、侧吊、暴风自存3种不同工况、不同水深条件下,优先考虑系泊缆3以减少设计计算中的工作量;起重船随着水深的增加,横荡、横摇变化最为明显,发生在90°浪向。该研究有助于设计者和船上工作人员预测船体运动,为系泊缆的进一步深入研究奠定基础。

基于无锚框的目标检测方法及其在复杂场景下的应用进展

基于深度学习的目标检测方法是目前计算机视觉领域的热点,在目标识别、跟踪等领域发挥了重要的作用.随着研究的深入开展,基于深度学习的目标检测方法主要分为有锚框的目标检测方法和无锚框的目标检测方法,其中无锚框的目标检测方法无需预定义大量锚框,具有更低的模型复杂度和更稳定的检测性能,是目前目标检测领域中较前沿的方法.在调研国内外相关文献的基础上,梳理基于无锚框的目标检测方法及各场景下的常用数据集,根据样本分配方式不同,分别从基于关键点组合、中心点回归、Transformer、锚框和无锚框融合等4个方面进行整体结构分析和总结,并结合COCO(Common objects in context)数据集上的性能指标进一步对比.在此基础上,介绍了无锚框目标检测方法在重叠目标、小目标和旋转目标等复杂场景情况下的应用,聚焦目标遮挡、尺寸过小和角度多等关键问题,综述现有方法的优缺点及难点.最后对无锚框目标检测方法中仍存在的问题进行总结并对未来发展的应用趋势进行展望.

基于改进区域生长法的洞室支护锚杆点云提取方法研究

针对施工期抽水蓄能电站地下厂房支护锚杆点云影响数据处理精度的问题,提出了一种基于反射强度改进区域生长算法提取锚杆点云的方法。通过分析锚杆表面曲率及反射强度特征,推导平滑及曲率阈值对提取结果的影响关系,确定基于反射强度的自适应阈值区域生长的改进方法,并通过采集洞室点云及提取支护锚杆点云分析该算法的提取结果。试验表明,该算法效率高,稳定性好,其提取点云簇准确率达98.09%,各簇锚杆的平均点云误提取率仅为2.61%。

Study on the Horizontal Movement State of Suction Anchor Piles for Offshore Wind Power during Horizontal Pulling

The subsea anchor piles of offshore wind power floating platform structures are mainly subjected to uplift and horizontal loads, and this paper focuses on the case of horizontal loads. A three-dimensional numerical simulation study of the horizontal pullout characteristics of wind power suction anchor piles in clay layers was carried out to reveal the horizontal movement state of the anchor piles during horizontal pile pullout, the range of pile depth at the pullout point where the horizontal movement is achieved (referred to as the horizontal movement range), the relationship between the pullout load and the ultimate load during the horizontal movement, and the optimal location of the pullout point for the horizontal movement. The results show that at certain pull-out points, the anchor pile produces an overall horizontal movement state under suitable horizontal pull-out loads. The depth of the pile pull-out point for horizontal movement is in the middle and lower part of the pile, i.e. 14.2 m to 14.5 m. The horizontal pull-out load of 24,000 kN at a depth of 14.5 m within the pile horizontal movement range of 14.2m to 14.5 m is the maximum ultimate horizontal pull-out load;the optimum pull-out point depth is 14.5 m at 0.275 L (L is the pile length). For each pull-out point of the anchor pile in horizontal movement, the horizontal pull-out load in horizontal movement and the horizontal ultimate pull-out load existed and it was found that the two values were not exactly the same, the values were compared and it was found that at the optimum pull-out point the value of the ultimate horizontal pull-out load/horizontal pull-out load in horizontal movement tended to 1.

基于锥形特征的二维横断面锚杆噪点剔除方法研究

本次研究针对地下洞室扫描采集的激光点云中存在支护锚杆噪点的问题,提出了一种基于锥形特征的二维横断面锚杆噪点剔除方法,设计了洞室二维横断面点云获取和锚杆点云精剔除的过程。首先提取洞室点云的二维中轴线,根据不同位置局部点云密度来自适应截取横断面点云切片,降低单次数据处理量;然后通过构建移动检测窗口,结合支护锚杆的结构参数,初步剔除锚杆点云;再采用基于锚杆锥形特征的噪点精剔除方法,对初步剔除的锚杆点云进行重采样。实验结果表明:构建的移动检测窗口初步识别锚杆噪点数量略多于目视判读的噪点数量,无锚杆漏识别现象;采用基于锚杆锥形特征的噪点精剔除方法,总体上,点云数据精确剔除噪点前后的平均误判值由37.32%降至18.19%,能有效地为地下洞室点云后续进行变形分析提供高质量的点云数据。