碎软煤层条带瓦斯地面高效抽采技术研究

为了提高碎软煤层条带瓦斯抽采效率和效果,基于目前地面瓦斯抽采主要采用垂直井或从式井的方式抽采效果差、效率低的现状,通过理论和实验分析论证了穿岩层压裂改造煤储层的可行性,提出了在目标煤层顶板岩层中钻水平井,并通过垂直向下射孔以及采用泵送桥塞分段进行压裂的方式进行地面瓦斯抽采。试验结果表明:顶板分段压裂水平井单井产量高、高稳产期更长、产量衰减更慢;有效水平井段控制区域内瓦斯下降均匀,更有利于进行条带瓦斯抽采;相同投资条件下,采用水平井的方式瓦斯抽采效率和投入产出比更高。     

Design of energy‐efficient train speed profiles considering fixed‐block signaling system

In dense traffic railway networks, trains may often slow down or stop between stations owing to previous train delays. If preceding train trajectory can be predicted, energy‐efficient driving can be achieved by suppressing unnecessary speed changes. In this paper, we propose an algorithm to find energy‐efficient driving considering fixed‐block signaling (FBS) system by using dynamic programming (DP). DP is suitable for use because it can optimize the control inputs with discrete and state constraints. In this paper, we discuss energy‐efficient driving by considering a FBS system using some case studies of simulation. In the simulation, we examine a technique to drive an express train in an energy‐efficient way when the preceding local train is running toward the station with passing loops. The results show that the proposed method can derive complex speed profiles for energy‐efficient driving and the train can be operated with a maximum reduced energy consumption of 8.3%.     

Crystallization-Directed Anisotropic Electroactuation in Selectively Solvated Olefinic Thermoplastic Elastomers: A Thermal and (Electro)Mechanical Property Study

Dielectric elastomers (DEs), a class of soft electroactive polymers that change size upon exposure to an external electric field, constitute an increasingly important class of stimuli-responsive polymers due primarily to their large actuation strains, facile and low-cost fabrication, scalability, and mechanical robustness. Unless purposefully constrained, most DEs exhibit isotropic actuation wherein size changes are the same in all actuation directions. Previous studies of DEs containing oriented, stiff fibers have demonstrated, however, that anisotropic actuation along a designated direction is more electromechanically efficient since this design eliminates energy expended in nonessential directions. To identify an alternative, supramolecular-level route to anisotropic electroactuation, we investigate the thermal and mechanical properties of novel thermoplastic elastomer gels composed of a selectively solvated olefinic block copolymer that not only microphase-separates but also crystallizes upon cooling from the solution state. While these materials possess remarkable mechanical attributes (e.g., giant strains in excess of 4000%), their ability to be strain-conditioned enables huge anisotropic actuation levels, measured to be greater than 30 from the ratio of orthogonal actuation strains. This work establishes that crystallization-induced anisotropic actuation can be achieved with these DEs.     

基于GFU和分层LSTM的组群行为识别研究方法

提出一种以"关键人物"为核心，使用门控融合单元（GFU，Gated Fusion Unit）进行特征融合的组群行为识别框架，旨在解决两个问题：①组群行为信息冗余，重点关注关键人物行为特征，忽略无关人员对组群行为的影响；②组群内部交互行为复杂，使用GFU有效融合以关键人物为核心的交互特征，再通过LSTM时序建模成为表征能力更强的组群特征.最终，通过softmax分类器进行组群行为类别分类.该算法在排球数据集上取得了86.7%的平均识别率.     

改进ILoG算子的故障检测方法

针对强背景噪声干扰下轮对轴承故障特征微弱、难以准确检测的问题,提出了一种自适应改进高斯拉普拉斯(improved Laplacian of Gaussian,简称ILoG)算子的微弱故障检测方法。ILoG算子滤波器具有优良的信号突变特征检测能力,将其用于轮对轴承故障信号的冲击特征检测,同时利用水循环算法(water cycle algorithm,简称WCA)的寻优特性,并行搜寻筛选最佳的ILoG算子影响参数,通过对参数优化后ILoG算子滤波后信号做进一步包络解调分析,提取出轮对轴承微弱的故障特征信息。对实际轮对轴承外圈和内圈故障信号分析的结果表明,该方法可以有效检测出轴承微弱故障特征频率,故障检测效果优于小波阈值和多尺度形态学差值滤波方法。     

基于阶段注意力机制的电力线提取算法

电力系统维护是电力系统稳定运行的重要保障，应用智能算法的无人机电力巡检则为电力系统维护提供便捷。电力线提取是自主电力巡检以及保障飞行器低空飞行安全的关键技术，结合深度学习理论进行电力线提取是电力巡检的重要突破点。本文将深度学习方法用于电力线提取任务，结合电力线图像特点嵌入改进的图像输入策略和注意力模块，提出一种基于阶段注意力机制的电力线提取模型（SA-Unet）。本文提出的SA-Unet模型编码阶段采用阶段输入融合策略（Stage input fusion strategy， SIFS），充分利用图像的多尺度信息减少空间位置信息丢失。解码阶段通过嵌入阶段注意力模块（Stage attention module，SAM）聚焦电力线特征，从大量信息中快速筛选出高价值信息。实验结果表明，该方法在复杂背景的多场景中具有良好的性能。     

自适应融合层级特征的混合退化图像复原算法

多种退化类型混合的图像比单一类型的退化图像降质更严重,很难建立精确模型对其复原,研究端到端的神经网络算法是复原的关键.现有的基于操作选择注意力网络的算法(operation-wiseattentionnetwork,OWAN)虽然有一定的性能提升,但是其网络过于复杂,运行较慢,复原图像缺乏高频细节,整体效果也有提升的空间.针对这些问题,提出一种基于层级特征融合的自适应复原算法.该算法直接融合不同感受野分支的特征,增强复原图像的结构;用注意力机制对不同层级的特征进行动态融合,增加模型的自适应性,降低了模型冗余;另外,结合L1损失和感知损失,增强了复原图像的视觉感知效果.在DIV2K,BSD500等数据集上的实验结果表明,该算法无论是在峰值信噪比和结构相似性上的定量分析,还是在主观视觉质量方面,均优于OWAN算法,充分证明了该算法的有效性.     

Rational design of novel carboxylic acid functionalized phosphonium based ionic liquids as high-performance extractants for rare earths

Developing the novel ionic liquids as the potential substitutes for conventional organic solvents in extraction of the rare-earth metals is highly desirable but challenges still remain. In this study, the well-designed carboxylic acid functionalized phosphonium based ionic liquids, (4-carboxyl)butyl-trioctyl-phosphonium chloride/nitrate, were synthesized and characterized. The as-prepared samples were tested as the undiluted hydrophobic acidic extractant for rare-earth metal ions, affording the maximal loading of 3 mol/mol towards Nd(III) in aqueous solution and the remarkable stripping performance. The results also reveal their excellent extractability and selectivity for Sc(III) in the mixtures of six rare-earth ions, as well as the outstanding separation properties between rare-earth and first row transition-metal ions (i.e., La/Ni, Sm/Co). Moreover, the extraction mechanism indicates that the extracted rare-earth complex via a proton exchange in the ionic liquid phase is structurally similar to the complexes obtained with neutral extractants. This work presents a prototype for the fabrication of the hydrophobic cation-functionalized ionic liquids for highly efficient rare-earth extraction and provides the future application in recycling of rare-earth metals from the spent magnets.     

数据中心内服务器能耗数据采集及特征分析

数据中心的高能耗低能效问题正受到广泛关注与研究，但目前没有公开的数据中心内服务器能耗数据集供研究人员使用，且现有过滤式特征选择并不能满足运维人员的需求。为此，提出了一套模拟数据中心内服务器运行状态的仿真环境架构，基于该架构采集了服务器运行各类任务时的多项性能指标和能耗数据。然后将基于因果关系的特征选择应用于能耗数据集的特征分析中，构建出可解释的特征子集和能耗预测结果。实验结果表明，因果特征子集大小约为过滤式特征子集大小的1/3到1/6，并且使用因果特征子集训练的模型在75%的情况下都取得了最优预测精度。     

隧道洞口上方危岩崩塌块石运动规律及其对隧道的危害

为揭示危岩崩塌块石的运动规律及其对下方隧道洞口的危害,以都香高速昭通段乐红隧道洞口高陡边坡危岩区崩塌体为研究对象,运用理论计算与Rockfall数值计算2种形式对危岩区崩塌体失稳后块石的运动特征进行预测,运用Abaqus有限元软件模拟不同初速度、不同冲击落点下块石冲击隧道结构的动力响应,揭示不同因素对块石冲击隧道结构的影响程度.结果 显示:理论计算与Rockfall数值计算结果吻合,崩塌块石经坠落、碰撞、滚动最终会与隧道引洞段上表面发生碰撞;与冲击落点相比,块石初速度对隧道结构的动力响应特性影响更大;块石冲击落点位于隧道圆拱圆心正上方位置时对隧道结构的损伤最大;不同冲击落点对碰撞回弹后块石的速度影响不大.