基于光流法的固体内部声传播方向测量

目前采用光弹设备测量超声传播方向多为人眼主观观测,其误差较大,难以实现精确的定量测量。将Farneback光流法应用于光弹图像的处理,并根据帧与帧之间的光流图来计算光弹图像中的超声传播方向,可以捕捉图像中的动态声场。在实验中对不同传播方向的超声进行了测量,其绝对偏差度数最大值为2.85°,测量结果较为准确。因此,Farneback光流法可用于光弹图像中声波传播方向的判断,且具有快速、准确和直观等特点。     

基于混合蚁群算法的异质车队低碳VRP研究

针对货运车辆在配送调度过程中产生大量碳排放的问题，建立模型将多种影响碳排放量的因素协同优化。模型中考虑了不同载重量的异质车队，两个节点之间有多条道路的柔性路径，以及车辆重量随卸货而减少的动态负载等因素，以碳排放量、行驶时间和行驶路程为优化目标，并加入了节点需求时间窗、根据速度变化划分路段、交接和卸货时间的约束。提出了一种混合蚁群算法，利用蚁群算法信息素强度更新方式保持群体记忆性，利用粒子群算法的快速收敛特性增加计算效率。通过随机数值算例的仿真优化与对比分析，验证了算法和模型的有效性。     

基于“动态学习数据分析”的智慧课堂 评价考核体系构建

This paper firstly describes the current situation and limitations of learning evaluation and assessment in colleges and uni- versities; then, it analyzes the possibility of implementing the learning evaluation and assessment based on "dynamic learning data analysis"; then, it describes the design process of the learning evaluation and assessment system in detail; finally, it determines the various assessment items and carries out quantitative processing to better apply them to actual teaching.     

An efficient method of generating vortex electromagnetic wave based on double‐layer transmissive metasurface

A broadband double‐layer transmissive metasurface (TMS) for effective generate vortex electromagnetic wave is presented in this paper. The proposed TMS consists of two types of elements. The first element is composed of a multi‐resonant dipole and four metal vias. The metal vias increase the coupling strength between the upper and lower layers to improve transmission efficiency. On the basis of the first element, the second element adds stubs to ensure sufficient phase shift. The far‐field cross polarisation is eliminated by special element arrangement. Then, a centre‐fed linear polarisation TMS is designed to generate orbital angular momentum beams with mode l = ? 1. The proposed TMS is designed, manufactured and measured to verify the proposed design. The measured results indicate that a maximum gain of 20.8 dBi and narrow divergence angle of ±5° are achieved at 18 GHz. Furthermore, mode purity is higher than 86.1% within the 17 to 19 GHz band. The proposed double‐layer TMS saved costs, reduced weight and without assembly error is a good candidate for OAM generator.     

视频侦查中人像智能分析应用及算法优化

人像智能分析指的是对视频或录像中的人像进行结构化和可视化分析,对目标人物进行性别、年龄、发型等特征的智能识别,这项技术在视频侦查中有极高的应用价值。人像识别早期的算法是通过人工提取特征,通过学习低级视觉特征来针对不同属性进行分类学习,这种基于传统方法的模型表现常常不尽如人意。在计算机视觉领域,通过海量图像数据学习的神经网络比传统方法有更丰富的信息量和特征可以被提取。文章尝试通过深度学习技术训练神经网络模型对行人进行检测和识别,对于衣着不同的行人进行智能识别,具有更好的鲁棒性,提升了视频人像识别的准确率,拓展了人工智能技术在身份识别领域的应用。     

高泉背斜地层压力测井多参数综合解释与异常高压成因

目前国内外地层压力测井解释方法较多,但大多数方法基于欠压实沉积理论并局限于单一的地层岩性,尤其是针对盆地边缘山前深层构造,单一测井参数确定地层压力误差较大,且无法解释沉积压实机制以外的异常高压。为此,将多种异常高压机制相关的测井参数进行综合解释,建立适合山前复杂构造的地层压力测井多参数计算模型,评价准噶尔盆地南缘高泉背斜高探1井的地层压力剖面,其计算结果与钻井实测数据吻合,相对误差小于3.00%.同时,基于原始沉积加载—卸载过程力学原理,利用地层压力测井多参数分析判别出高泉背斜深层各层系异常高压的形成机制。     

广西伴生放射性矿开发利用中的放射性水平初步调查

本文围绕广西境内17类矿产资源的开发利用展开调查,并对280家详查企业开发利用中的放射性水平进行了监测分析。结果表明:在原料与产品中,锆石和氧化锆、铅/锌、锗/钛、钒等行业²³⁸U、²²⁶Ra、²³²Th任一核素比活度大于1 000 Bq/kg。在固体废物中,稀土、锆石和氧化锆、铅/锌、锗/钛、锡、钒等行业²³⁸U、²²⁶Ra、²³²Th任一核素比活度大于1 000 Bq/kg。花岗岩企业的废水中总α、总β放射性水平偏高。研究发现,在对矿产开发利用企业开展放射性水平调查时,将γ辐射剂量率现场监测筛选标准下调50 nGy/h,即设定为在固体表面1 m处的γ辐射剂量率超过“当地本底水平”＋100 nGy/h时,可更为保守地避免出现遗漏放射性水平较高企业的情况。     

基于广义深度学习的含DG配网故障诊断方法

针对传统故障定位方法难以满足含分布式电源配电网的问题，提出一种基于广义深度学习的故障定位方法。利用广义深度学习在逼近能力和容错性方面的优势，挖掘响应数据与故障位置之间的映射关系，建立含分布式电源配电网故障定位的模型。IEEE34节点仿真结果表明，该方法可有效实现含分布式电源配电网的故障定位，准确率高，速度快，且在信息畸变或缺失时容错性好。     

Hollow mesoporous silica nanoparticles as nanocarriers employed in cancer therapy: A review

Hollow mesoporous silica nanoparticles (HMSNs) have become an attractive drug carrier because of their unique characteristics including stable physicochemical properties, large specific surface area and facile functionalization, especially made into intelligent drug delivery systems (DDSs) for cancer therapy. HMSNs are employed to transport traditional anti-tumor drugs, which can solve the problems of drugs with instability, poor solubility and lack of recognition, etc., while significantly improving the anti-tumor effect. And an unexpected good result will be obtained by combining functional molecules and metal species with HMSNs for cancer diagnosis and treatment. Actually, HMSNs-based DDSs have developed relatively mature in recent years. This review briefly describes how to successfully prepare an ordinary HMSNs-based DDS, as well as its degradation, different stimuli-responses, targets and combination therapy. These versatile intelligent nanoparticles show great potential in clinical aspects.