Scale Invariant Theory of Gravitation in Einstein-Rosen Space-Time

In this paper, we have studied the perfect fluid distribution in the scale invariant theory of gravitation, when the space-time described by Einstein-Rosen metric with a time dependent gauge function. The cosmological equations for this space-time with gauge function are solved and some physical properties of the model are studied.

Convective Scale Structure and Evolution of a Squall Line Observed by C-Band Dual Doppler Radar in an Arid Region of Northwestern China

A long-lived and loosely organized squall line moved rapidly across U¨ru¨mqi, the capital city of Xinjiang Uygur Autonomous Region of China on 26 June 2005, generating hail and strong winds. The squall line was observed by a dual Doppler radar system in a field experiment conducted in 2004 and 2005 by the Chinese Academy of Meteorological Sciences and the local meteorological bureau in northwestern China. The 3D wind fields within the squall line were retrieved through dual Doppler analyses and a variational Doppler radar analysis system （VDRAS）. The formation and structure of the squall line as well as the genesis and evolution of embedded convective cells were investigated. During its life period, the squall line consisted of six storm cells extending about 100 km in length, and produced hail of about 25 mm in diameter and strong surface winds up to 11 m s^-1. Radar observations revealed a broad region of stratiform rain in a meso-β cyclone, with the squall line located to the west of this. Two meso-γ scale vortices were found within the squall line. Compared to typical squall lines in moist regions, such as Guangdong Province and Shanghai, which tend to be around 300–400 km in length, have echo tops of 17–19 km, and produce maximum surface winds of about 25 m s^-1 and temperature variations of about 8-C this squall line system had weaker maximum reflectivity （55 dBZ）, a lower echo top （13 km） and smaller extension （about 100 km）, relatively little stratiform rainfall preceding the convective line, and a similar moving speed and temperature variation at the surface.

LES prediction of space-time correlations in turbulent shear flows

We compare the space-time correlations calculated from direct numerical simulation （DNS） and large-eddy simulation （LES） of turbulent channel flows. It is found from the comparisons that the LES with an eddy-viscosity subgrid scale （SGS） model over-predicts the space-time corre- lations than the DNS. The overpredictions are further quantified by the integral scales of directional correlations and convection velocities. A physical argument for the overpre- diction is provided that the eddy-viscosity SGS model alone does not includes the backscatter effects although it correctly represents the energy dissipations of SGS motions. This argument is confirmed by the recently developed elliptic model for space-time correlations in turbulent shear flows. It suggests that enstrophy is crucial to the LES prediction of spacetime correlations. The random forcing models and stochastic SGS models are proposed to overcome the overpredictions on space-time correlations.

Dual metal atom catalysts:Advantages in electrocatalytic reactions

The dual-metal-atom catalysts(DACs)have aroused much attention as they possess the advantages of single-atom and metal alloy catalysts.And the DACs have exhibited enhanced performance in various electrocatalytic reactions,such as hydrogen/oxygen evolution and oxygen/carbon dioxide/nitrogen reduction.In this review,we mainly overview the latest understanding of the advantages of DACs for these reactions.This review will start with the familiar characterization methods for DACs,then the primary synthesis strategies for DACs will be discussed.Emphasis is given to the advantages of DACs in catalytic reactions,including the adsorption and activation,electronic structure regulation,breaking scaling relations,reducing energy barriers,cascading and coupling,synergy effect,and providing mechanism research platforms.Finally,personal perspectives and challenges for the further development of DACs are briefly discussed.

Three Dimensional Space-Time Gravitational Metric, 3 Space + 3 Time Dimensions

We have recently suggested a new quantum gravity theory that can be unified with quantum mechanics. We have coined this theory collision space-time. This new theory seems to be fully consistent with a 3-dimensional space-time, that is, three space dimensions and three time-dimensions, so some would call it six-dimensional. However, we have shown that collision-time and collision-length (space) are just two different sides of the same “coin” (space-time), so it is more intuitive to think of them as 3-dimensional space-time. In previous papers, we have not laid out a geometric coordinate system for our theory that also considers gravity, but we will do that here. We are pointing out that Einstein’s negative attitude towards relativistic mass can perhaps cause a weakness in the foundation of general relativity theory. When a relativistic mass is incorporated in the theory, this mass also seems to indicate one needs to move to three-dimensional space-time. Then, for example, our new theory matches fully up with all the properties of the Planck scale in relation to the mathematical properties of micro black holes, not only mathematically but also logically, something we demonstrate clearly that it is not the case of general relativity theory. Our new metric has many benefits as an alternative to the Schwarzschild metric and general relativity theory. It seems to be more consistent with the Planck units than the Schwarzschild metric. Most importantly, it seems to be fully consistent with a new quantum gravity theory that seems to unify gravity with quantum mechanics.

Gauge Gravitational Field in a Fractal Space-Time

Considering the fractal structure of space-time, the scale relativity theory in the topological dimension DT = 2 is built. In such a conjecture, the geodesics of this space-time imply the hydrodynamic model of the quantum mechanics. Subsequently, the gauge gravitational field on a fractal space-time is given. Then, the gauge group, the gauge-covariant derivative, the strength tensor of the gauge field, the gauge-invariant Lagrangean, the field equations of the gauge potentials and the gauge energy-momentum tensor are determined. Finally, using this model, a Reissner- Nordstrom type metric is obtained.

METHOD BASED ON DUAL-QUADRATIC PROGRAMMING FOR FRAME STRUCTURAL OPTIMIZATION WITH LARGE SCALE

The optimality criteria （OC） method and mathematical programming （MP） were combined to found the sectional optimization model of frame structures. Different methods were adopted to deal with the different constraints. The stress constraints as local constraints were approached by zero-order approximation and transformed into movable sectional lower limits with the full stress criterion. The displacement constraints as global constraints were transformed into explicit expressions with the unit virtual load method. Thus an approximate explicit model for the sectional optimization of frame structures was built with stress and displacement constraints. To improve the resolution efficiency, the dual-quadratic programming was adopted to transform the original optimization model into a dual problem according to the dual theory and solved iteratively in its dual space. A method called approximate scaling step was adopted to reduce computations and smooth the iterative process. Negative constraints were deleted to reduce the size of the optimization model. With MSC/Nastran software as structural solver and MSC/Patran software as developing platform, the sectional optimization software of frame structures was accomplished, considering stress and displacement constraints. The examples show that the efficiency and accuracy are improved.

Free-Interface Dual-Compatibility Modal Synthesis Substructure Method in Large-Scale Structures

Free-interface dual-compatibility modal synthesis method(compatibility of both force and displacement on interfaces)is introduced to large-scale civil engineering structure to enhance computation efficiency. The basic equations of the method are first set up, and then the mode cut-off principle and the dividing principle are proposed. MATLAB is used for simulation in different frame structures. The simulation results demonstrate the applicability of this substructure method to civil engineering structures and the correctness of the proposed mode cut-off principle. Studies are also conducted on how to divide the whole structure for better computation efficiency while maintaining better precision. It is observed that the geometry and material properties should be considered, and the synthesis results would be more precise when the inflection points of the mode shapes are taken into consideration. Furthermore, the simulation performed on a large-scale high-rise connected structure further proves the feasibility and efficiency of this modal synthesis method compared with the traditional global method. It is also concluded from the simulation results that the fewer number of DOFs in each substructure will result in better computation efficiency, but too many substructures will be time-consuming due to the tedious synthesis procedures. Moreover, the substructures with free interface will introduce errors and reduce the precision dramatically, which should be avoided.

中国水层CO_(2)地质封存技术攻关方向

在国家“碳达峰碳中和”战略目标下,为了延长化石能源行业的生命周期,亟需进行大规模碳减排的重大革命,其中CO_(2)地质封存(CCS)这一大规模碳减排的托底技术,已在全球达成共识。为此,系统梳理了全球CCS发展趋势,并针对中国CCS地质特点和工程技术现状,分析了中国发展水层CCS面临的挑战和难点,提出了中国水层CO_(2)地质封存技术的攻关方向。研究结果表明:①中国水层CCS起步早,当前正在开展世界级规模水层CCS先导试验,技术水平与国外并行,CCS有望成为中国未来大规模碳减排的主要手段;②鉴于中国水层多无法回避陆相沉积、断裂发育、非均质性强等先天条件,中国未来建设大规模水层CCS将面临封存规模大、封存成本高和长期封存安全性等3个方面关键技术的挑战;③从推动和形成CCS负碳新产业角度,中国未来水层CCS碳埋存需要攻克CCS地质体评估、CO_(2)高速注入、低浓度CO_(2)封存、降低成本和地质监测等5个关键技术。结论认为,该研究认识进一步坚定了中国发展水层CCS技术的信心,提出的5个方面关键技术为快速形成负碳领域的国家战略科技力量明确了攻关方向,同时为后续推动大规模地下储氢、储氦、压缩空气储能等战略性新兴行业的发展提供了前瞻性技术支撑。

基于双尺度海谱的动态海面场景仿真建模分析

针对典型复杂环境下动态海面目标场景,采用双尺度海谱用于风驱海面的建模构建公里级别复杂场景电磁仿真模型,研究面向体-面复合结构的自适应弹跳射线法,结合区域分解和并行加速技术实现典型目标及地物环境的耦合电磁散射特性分析,形成公里级别复杂场景多维复合散射特性仿真能力,实现大场景复杂环境与目标一体化模型的高效电磁散射计算与分析。通过测量海面邮轮在实际海面背景下的电磁散射特性,进行验证,为复杂海面背景下大尺度目标的动态场景生成和目标特性研究提供重要支撑作用。

面向城市复杂场景的多尺度监督融合变化检测

城市复杂场景中,地物形状多样,光照和成像角度变化大会导致变化检测结果受到干扰。为解决这些问题,文章提出了一种双上下文多尺度监督融合的网络模型(dual context multi-scale supervised fusion network model,DCMSFNet)。首先,在编码部分使用双上下文增强模块获得地物丰富的全局上下文信息。在解码部分,采用级联的方法组合特征,然后通过自适应注意力模块捕捉不同尺度的变化关系,设计多尺度监督融合模块,增强深度网络融合,获得具有更高辨别能力的变化区域特征,将不同层级的输出结果与主网络的重构变化图融合形成最终的变化检测结果。该模型在LEVIR-CD和SYSU-CD变化检测数据集取得了较好的结果,F1-score分别提高了1.58%和2.17%,可更加精确识别复杂场景的变化区域,进一步减少无关因素引起的误检和漏检,且对目标地物边缘的检测更加平滑。

时间尺度上约束力学系统的Noether型绝热不变量

时间尺度微积分是近年来的研究热点之一,其不仅统一连续分析与离散分析,还能协助完成更复杂动力学系统的建模.Noether对称性方法是一种近代的积分方法,揭示了力学系统守恒量与其内在的动力学对称性之间的潜在关系,Noether对称性的摄动与绝热不变量和系统的可积性之间也有着密切的联系.时间尺度上约束力学系统的Noether对称性问题虽然已有学者研究,但是由于时间尺度微积分理论的不成熟,研究成果的深度及正确性均有待探究.文章的重点是探讨时间尺度上约束力学系统Noether对称性的摄动与绝热不变量,这其中包含了Lagrange系统、Hamilton系统以及Birkhoff系统.首先,探讨了3个受小扰动作用的系统,其Noether对称性的变化,并给出了相应的绝热不变量;然后提供了在无扰动条件下,3个系统的精确不变量.Lagrange系统得到的精确不变量与原有结果吻合,Hamilton系统和Birkhoff系统得到的精确不变量是新的.其次,时间尺度上的导数分为delta导数和nabla导数,由这两种导数所组成的系统,互为对偶系统.基于文章delta导数下所得到的结果,采用对偶原理的方法,给出了3个系统对偶空间的绝热不变量和精确不变量.最后,文末分别讨论了时间尺度上Kepler问题和Hojman-Urrutia问题的Noether型绝热不变量,从而借助例题对3个系统中所得到的结果和所采用的方法进行说明.

拉普拉斯卷积的双路径特征融合遥感图像智能解译方法

由于遥感图像存在多尺度变化和目标边缘模糊等问题,对其进行智能解译仍然是一项极具挑战性的工作。传统的语义分割方法在处理这些问题时存在局限性,难以有效捕捉全局和局部信息。针对上述问题,文中提出一种双路径特征融合分割方法 DFNet。首先,使用Swin Transformer作为主干提取全局语义特征,以处理像素之间的长距离依赖关系,从而促进对图像中不同区域相关性的理解;其次,将拉普拉斯卷积嵌入到空间分支,以捕获局部细节信息,加强目标地物边缘信息表达;最后,引入多尺度双向特征融合模块,充分利用图像中的全局和局部信息,以增强多尺度信息的获取能力。在实验中,使用了三个公开的高分辨率遥感图像数据集进行验证,并通过消融实验验证了所提模型不同模块的作用。实验结果表明,所提方法在Uavid数据集、Potsdam数据集、LoveDA数据集的mIoU达到了71.32%、85.58%、54.01%,提高了语义分割的性能,使分割结果更为精细。

基于双流YOLOv4的金属表面缺陷检测方法

目前有许多学者使用深度学习进行表面缺陷检测研究,由于这些研究大都沿用主流目标检测算法的思路,注重高级语义特征,而忽视了低级语义信息(色彩、形状)对表面缺陷检测的重要性,因此导致缺陷检测效果不够理想。为解决上述问题,提出了一种金属表面缺陷检测网络——双流YOLOv4网络,骨干网络分成两个分支,输入分为高分辨率图像和低分辨率图像,浅分支负责从高分辨率图像中提取低级特征,深分支负责从低分辨率图像中提取高级特征,通过削减两分支的层数和通道数来减少模型总参数量;为了强化低级语义特征,提出了一种树形多尺度融合方法(Tree-structured Multi-scale Feature Fusion Me-thod,TMFF),并设计了一个结合极化自注意力机制和空间金字塔池化的特征融合模块(Feature Fusion Module with Polarized Self-Attention Mechanism and Spatial Pyramid Pooling,FFM-PSASPP)应用到TMFF中。在东北大学热轧带表面缺陷数据集NEU-DET、金属表面缺陷数据集GC10-DET和伊莱特电饭煲内胆缺陷数据集Enaiter的测试集上对所提算法进行了测试,测得的map@50结果分别为0.80,0.66和0.57,相比大部分主流的用于缺陷检测的目标检测算法均有提升,且模型参数量仅为原YOLOv4的一半,速度与YOLOv4接近,可满足实际使用需求。

四川盆地中部地区震旦系大型碳酸盐岩气藏开发技术新进展

四川盆地中部地区(以下简称川中地区)震旦系灯影组气藏为大型古老深层岩溶风化壳型碳酸盐岩气藏,气藏整体表现为低孔隙度低渗透率,储集空间小尺度缝洞发育,储层非均质性强特征,气藏开发面临储层表征难、井位部署和效益开发难度大等系列挑战。为此,以川中地区安岳气田灯影组气藏为对象,通过地质工程一体技术化攻关,深化了对储层发育特征和成因机理的认识,提出了气藏开发新模式,创建形成了气藏高效开发的关键技术。研究结果表明:①灯影组储层主要为岩溶成因储层,具有“叠合岩溶差异控储”的发育特征;②岩溶储层非均质性强,“溶蚀相成因控储”可精细刻画储层的非均质性;③灯影组气藏宜采用“单井指标叠加论证气藏规模”的开发新模式;④创新形成了“双界面”岩溶古地貌恢复及定量刻画方法,并指导平面选区,其中Ⅰ+Ⅱ类井比例由评价期的40%提高到方案建设期的100%;⑤创建了小尺度缝洞识别、刻画及表征技术,并指导钻井轨迹设计,测试百万立方米气井比例由开发评价期的41.6%提高到建产期的60%;⑥配套完善了分段酸压工艺技术,高效动用了优质储层,并解放了低品位储层,储层渗透率由改造前的0.62 mD提高到酸压改造后的6.77 mD。结论认为,“十四五”期间,在上述开发技术的支持下,安岳气田由台缘带灯四段高效开发转向长期稳产和台内地区多层系效益开发,蓬莱含气区由单井高产转向区块高产和气藏规模建产,将强力支撑中国石油西南油气田公司“十四五”末天然气上产500×10^(8)m^(3)。

基于双路径编码的遥感建筑物图像分割方法

高分辨率遥感图像建筑物分割是遥感影像研究的热点之一,而高分辨率遥感图像中建筑物尺度多样容易导致错分割、漏分割和边界模糊。针对上述问题,基于U-Net网络结构提出了一种双路径编码的遥感建筑物图像分割网络(DCU-Net)。DCU-Net在U-Net上加入一条并行编码路径,形成双路径编码结构。在编码阶段设计了密集残差编码模块(DRCM)和多尺度空洞卷积编码模块(MDCCM)以增强多尺度特征提取。在网络中加入双路融合注意力模块(DFAM),增强网络对特征的表达能力。为验证网络有效性,在WHU与Massachusetts数据集上进行实验,召回率、F1分数和交并比指标在WHU上达到91.26%、92.33%和86.15%,在Massachusetts Buildings上达到81.64%、84.33%和82.72%。结果表明,DCU-Net对于不同尺度的建筑物提取有较高的提取精度。

基于多分支空谱特征增强的高光谱图像分类

为了解决高光谱图像自身及分类过程中噪声干扰大、空间-光谱特征信息提取不足以及有限样本下分类性能不佳等问题,提出一种基于多分支空谱特征增强的高光谱图像分类模型SSFE-MBACNN。首先,利用多分支特征提取模块分别提取浅层空谱特征和深层空间特征信息,并引入注意力机制抑制噪声干扰。其次,设计一种改进多尺度空谱特征提取融合模块及结合双池化和空洞卷积的空间特征增强模块实现空谱特征增强,减少模型参数量和提高分类性能。最后,用全局平均池化层代替全连接层,进一步降低参数量,缓解模型过拟合问题。实验结果表明,在Indian Pines(10%训练样本)、Pavia University (5%训练样本)和Salinas(1%训练样本)数据集分别取得了0.990 7、0.997 5和0.994 7的总体分类精度。SSFE-MBACNN不仅能充分利用空谱特征信息,而且在有限样本下也取得了优秀的分类性能,明显高于其他对比方法。

基于改进Swin Transformer的遥感图像语义分割方法

在高分辨率的遥感图像中提取出准确的地物信息对城市规划以及土地资源利用有重要作用。然而,遥感图像具有目标物体之间尺度差异大,背景复杂等特点,易导致提取结果不准确,特别是对小尺度地物的提取精度较低。为了解决这些问题,提出一种新型双编码结构,充分获取全局语义信息以及空间细节信息,分阶段融合不同尺度的特征信息,增强特征表示能力。构造了特征加强模块(FEM),以减少下采样中细节信息的丢失,关注更多小尺度特征。为了更好地细化特征信息,融合了通道注意力和内核注意力后进行上采样,能够将局部特征与对应的全局空间依赖关系整合,提升目标物体的分割精度。在Potsdam数据集和Vaihingen数据集上的mIoU分别为86.1%和82.4%,与流行的语义分割模型进行对比分析,结果表明,该方法能够有效解决遥感图像中小尺度物体分割不准确的问题,适合处理遥感图像语义分割任务。

基于再感知双模型联合训练的散焦模糊检测

针对散焦模糊检测(defocus blur etection, DBD)模型训练时没有对响应错误区域学习优化,且在识别过程中部分图像的模糊同质区域和处理边界过渡等位置仍然具有挑战性等问题,提出了再感知双模型联合训练方法和基于注意力机制多尺度语义融合散焦模糊检测网络。将未正确响应的预测区域映射到全新合成图像中驱动模型学习,实现再感知错误位置的图像特征;利用DBD任务的互补性质构建预测网络,组成对焦预测和模糊预测双模型,将互补网络中多余响应的区域反馈到另一个模型上从而提升训练效果;利用多尺度特征融合模块逐渐整合不同尺度的语义信息;在特征提取时设计了全局通道注意力模块,使模型关注预测结果的有效特征信息,增强网络在不同输入场景下的灵活性。在DUT、CUHK和CTCUG数据集上进行的对比实验表明,提出的方法与对比方法中性能最优者相比,F-Measure指标分别提高了0.082、0.051、0.264,MAE指标分别降低了0.032、0.018、0.144。

基于双注意力机制的MSCN-BiGRU的滚动轴承故障诊断方法

针对滚动轴承故障诊断模型在变工况和环境噪声干扰下诊断精度降低的问题,提出一种基于双注意力机制的多尺度卷积网络(dual attention and multi-scale convolutional networks,DAMSCN)与改进的双向门控循环单元(bidirectional gated recurrent unit,BiGRU)组成的故障诊断模型DAMSCN-BiGRU。首先,多尺度特征融合模块使用不同大小的卷积核,获得多种感受野,从而提取到轴承原始振动信号的多尺度特征信息,并根据重要性对其进行自适应融合,然后利用通道注意力和空间注意力组成的双注意力模块(dual attention module,DAM)对多尺度特征进行重新标定,分配注意力权重,削弱融合特征中的冗余特征;然后,增加注意力层和利用分段激活改进BiGRU进而挖掘信号的时域特征,以提高轴承故障诊断的性能;最后,通过Softmax层完成对不同故障的分类。试验结果表明,与其他智能诊断模型相比,DAMSCN-BiGRU在变工况环境下,平均诊断精度达到98.2%,在强噪声背景下仍然有着85.3%的准确率,且在不同程度的噪声强度下效果均优于其他常用模型,有利于促进滚动轴承的智能故障诊断研究和实际应用。