首页 | 本学科首页   官方微博 | 高级检索  
相似文献
 共查询到20条相似文献,搜索用时 62 毫秒
1.
闪电哨声波是由闪电激发的宽频带电磁脉冲,在穿越南北两半球沿地磁线传播时携带了大量的空间环境信息.由于在闪电哨声波中高频的部分先到达低频的部分后到达,导致其在时频图中呈现色散状.其传播路径的长短、传播路径中的电子密度成分以及反射次数等不同使得其存在多种扩散状态.为了进一步探索闪电哨声波的扩散状态与空间环境之间的关系,将闪电哨声波依据其扩散状态进行分类是现在一个亟需解决的问题.然而,目前的检测算法还无法对闪电哨声波扩散状态进行细分.基于此,本文以张衡一号(ZH-1)卫星感应磁力仪(SCM)的VLF波段的波形数据为研究对象,首次将闪电哨声波的散度信息作为其特征,统计并制作了2303张闪电哨声波的时频图像数据集并对其进行特征提取,最后将提取到的特征输入SVM分类器对闪电哨声波进行分类.实验结果表明该算法具有较好的分类效果,其准确率为94.35%、精确率为95.42%、召回率为98.58%、F1-Score为96.97%.  相似文献   

2.
闪电哨声波是一种重要的电磁波动,了解其传播特征及传播过程有助于揭开圈层电磁耦合机理.从卫星观测资料识别闪电哨声波通常需要将原始电磁波形进行滤波处理再转化为时频图像,最后采用目视方法识别图像中的色散状形态,整个过程消耗大量人机时间和内存资源,不能满足张衡一号(ZH-1)卫星观测的海量电磁场数据处理的需求.针对该问题,鉴于...  相似文献   

3.
闪电哨声波作为探索空间物理环境的重要媒介,淹没在海量的电磁卫星数据中.近年来随着计算机视觉和深度学习等人工智能技术的发展,从电磁卫星的存档数据中自动检测闪电哨声波的算法取得了一定的效果.本文对近年来闪电哨声波智能检测算法的文献进行了整理和总结.首先,阐述闪电哨声波在电磁卫星数据中呈现的时频特征和类型;然后,介绍了闪电哨声波智能检测算法的流程并从闪电哨声波的特征提取、分类和定位三个方面对主要的智能检测算法进行归纳、总结和评述;其次,简述了闪电哨声波智能检测模型的评价指标;接着,在张衡一号(ZH-1)卫星的磁场数据上对三种典型的闪电哨声波智能检测算法进行复现,并对三种算法的优缺点进行了较深入的分析;最后,对基于电磁卫星的闪电哨声波智能检测的研究领域进行总结和展望.  相似文献   

4.
频高图F层异常分层(增层)是指在常规电离层结构的基础上,由于电子密度被扰动而在频高图中出现多层回波结构,该F层异常分层(增层)现象在白天和晚上均可观测到.电离层异常分层(增层)主要由某种电动力学或者突发电离源导致电子密度出现扰动而形成,已经引起国内外学者的广泛关注和研究.目前从频高图中识别F层异常分层(增层)主要还是通过人工的方法,海量的频高图数据使该方法在研究F层异常分层(增层)统计特征时变得非常困难.有别于传统的图像分类算法中需要对图像特征进行提取,本文通过引入卷积神经网络(Convolution Neural Network, CNN),设计开发出一种能够自动识别存在F层异常分层(增层)频高图的方法,该方法可以直接输入频高图样本数据,省去了复杂的图像特征提取过程.本文设计的卷积神经网络经过训练以后F层异常分层(增层)识别率为85.82%,准确率90.36%.实验结果表明该卷积神经网络可以较好地自动识别出F层异常分层(增层)的频高图.  相似文献   

5.
脱空缺陷在水泥路面中不可避免,严重影响水泥路面的结构安全性,迫切需要建立脱空缺陷早期识别定位方法,为路面精准预养护提供依据.根据探地雷达(GPR)数据的成像原理,提出了标准化和去背景的脱空特征增强算法.依据GPR发射Ricker子波的工作特点,借鉴深度学习中卷积网络的特征提取特性,以A-Scan数据集为对象,设计了提取脱空特征的二维卷积核,并提出基于卷积核和阈值判断的脱空定位算法.采用数值模拟和室内验证相结合的方法,研究了不同卷积尺度的影响,并将设计的卷积识别算法在水泥路面进行现场验证.实验结果表明,提出的标准化和去背景的预处理算法可替换传统的GPR多步骤数据处理的算法组合,有效实现脱空区域的特征增强作用;卷积算法有效突出了脱空特征,卷积核尺度过大会降低图谱分辨率,卷积尺度过小无法突出脱空特征,合适卷积纵向尺寸为1~1.5倍fs/f0,横向尺寸根据检测对象一般取对应实际距离10~20 cm的采样点.实验统计方法设置了卷积核判断阈值,可有效确定脱空区域的深度以及判断脱空区域是否含水,为水泥路面病害的自动识别奠定基础.  相似文献   

6.
小地震与人工爆破记录的时频分析   总被引:2,自引:0,他引:2       下载免费PDF全文
基于Matlab软件,编制了对近震记录进行时-频分析试验研究的程序,搜集2006—2008年间首都圈地区震级介于ML2.4~2.7的5次天然地震和5次人工爆破的27个台站的宽频带记录,根据震中距离的相似性挑选5组天然地震和人工爆破地震,分析其时频特征。对于所有的数据都给出其三分向的时频图、原始波形图和频谱图,以及水平向和三分向的叠加图。分析结果表明,天然地震和人工爆破的记录具有明显不同的时频特征,天然地震的时频呈现出"多峰"特征,而人工爆破的时频分布则相对"少峰";天然地震的时频峰值分布在较宽的频率范围内,人工爆破的峰值则分布在频率较低的相对狭窄的范围内。通过以上实践,发现时频分析在分析不同时间点的频谱信息以及整个数据的时频分布特点时比傅里叶变换具有优势。  相似文献   

7.
地震勘探信号时频分析方法对比与应用分析   总被引:16,自引:7,他引:9       下载免费PDF全文
针对地震勘探信号,对非平稳信号时频分析几种适效方法:短时Fourier变换、小波变换、S变换、Wigner分布、平滑伪Wigner分布、锥形核时频分布、AOK(adaptive optimum kernel,自适应最优核函数)分布等进行对比与应用研究.在阐明各种方法基本原理的基础上,进行数值分析与应用研究.首先对非平稳地震勘探模拟信号进行试算及时频属性提取,结合各类方法的信号表示理论,在时频局部化的精度和交叉项抑制等方面对计算结果进行对比分析;进一步应用于实际二维地震数据,提取瞬时频率和瞬时带宽等时频属性,进行比较研究.研究表明:对于地震勘探信号,就本文涉及的几种时频分析方法而言,AOK分布是时频局部化精度最高、交叉项抑制最好、时频匹配最优的方法,值得在地震勘探信号分析和地震属性提取、频谱分解等应用中深入研究和应用.  相似文献   

8.
小波变换作为一种具有多分辨率特征的时频分析方法已被地球物理学家广泛应用,受不确定性原理影响,其时频分辨率有限.能量重排小波变换通过对小波变换能谱向其局部重心进行重新分配,进一步提高了时频谱图的精度.拟合信号和实际地震数据表明,利用该方法能够更准确地刻画信号的时频特征,有助于获得更好的解释效果.  相似文献   

9.
地震信号分类——即信号震源类型的分类,尤其是天然地震和人工爆破的分类对于地震目录的清洗和强震实时预警等具有重要意义.本文首先对原始波形信号进行必要的预处理,然后对预处理后的信号进行分帧加窗,对窗内信号采用短时傅里叶变换,将原始波形信号从时域转换成时频域信号,生成时频谱图,每条波形生成一个时频图像;经过反复多次试验比较,将原尺寸时频谱图统一缩放为32×32像素大小的灰度图像,该灰度图像作为卷积神经网络(CNN)的输入,此时的分类效果和计算效率最优.采用以上方法对2010年至2016年发生在河北三河和北京通州等地区,震级为1.5 ~2.8级之间的54个天然地震事件的1674条波形和63个人工爆破事件的1509条波形,采用五折交叉验证法进行分类,得到的平均准确率为97.39%,与传统的支持向量机(SVM)方法和多层感知器(MLP)方法对比,分类准确率有大幅提升,与使用梅尔倒谱系数(MFCC)构建的CNN分类器方法对比,本文方法信噪比更低,分类准确率提高约1.5%.实验结果表明,采用短时傅里叶变换提取地震信号时频域特征,生成时频谱图,使用卷积神经网络对地震信号进行分类具有良好的分类效果.  相似文献   

10.
地震数据通常存在数据缺失问题,严重影响地震数据各个处理环节,需采用适当的手段对其重构.本文提出了一种基于深度学习卷积神经网络(CNN)的智能化地震数据插值技术.算法的关键在于构建一个适用于地震资料插值的CNN模型,该技术以缺失地震数据作为输入层,由卷积算法提取地震数据的特征信息,并通过池化层实现数据压缩降维,同时引入修正线性函数(ReLU)提高模型的非线性表达能力,再通过反卷积层恢复数据尺寸,最终搭建卷积自编码器模型(CAE),实现数据-数据的映射关系.该模型通过残差学习获得缺失数据特征并实现重构数据输出,与现有技术相比,该方法采用自监督学习方式,利用大量数据训练卷积自编码器模型,通过所得模型实现缺失地震道的数据重构.分别利用CAE模型及POCS插值技术对模型资料和实际数据进行插值,测试结果表明,CAE能有效实现地震数据插值,且与POCS方法相比具有更高的精度,验证了算法的可行性和有效性.  相似文献   

11.
The data on geomagnetic reversals are compared with the changes in the organic world and with the lower-mantle plumes. The times of the formation of plumes and the times of their appearance on the Earth’s surface relate to the intervals characterized by the different frequencies of geomagnetic reversals, i.e., there is no interrelation between the formation of plumes and the frequency of the changes in the geomagnetic field polarity. At the same time, a certain synchronism is observed between the frequency of the geomagnetic reversals and the boundaries of the biostratigraphic ages, i.e., the changes in the organic world in the long-period range. A hypothesis is proposed, which explains the change in the sign of the geomagnetic field by the combined effect of the irregular rotation of the internal core relative to the mantle and the changes in the slope angle of the axis of the Earth’s rotation, which, in turn, results in synchronous events on the Earth’s surface: the rates of changes in the organic world.  相似文献   

12.
Summary The author mentions the aims of the World-wide gravity project he established in the Ohio State University in Columbus, in 1950. He outlines the practical procedure of the gravimetric computations of the undulationsN and the vertical deflection components and and emphasizes that only by the global international cooperation and additional gravity observations at sea carried out during the last decade it has been possible to gather to Columbus the needed gravity material. Since there exist still large gravimetrically unsurveyed areas it is of vital significance to study what gravity anomalies are best to be used for these regions. The given figures concerning the accuracy of theN, and , estimated theoretically and obtained in practice, indicate that in the gravimetrically well surveyed parts of the world like in Europe and the United States we can get gravimetrically on basis of existing gravity material theN-values with accuracy of about 5–10 meters, and and with the accuracy of about 1. The geoid undulationsN are already computed in Columbus for more than 6000 points of the northern hemisphere. The sample maps show the interesting geoid of Europe and vicinity between the latitudes 60° and 30° and longitude 5° W and 30° E, drafted on basis of more than 1000N-values computed at the corners of 1°×1° squares. It is interesting to realize that the geoid undulations in all this area are positive, the extreme values being between 40 and 50 meters. The geophysical significance of the geoid maps of this kind is pointed out.  相似文献   

13.
Summary The maintenance of the axisymmetric component of the flow in the atmosphere is investigated by means of a steady-state, quasi-geostrophic formulation of the meteorological equations. It is shown that the meridional variations in the time-averaged axisymmetric variables can be expressed as the sum of three contributions, one being due to the eddy heat transport, another to the eddy momentum transport, and a third to the convective-radiative equilibrium temperature which enters the problem through the specification of a Newtonian form of diabatic heating. The contributions by the large scale eddies are evaluated through the use of observed values for the eddy heat and momentum transports.The contributions from each of the three forcing mechanisms to the temperature and zonal wind fields are invstigated individually and found to be of about equal importance. The sum of the three contributions are also presented for the temperature, the zonal wind, the stream function associated with the mean meridional circulation and the corresponding vertical motion. Although the results fail to reproduce the main observed features of the lower stratosphere, they are found to be in good agreement with observations in the middle latitude troposphere. At any pressure level, for example, the computed mean zonal wind has a jet-like profile and the axis of the jet is found to slope to the south with height, as observed in the atmosphere.Based in part on a thesis submitted by the first author as partial fulfillment of the requirements for the Ph.D. degree at the University of Michigan. — Publication No. 194 from the Department of Meteorology and Oceanography, The University of Michigan.  相似文献   

14.
15.
陈伯舫 《华南地震》2004,24(4):8-10
综合APIA、HONOLULU、PAMATAI 3个地磁台三分量的11年周期变化的形态后,可定性地提出该变化源于内场。如内源为核幔边界的电涡流,则其中心点应位于3台之间。  相似文献   

16.
The nutations of the planets Mars andEarth are investigated and compared. Alarge number of interior structureparameters are involved in the nutationcomputations. The comparison between the observations and the computationsprovides several constraints on these parmeters andtherefore allows a better understanding of the physics of the interior of theplanet. For the Earth, the high precision of the observations of the nutationshas led to a very good determination of interior properties of the planet. ForMars, observations of nutations are not yet available, and we review how theamplitude of the Martian nutations depends on the hypotheses consideredfor its interior. Although Mars is very similar to the Earth, its interior is not well known;for example, we don't knowif its core is liquid or solid. Only if the core is liquid,the Free Core Nutation (FCN) normal mode exists and can alter the nutationswhich are close to the resonance. From the observed geoids, it is known thatboth planets are not in hydrostatic equilibrium. The departure is larger forMars than for the Earth, and consequently, the implication of considering a convective mantle instead of a mantle in hydrostatic equilibrium described byClairaut's equation for the initial equilibrium state of the planet is largeron the Martian nutations than on the Earth nutations. The consequences of theuncertainty in the core dimensions are also examined and shown to be of a veryhigh influence for Mars if the core is liquid, due to the potential changes inthe FCN resonance. The influence of the presence of an inner core, which isknown to exist for the Earth, could be more important for Mars than for theEarth if the inner core is large. Due to the presence of Tharsis on Mars, thetriaxiality of this planet has, additionally, larger effects than on Earth.  相似文献   

17.
Until recently, the existing data prevented the geophysicists from accurately dating the Bysy-Yuryakh stratum, which outcrops in the middle reach of the Kotuy River, constraining the time of its formation to a wide interval from the end of the Late Cambrian to the beginning of the Silurian. The obtained paleomagnetic data unambiguously correlate the Bysy-Yuryakh stratum to the Nyaian regional stage and constrain its formation, at least a considerable part of it, by the Tremadocian. This result perfectly agrees with the data on the Bysy-Yuryakh conodonts studied in this work and yields a spectacular example of the successful application of paleomagnetic studies in solving important tasks of stratigraphy and, correspondingly, petroleum geology. Within the Bysy-Yuryakh stratum, we revealed a large normal-polarity interval corresponding to the long (>1 Ma) period when the geomagnetic reversals were absent. This result, in combination with the data for the Tremadocian and Middle–Upper Cambrian sequences of the other regions, indicates that (1) the rate of occurrence of the geomagnetic reversals on the eve of the Ordovician Moyero superchron of reversed polarity was at most one reversal per Ma; (2) the superchron does not switch on instantaneously but is preceded by a certain gradual change in the operation conditions of the dynamo mechanism which, inter alia, manifests itself by the reduction of the frequency of geomagnetic reversals with the approach of the superchron. This finding supports the views according to which a process preparing the establishment of the superchrons takes place at the core–mantle boundary.  相似文献   

18.
19.
The global database on the paleointensity, containing determinations of the virtual dipole moment (VDM) for a stable (normal) regime of the geomagnetic field in a time interval of up to 3.5 Ga, is supplemented by new VDM determinations and analyzed. The field generation process started no later than 3–3.5 Ga (earlier data are absent) at the stage of the Earth’s core formation. Since that time, the dipole value has differed from its present value by no more than an order of magnitude, and the deviations that have already been detected tend toward smaller values. The distribution of VDM values in the time interval 0–400 Ma is bimodal, which apparently reflects the presence of two different generation levels of the geomagnetic field distinguished by a relatively large value (close to the present field value) and a relatively small value (approximately half as large as the present value). The total duration of decreased VDM values appreciably exceeds that of increased VDM values (179.1 and 28.6 Myr, respectively). On the whole, data on the paleointensity do not contradict the hypothesis about the dipole nature of the field over the last 400 Myr; however, the number of determinations at high paleolatitudes is too small to draw decisive conclusions on the validity (or invalidity) of the dipole field approximation based solely on paleointensity data.  相似文献   

20.
A special high-resolution routine of the EISCAT radar has been used to investigate the structure and development of the poleward wall of a deep trough in electron density. The feature was tracked by the radar during a 7-hour period under very quiet geomagnetic conditions. The field-aligned nature of the structure enabled an estimate to be made of the inclination of the geomagnetic field above EISCAT that was in good agreement with the current model. Observations of narrow field-aligned enhancements in electron temperature demonstrated that the wall of this trough is a dynamic feature, reforming regularly as the electron density responds on a time scale of tens of minutes to energy input from soft-particle precipitation.  相似文献   

设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号