自然时间域中地震序列参数及地电场数据分析

发布时间：2018-04-11 00:16

本文选题：自然时间分析 + 地震电信号　；参考：《中国地震局兰州地震研究所》2017年硕士论文

【摘要】：自然时间理论是一门新兴理论,由Varotsos P A等人于2001年提出的自然时间域,是对时间模型的一种新观点。该观点认为时间不连续,在形式上区别于传统的时间域—由实数集组成的一维连续模型。将复杂系统时间序列在自然时间域中进行分析,能够更好地揭示时间序列的动态特征,提高分析过程准确率,提取更多有用的信号。利用自然时间域还可以研究复杂系统的动态变化过程,在自然时间域中计算出时间序列的归一化功率谱Π(ω),当自然频率ω趋近于0时,对Π(ω)进行泰勒展开,得到Π(ω)≈1-κ1ω2,系数κ1恰好等于自然时间序列的方差,这一特征可用于判断复杂系统是否进入临界阶段。地震的三要素(时间、地点和震级)之间具有复杂联系,地震能量的释放以及震前电信号的产生具有临界特性。对地震目录进行自然时间分析时,将方差κ1作为序列参数。结合Varotsos PA和SarlisNV等人研究成果,在主震发生前的几小时至数月,震中区域的κ1值会接近0.07,其变异系数会呈现一个小值,这对于我们缩短地震预报窗口很有利用价值。本文在对自然时间方法理论,以及地震电信号的产生机理和识别方法等阐述的基础上,对我国地震目录资料及地电场观测进行了自然时间分析,得到了一些有意义的结果:(1)选取地震目录中发生在中国及周边地区震级前30位的地震,将所选震例的前震序列转化至自然时间域,计算每次地震的归一化能量在自然时间域中的方差κ1及其变异系数β,将方差κ1作为地震序列参数,利用变异系数β对序列参数的波动性进行了定性分析,在此基础上,又对κ1进行了定量分析。研究发现,在分析的30次震例中有21次从震前半小时至几月开始,地震序列参数会具有持续在0.07上下波动,并在主震发生前迅速下降至0附近的特点。还对定量分析中时间窗口的选择以及分析结果中出现的特殊情况进行了简单讨论。相关研究结果对于我国地震事件的回溯性研究以及未来地震预测研究具有较一定的指导意义。(2)介绍地震电信号的物理性质及产生机制,给出了地震电信号具有临界性的理论依据。引入自然时间理论,对中法合作天祝电磁台记录到的地震电信号及人工干扰信号进行了自然时间分析,得到上述两种信号及均匀分布在自然时间域中方差、熵和归一化功率谱间的关系。该结论为地震电信号的识别提供了理论和技术支持。
[Abstract]:Natural time theory is a new theory, which was put forward by Varotsos PA et al in 2001. It is a new view of time model.According to this view, the time discontinuity is different from the traditional time domain in form, which is a one-dimensional continuous model composed of real number sets.By analyzing the time series of complex systems in the natural time domain, we can better reveal the dynamic characteristics of the time series, improve the accuracy of the analysis process, and extract more useful signals.The dynamic process of complex system can also be studied by using natural time domain. In the natural time domain, the normalized power spectrum of time series is calculated, and when the natural frequency 蠅 approaches 0, the Taylor expansion is carried out for the 蟺 (蠅).The coefficient 魏 1 is exactly equal to the variance of the natural time series, which can be used to judge whether the complex system has entered the critical stage.The three elements of an earthquake (time, place and magnitude) have complex relationships, and the release of seismic energy and the generation of electrical signals before earthquakes have critical characteristics.In the natural time analysis of the earthquake catalogue, the variance 魏 1 is taken as the sequence parameter.Combined with the research results of Varotsos PA and SarlisNV, the value of 魏 1 in the epicenter region will be close to 0.07 from a few hours to a few months before the occurrence of the main earthquake, and the coefficient of variation will present a small value, which is of great value for us to shorten the earthquake prediction window.Based on the theory of natural time method, the generation mechanism and identification method of seismic electrical signals, the natural time analysis of earthquake catalogue data and geoelectric field observation in China is carried out in this paper.Some meaningful results are obtained: (1) selecting the earthquake occurring in the first 30 places of magnitude in China and its surrounding areas in the earthquake catalogue, and transforming the foreshock sequence of the selected earthquake example into natural time domain.The normalized energy of each earthquake is calculated in the natural time domain, and the variance 魏 1 and its coefficient of variation 尾 are calculated. The variance 魏 1 is taken as the parameter of the earthquake sequence, and the volatility of the sequence parameters is analyzed qualitatively by using the coefficient of variation 尾.The quantitative analysis of 魏 1 was also carried out.It is found that 21 of the 30 earthquake cases analysed from half an hour to a few months before the earthquake, the parameters of the earthquake sequence will fluctuate continuously in the range of 0.07 or so, and will drop rapidly to near zero before the occurrence of the main shock.The selection of time window in quantitative analysis and the special situation in the analysis result are also discussed briefly.The related research results have certain guiding significance for the backtracking study of earthquake events in China and the study of earthquake prediction in the future. It also introduces the physical properties and generating mechanism of seismic electrical signals.The theoretical basis for the criticality of seismic signals is given.By introducing the theory of natural time, the natural time analysis of seismic signal and artificial interference signal recorded by Tianzhu electromagnetic Station of Sino-French Cooperation is carried out, and the variance of these two signals and their uniform distribution in natural time domain are obtained.The relationship between entropy and normalized power spectrum.This conclusion provides theoretical and technical support for seismic signal recognition.
【学位授予单位】：中国地震局兰州地震研究所
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P315.722

【参考文献】