小震级地震事件的倒谱差异

从数学、几何及物理方面,简要介绍了地震事件震源倒谱分析和震源性质判别的基本原理和方法。将该方法应用于小震级的地震与爆炸事件的分析,结果表明,朝鲜半岛6次较小地震与爆炸的震源倒谱具有显著的峰值差异,倒谱差异的定量描述C值参量分别大于和小于1。     

识别震源性质的一种新方法

提出了利用地震事件的震源差异在倒谱域所呈现的不同特征识别震源性质的新方法。识别结果表明新方法是有效的 ,展示出良好的应用前景     

使用地面区域台阵识别水下浅层化学爆破

区域地震台阵NORESS和FINESA记录到的波罗的海波斯尼亚湾以及北海的地震事件具有声观测到的水下爆破特征。对这些地震的区域震相的谱分析。揭示出很强的与时间无关的频谱边缘特征。表明波形是由相关的“脉冲响应”型信号组成。倒谱逆分析给出的脉冲响应对滞后时间与最常见的介于400－600ms的滞后是一致的，一些倒谱在150－350ms之间还有负峰。倒谱峰在与事件有关的所有震相中都被观测到，并且这两个阵所记录到的地震事件的延迟时间都相同。相对长的时间延迟与水声观测记录到的水下爆破中普遍存在的气泡脉冲延迟是一致的。低频负峰是由相对于主脉冲而言极性逆转了回波造成的，并与从水列顶部的反射一致。一般而言，这些事件的Pn/Lg振幅比较大，与采石场爆破所观测到的相似。波罗的海事件的高频Pn/Lg比值比北海的要小些，这可能是由于这两个地区水底地质构造的差异引起的。一些简单的水下爆破模型可以解释观测到的倒谱的许多特征，并揭示了75－150kg的爆破规模的爆破源较浅（约40－80m)。本研究表明，全面禁止核试验条约（CTBT）所要求布置的地震台网，可以对用于近海环境水下核爆破试验的检测和识别的水声观测台站的小网形成补充。     

基于倒谱的2015年天津港“8·12”爆炸分析

本文通过对记录到2015年天津港“8·12”爆炸的32个台站的三分量数字地震波形进行倒谱分析,得到了以下结论:①2个主要爆炸的发生时间间隔约为32.3s;②爆炸-2发生在爆炸-1的西北侧约353°处;③依据这些记录的倒谱,无法判定在爆炸-1前是否还存在微小的爆炸。以上结论均与前人研究成果吻合。由于倒谱叠加采用的是全波形,对滤波频带不敏感,因此,在检测发生在同一地点的多次爆炸或其它类似事件（如核爆）上有一定的优势。     

震源性质的倒谱分析ht

介绍了倒谱的概念．从天然地震与人工爆炸的震源差异入手，推断震源差异在各域的反映形式，探索表征这种震源差异的有效途径．将震源识别的研究方法从时间域、频率域扩展到倒谱域，提出了识别震源性质的倒谱分析方法，建立了一个震源类型的识别判据．对近年来发生的一些地震及爆炸事件进行了倒谱分析，结果表明该方法具有良好的效能．      

利用倒谱辅助识别终止相——以9·21集集地震为例

应用倒谱对终止相识别的理论依据,根据厦门地震遥测中心记录的9·21集集地震进行实例分析,认为倒谱分析可以辅助识别终止相,结合传统的终止相走时特征和动力学特征识别方法,进而提高终止相识别的准确度。     

基于MFCC样本熵和灰狼算法优化支持向量机的天然地震与人工爆破自动识别

针对天然地震与人工爆破波形特征相似、难以区分的情况,结合灰狼优化算法和支持向量机,提出一种地震事件性质辨识新方法。通过梅尔频率倒谱系数法对2013年四川芦山地震地震事件信号和人工爆破信号进行分析,进过预加重、FFT、梅尔滤波及离散余弦变换等步骤,提取静态系数样本熵、一阶差分系数样本熵和二阶差分系数样本熵等作为样本特征集。使用灰狼算法优化支持向量机径向基核函数RBF中的惩罚系数和核函数半径形成新的GWO-SVM分类器,然后对事件进行辨识。结果表明：GWO-SVM分类器辨识效果明显优于SVM、RobustBoost集成学习、LDA、PLDA等分类器,其在1 000次循环识别实验下的准确率均值相对SVM提高了9.2个百分点,标准差降低了3.2以上;t检验证明MFCC样本熵各特征具有可靠的地震事件分类效果;GWO-SVM与MFCC样本熵可作为天然地震事件与人工爆破事件的辨识方法与分类判据。     

震级——频度关系的b值谱及其在地震预报中的应用

本文引进了震级——频度关系的b值谱的概念,给出了计算b值谱的矩方法,对我国华北和西北地区分别计算和分析了以半年为单位的不同时段M_L≥3.0地震的b值谱线。所得结果表明,在研究区内发生6级左右以上大震前半年至一年半之内,b值谱的形态要出现较明显的前兆异常变化,因而可在地震预报中加以应用。     

2009年3月25日黄海M_L 4.0地震震源机制的讨论

应用振幅谱反演方法,对2009年3月25日黄海M_L 4.0地震的震源机制进行求解。在震源位置固定情况下,得到此次地震的地震矩为7.17×10~(14)N·m,矩震级约3.8级,最佳震源机制解为:走向40°,倾角80°,滑动角150°,与震中附近区域的中国断层线走向基本吻合。     

地震分倒频处理技术

为刻画地震信号的局部层次特征,提高复杂地质体的地震分辨率,本文利用倒谱变换能够提高分辨率的特点,结合短时傅里叶变换,构造了倒时-频变换算法,并提出地震分倒频处理技术.文中对地震分倒频处理技术的公式推导、算法设计、分倒频处理技术提高地震纵向分辨率的机理进行了详细研究.设计出了适合三维地震资料处理的软件,并从理论上阐述了地震分倒频处理技术能够提高地震分辨率.利用倒时-频变换对多分量线性调谐信号模型做倒时-频分析,可以得到良好的时频分辨率.实际三维地震资料计算表明,该方法能够提高地震纵向分辨率,且计算效率高,适合大规模三维地震资料的处理.     

Cepstrum analysis of seismic source characteristics

This paper introduces the concept of cepstrum.By investigating the difference in source characteristics between earthquakes and explosions the paper infers the manifestation of source difference in various variable domains,and seeks for effective means to express such source difference.Extending the approach of source discrimination from time and frequency domain to the cepstrum domain,the paper proposes a method of cepstrum analysis for recognizing the characteristics of seismic sources and establishes criteria for identifying the type of seismic sources.Cepstrum analysis on some recent earthquakes and explosions has been made,and the result shows that the method is quite effective in practice.     

Initial Experimental Result of Accurately Controlled Routinely Operated Signal System （ACROSS）

The principle of Accurately Controlled Routinely Operated Signal System (ACROSS) is introduced in this paper, A sample machine is made and tested. The experiment shows that the signal-stacking technique is effective in improving signal-to-noise ratio and the sompi cepstrum method is applicable to deconvolute a set of travel times of wave elements from accurate transfer function data in frequency domain.     

只测天然水平电场提取地下各深度信息的一种可能方案

一、引言 对于探测深度大于几十米的地下的电波传播理论研究,从电磁感应的原理出发是最方便的.由于有了无噪声情况下地电反演的唯一性定理,从地面通过测量天然宽频带电场和磁场探测地下较细微的电性结构并不是不可能的.现行的地下电磁探测的主要问题之一是需要提高分辨力.一种可能的思路是试图将以前测不到的弱信息测出来.利用精密电子技术,先将微弱信息与强信息分离,再通过信号检测技术（例如多次叠加）将弱信息从各种干扰噪声中提取出来.在地电探测领域,关键之一是要解决弱信息与强信息分离的方法.本文就是探索分离水平电场中各层次信息的方法.假如水平磁场也同时测量,并也采用本文的方案进行处理,则可能得到改进的大地电磁法. 二、输入阻抗模值的模拟反演公式 弱信息与强信息的分离,相当于用模拟电路完成通常用数字处理完成的反演处理,因而需要简易的反演公式.只测电场的反演需要输入阻抗模值的简易反演公式,这里只给出结果,推导见文献1). 类似于地震学中的Goupilloud方法,可以建立起一种导电介质反演的理论公式,在这些公式基础上,可以推导出输入阻抗模值的反演的倒谱公式,从倒谱公式出发便得到适合于模拟反演的公式. 设地面所测输入阻抗为 ρ（ω）=E_x（ω,0）/H_y（ω,0）,（1） 第一层波阻抗为 ρ₁（ω）=(     

Wavelet‐based cepstrum decomposition of seismic data and its application in hydrocarbon detection

How to use cepstrum analysis for reservoir characterization and hydrocarbon detection is an initial question of great interest to exploration seismologists. In this paper, wavelet‐based cepstrum decomposition is proposed as a valid technology for enhancing geophysical responses in specific frequency bands, in the same way as traditional spectrum decomposition methods do. The calculation of wavelet‐based cepstrum decomposition, which decomposes the original seismic volume into a series of common quefrency volumes, employs a sliding window to move over each seismic trace sample by sample. The key factor in wavelet‐based cepstrum decomposition is the selection of the sliding‐window length as it limits the frequency ranges of the common quefrency section. Comparison of the wavelet‐based cepstrum decomposition with traditional spectrum decomposition methods, such as short‐time Fourier transform and wavelet transform, is conducted to demonstrate the effectiveness of the wavelet‐based cepstrum decomposition and the relation between these two technologies. In hydrocarbon detection, seismic amplitude anomalies are detected using wavelet‐based cepstrum decomposition by utilizing the first and second common quefrency sections. This reduces the burden of needing dozens of seismic volumes to represent the response to different mono‐frequency sections in the interpretation of spectrum decomposition in conventional spectrum decomposition methods. The model test and the application of real data acquired from the Sulige gas field in the Ordos Basin, China, confirm the effectiveness of the seismic amplitude anomaly section using wavelet‐based cepstrum decomposition for discerning the strong amplitude anomalies at a particular quefrency buried in the broadband seismic response. Wavelet‐based cepstrum decomposition provides a new method for measuring the instantaneous cepstrum properties of a reservoir and offers a new field of processing and interpretation of seismic reflection data.     

基于线性预测倒谱系数的地震相分析

本文借鉴语音识别技术中的线性预测倒谱系数(LPCC系数)特征参数提取方法对地震数据进行分解,这种方法的优点是:可以获得将子波和反射系数信息分离的地震语音特征参数,对地质现象边界具有较好的描述能力,使我们可以从不同维度更细致地观察隐藏在地震数据中的地质特征.理论模型分析表明,基于LPCC系数的地震分析具有较高的地震相划分能力.实际地震资料应用表明,LPCC系数对储层特征的描述比常规三瞬属性更为细致,不同阶次LPCC系数在描述储层不同特征时也保持了内在的联系.采用K均值聚类方法对提取的12阶和24阶LPCC系数进行聚类分析,聚类结果与目的层段古地形较为吻合,较好地反映了研究区的断裂、礁滩相带、深水扇和储层的分布特征,说明在地震相分析中采用LPCC系数作为特征参数是可行和有效的.     

Source characterisation of low frequency events at Stromboli and Vulcano Islands (Isole Eolie Italy)

The discussion about the source of low frequency events generally wrecorded on active volcanoes is still open and needs deeper understanding of the phenomena involved in their generation. Most of the models view such a phenomenon as a source effect (oscillation of volcanic fluids in conduits or cracks), although a different explanation as a path or site effect exists. In the present paper we analysed 26 seismic signals recorded at Vulcano and 60 at Stromboli in order to put some constraints on the functional shape of the recorded signals. They evidence all the characteristics of the low frequency events. The spectral analysis reveals sharp peaks in the range 0.5–4 Hz, while the cepstra suggest that the signals are composed by a two-sine kind function. This suggestion is confirmed by the two-dimensional projections of the reconstructed phase space. The correlation dimension of the attractor is very close to 3 for both the volcanoes confirming the existence of a toroidal structure in the phase space. This implies that we can suggest as a model for the low frequency events source a physical mechanism very similar to a Duffing oscillator or any other quasiperiodic one. In particular in the cases of the analysed volcanoes we recognise two independent oscillations.     

HOMOMORPHIC FILTERING — THEORY AND PRACTICE*

The application of homomorphic filtering in marine seismic reflection work is investigated with the aims to achieve the estimation of the basic wavelet, the wavelet deconvolution and the elimination of multiples. Each of these deconvolution problems can be subdivided into two parts: The first problem is the detection of those parts in the cepstrum which ought to be suppressed in processing. The second part includes the actual filtering process and the problem of minimizing the random noise which generally is enhanced during the homomorphic procedure. The application of homomorphic filters to synthetic seismograms and air-gun measurements shows the possibilities for the practical application of the method as well as the critical parameters which determine the quality of the results. These parameters are:

• a) the signal-to-noise ratio (SNR) of the input data
• b) the window width and the cepstrum components for the separation of the individual parts
• c) the time invariance of the signal in the trace.

In the presence of random noise the power cepstrum is most efficient for the detection of wavelet arrival times. For wavelet estimation, overlapping signals can be detected with the power cepstrum up to a SNR of three. In comparison with this, the detection of long period multiples is much more complicated. While the exact determination of the water reverberation arrival times can be realized with the power cepstrum up to a multiples-to-primaries ratio of three to five, the detection of the internal multiples is generally not possible, since for these multiples this threshold value of detectibility and arrival time determination is generally not realized. For wavelet estimation, comb filtering of the complex cepstrum is most valuable. The wavelet estimation gives no problems up to a SNR of ten. Even in the presence of larger noise a reasonable estimation can be obtained up to a SNR of five by filtering the phase spectrum during the computation of the complex cepstrum. In contrast to this, the successful application of the method for the multiple reduction is confined to a SNR of ten, since the filtering of the phase spectrum for noise reduction cannot be applied. Even if the threshold results are empirical, they show the limits fór the successful application of the method.