宽频带超导重力仪的研制进展

本文报道一种构建宽频带高分辨率超导重力仪（SG）的方法.与已商业化的GWR超导重力仪类似,新型超导重力仪同样使用磁悬浮超导检验质量来构建弹簧振子结构,不同的是弹簧振子的固有频率更高,可以有效覆盖地脉动频带.同时,利用基于超导量子干涉器件（SQUID）的位移传感技术,大幅提高了位移传感灵敏度,以保证仪器在测量带宽扩展之后仍具有与GWR仪器相当的测量分辨率,从而实现对时变重力信号与地球背景噪声信号的同时高分辨观测.宽频带超导重力仪核心部件垂向超导加速度计（VSA）的设计、组装和初步测试已经完成.结果表明,加速度计的噪声本底为8 μGal/Hz^1/2@0.1 Hz,实现了对地脉动信号的高分辨率测量.而为了将仪器噪声本底降低到GWR仪器水平,加速度计的固有频率需要进一步减小,该部分参数优化工作正在进行.本文将对垂向超导加速度计的工作原理、结构和测试结果进行详细介绍.     

Analysis of the Frequency Domain Water-level Deconvolution Method Based on Reservoir Airgun Source Data

Based on the 2016 airgun experimental data of the Fujian Nanyi reservoir, we adopted the frequency domain water-level deconvolution method and cross-correlation time delay detection technique to study the influence of level scaling factor and the background noise level of the station on deconvolution calculation results, and analyze the effect of deconvolution on eliminating the influence of the source caused by different air-gun pressures. The results show that:(1) When the level scaling factor is smaller, the signal to noise ratio of the waveform after the deconvolution is smaller, and when the level scaling factor is over smaller, the identification error of travel time is greater. (2) When the SNR of the station record is higher, the recognition accuracy of travel time is higher, the influence of SNR on the reference station record is far greater than the far station, when the SNR of the far station record is more than 10, the error of travel time is within 6ms, but when the SNR of the reference station record is 30, the travel time error may reach to 20ms. (3) When the airgun source difference is big, the frequency domain water-level deconvolution method has better effect on eliminating the source influence, but the method error may be introduced when the source difference is small.     

IMPULSE RESPONSE MODELS FOR NOISY VIBROSEIS† DATA*

A new method of Vibroseis deconvolution has been recently proposed by the authors. This discussion describes the effects of noise on the application of this method. The initial deconvolution step involves estimating the spectrum of the Vibroseis wavelet by homomorphic filtering. It is shown that noise causes problems with phase estimation. Hence, the Vibroseis wavelet is assumed to be zero phase. Examples demonstrate that zero phase cepstral filtering is a robust wavelet estimation approach for noisy data. The second step of the deconvolution method forms an impulse response model by a spectral extension method. Although this step can improve the resolution of seismic arrivals, it must be applied with caution in view of the deleterious effects of noise.     

提高自然伽马测井曲线分辨率的正则化方法

应用正则化反演理论研究了提高自然伽马测井曲线纵向分辨率的正则反褶积处理技术，通过正则化因子对噪音和信号影响的分析和研究，给出了估计反褶积结果与地层真值偏差大小的新公式，提出了直接根据信噪比大小选择正则化因子的新方法．由于正则化因子保证使反褶积结果与地层真值偏差为最小，使计算结果的稳定性和精度均得到明显提高．此外，大部分测井记录的信噪比均可以很容易估计，从而大大提高了正则反褶积的适用性．最后通过理论模型实验和井场数据处理证明了正则反褶积技术是十分有效的．     

自适应Kalman滤波反褶积的快速实现方法

提出了以二进小波变换为基础的自适应Kalman滤波反褶积(AKFD)新方法，针对该方法的计算复杂程度，提出了一种快速实现方法.二进小波变换的AKFD抛弃了传统预测反褶积对信号平稳性的假设，克服了提高分辨率而信噪比明显降低的问题，具有很好的抗噪性能.在小波域进行的AKFD在压制假反射以及提高分辨率方面比时间域的AKFD好，克服了在时域内进行AKFD抬升低频成分的缺陷.利用二维地震数据的局部平稳性的假设提出了快速实现方法，通过分段求取自适应预测算子，分别于横向及纵向采用样条插值的方法进行插值，来减少求取自适应预测算子的计算量，达到快速实现的目的.经过大量实验表明计算速度提高数百倍，仍能保持原来的计算效果.     

Linking ground motion measurements and macroseismic observations in France: a case study based on accelerometric and macroseismic databases

Comparison between accelerometric and macroseismic observations is made for three M _w?=?4.5 earthquakes, which occurred in north-eastern France and south-western Germany in 2003 and 2004. Scalar and spectral instrumental parameters are processed from the accelerometric data recorded by nine accelerometric stations located between 29 and 180 km from the epicentres. Macroseismic data are based on French Internet reports. In addition to the single questionnaire intensity, analysis of the internal correlation between the encoded answers highlights four predominant fields of questions bearing different physical meanings: (1) “vibratory motions of small objects”, (2) “displacement and fall of objects”, (3) “acoustic noise” and (4) “personal feelings”. Best correlations between macroseismic and instrumental observations are obtained when the macroseismic parameters are averaged over 10-km-radius circles around each station. Macroseismic intensities predicted by published peak ground velocity (PGV)–intensity relationships agree with our observed intensities, contrary to those based on peak ground acceleration (PGA). Correlation between the macroseismic and instrumental data for intensities between II and V (EMS-98) is better for PGV than for PGA. Correlation with the response spectra exhibits clear frequency dependence for all macroseismic parameters. Horizontal and vertical components are significantly correlated with the macroseismic parameters between 1 and 10 Hz, a range corresponding to both natural frequencies of most buildings and high energy content in the seismic ground motion. Between 10 and 25 Hz, a clear lack of correlation between macroseismic and instrumental observations exists. It could be due to a combination of the decrease in the energy signal above 10 Hz, a high level of anthropogenic noise and an increase in variability in soil conditions. Above 25 Hz, the correlation coefficients between the acceleration response spectra and the macroseismic parameters are close to the PGA correlation level.     

Widely linear denoising of multicomponent seismic data

Seismic data processing is a challenging task, especially when dealing with vector-valued datasets. These data are characterized by correlated components, where different levels of uncorrelated random noise corrupt each one of the components. Mitigating such noise while preserving the signal of interest is a primary goal in the seismic-processing workflow. The frequency-space deconvolution is a well-known linear prediction technique, which is commonly used for random noise suppression. This paper represents vector-field seismic data through quaternion arrays and shows how to mitigate random noise by proposing the extension of the frequency-space deconvolution to its hypercomplex version, the quaternion frequency-space deconvolution. It also shows how a widely linear prediction model exploits the correlation between data components of improper signals. The widely linear scheme, named widely-linear quaternion frequency-space deconvolution, produces longer prediction filters, which have enhanced signal preservation capabilities shown through synthetic and field vector-valued data examples.     

基于高通滤波的频率-空间域经验模态分解压制高频噪声

高频噪声压制是高分辨率地震数据处理中提高信噪比的关键性问题.本文针对f-x(频率-空间)反褶积空间预测滤波器无法处理非平稳、非线性信号的缺点,提出了一种基于高通滤波的频率-空间域经验模态分解(Empirical Mode Decomposition in the frequency-space domain,f-xEMD)压制地震剖面中高频噪声的方法.该方法采用全域高通滤波从原始数据中分离出含有部分有效信号的高频数据,将其变换到f-x域,然后在滑动的短窗口内提取每一个频率的空变数据序列进行EMD分解得到高频复本征模态函数(Intrinsic Mode Function,IMF)IMF1,将所有频率的IMF1序列反Fourier变换到时间域得到噪声剖面,将其与原始数据相减,达到高频噪声压制的目的.该方法可克服传统EMD分解方法中的模态混叠现象,保护陡倾角反射同相轴;压制后的噪声剖面中不包含有效信号能量,地震剖面的信噪比得到了提高.模拟数据和实际数据处理结果充分证明了该方法的有效性.     

Vibroseis deconvolution: A comparison of pre and post correlation vibroseis deconvolution data in real noisy data

Vibroseis is a source used commonly for inland seismic exploration. This non-destructive source is often used in urban areas with strong environmental noise. The main goal of seismic data processing is to increase the signal/noise ratio where a determinant step is deconvolution. Vibroseis seismic data do not meet the basic minimum-phase assumption for the application of spiking and predictive deconvolution, therefore various techniques, such as phase shift, are applied to the data, to be able to successfully perform deconvolution of vibroseis data.This work analyzes the application of deconvolution techniques before and after cross-correlation on a real data set acquired for high resolution prospection of deep aquifers. In particular, we compare pre-correlation spiking and predictive deconvolution with Wiener filtering and with post-correlation time variant spectral whitening deconvolution. The main result is that at small offsets, post cross-correlation spectral whitening deconvolution and pre-correlation spiking deconvolution yield comparable results, while for large offsets the best result is obtained by applying a pre-cross-correlation predictive deconvolution.     

Further improvement of temporal resolution of seismic data by autoregressive (AR) spectral extrapolation

Seismic data have still no enough temporal resolution because of band-limited nature of available data even if it is deconvolved. However, lower and higher frequency information belonging to seismic data is missing and it is not directly recovered from seismic data. In this paper, a method originally applied by Honarvar et al. [Honarvar, F., Sheikhzadeh, H., Moles, M., Sinclair, A.N., 2004. Improving the time-resolution and signal–noise ratio of ultrasonic NDE signals. Ultrasonics 41, 755–763.] which is the combination of the most widely used Wiener deconvolution and AR spectral extrapolation in frequency domain is briefly reviewed and is applied to seismic data to improve temporal resolution further. The missing frequency information is optimally recovered by forward and backward extrapolation based on the selection of a high signal–noise ratio (SNR) of signal spectrum deconvolved in signal processing technique. The combination of the two methods is firstly tested on a variety of synthetic examples and then applied to a stacked real trace. The selection of necessary parameters in Wiener filtering and in extrapolation are discussed in detail. It is used an optimum frequency windows between 3 and 10 dB drops by comparing results from these drops, while frequency windows are used as standard between 2.8 and 3.2 dB drops in study of Honarvar et al. [Honarvar, F., Sheikhzadeh, H., Moles, M., Sinclair, A.N., 2004. Improving the time-resolution and signal–noise ratio of ultrasonic NDE signals. Ultrasonics 41, 755–763.]. The results obtained show that the application of the purposed signal processing technique considerably improves temporal resolution of seismic data when compared with the original seismic data. Furthermore, AR based spectral extrapolated data can be almost considered as reflectivity sequence of layered medium. Consequently, the combination of Wiener deconvolution and AR spectral extrapolation can reveal some details of seismic data that cannot be observed in raw signal or which lost during the previous processing.     

矿井瞬变电磁探测中空芯线圈传感器的研制

针对矿井瞬变电磁法中探测空间的有限性及探测目标分布的全空间性,研制一种适合该方法的感应式空芯线圈传感器.根据矿井瞬变电磁探测需要,确定了感应式空芯线圈传感器的基本技术指标.通过对空芯线圈的物理结构和前置放大电路的等效噪声电路进行分析,获取了引入噪声的主要来源.构建感应式空芯线圈传感器的等效模型,实现传感器的幅频特性及低噪声的分析,给出影响传感器噪声水平的各噪声源随频率变化的规律,设计矿井瞬变电磁中降低传感器噪声的方法,采用该优化方法,传感器的总噪声在谐振频率处信噪比提高了15.6dB.测试结果表明:感应式空芯线圈传感器噪声性能和输出特性优于加拿大Geonics公司的3D-1HF-coil传感器,满足矿井瞬变电磁探测需要.     

Bayesian frequency-domain blind deconvolution of ground-penetrating radar data

Enhancing the resolution and accuracy of surface ground-penetrating radar (GPR) reflection data by inverse filtering to recover a zero-phased band-limited reflectivity image requires a deconvolution technique that takes the mixed-phase character of the embedded wavelet into account. In contrast, standard stochastic deconvolution techniques assume that the wavelet is minimum phase and, hence, often meet with limited success when applied to GPR data. We present a new general-purpose blind deconvolution algorithm for mixed-phase wavelet estimation and deconvolution that (1) uses the parametrization of a mixed-phase wavelet as the convolution of the wavelet's minimum-phase equivalent with a dispersive all-pass filter, (2) includes prior information about the wavelet to be estimated in a Bayesian framework, and (3) relies on the assumption of a sparse reflectivity. Solving the normal equations using the data autocorrelation function provides an inverse filter that optimally removes the minimum-phase equivalent of the wavelet from the data, which leaves traces with a balanced amplitude spectrum but distorted phase. To compensate for the remaining phase errors, we invert in the frequency domain for an all-pass filter thereby taking advantage of the fact that the action of the all-pass filter is exclusively contained in its phase spectrum. A key element of our algorithm and a novelty in blind deconvolution is the inclusion of prior information that allows resolving ambiguities in polarity and timing that cannot be resolved using the sparseness measure alone. We employ a global inversion approach for non-linear optimization to find the all-pass filter phase values for each signal frequency. We tested the robustness and reliability of our algorithm on synthetic data with different wavelets, 1-D reflectivity models of different complexity, varying levels of added noise, and different types of prior information. When applied to realistic synthetic 2-D data and 2-D field data, we obtain images with increased temporal resolution compared to the results of standard processing.     

Elimination of ghost noise in vibroseis data by deconvolution

A deconvolution approach is presented to process uncorrelated vibroseis data. The recorded ground force signal, known to be a better correlator for the vibrogram than the pilot sweep itself, is used to deconvolve rather than to correlate the vibrogram. In such a way the ghost sweep (correlation noise), produced by severe harmonic distortions in the ground force signal both at negative and positive correlation times if such a signal is used as the correlator, is eliminated automatically. This type of deconvolution can eliminate the ghost sweep caused by both the upsweep or by the downsweep signal. Synthetic and real data are used to demonstrate the application of the deconvolution procedure. The results are compared with the traditional correlation procedure and show the superiority of the deconvolution approach.     

Lp-NORM DECONVOLUTION1

The purpose of deconvolution is to retrieve the reflectivity from seismic data. To do this requires an estimate of the seismic wavelet, which in some techniques is estimated simultaneously with the reflectivity, and in others is assumed known. The most popular deconvolution technique is inverse filtering. It has the property that the deconvolved reflectivity is band-limited. Band-limitation implies that reflectors are not sharply resolved, which can lead to serious interpretation problems in detailed delineation. To overcome the adverse effects of band-limitation, various alternatives for inverse filtering have been proposed. One class of alternatives is L_p-norm deconvolution, L₁norm deconvolution being the best-known of this class. We show that for an exact convolutional forward model and statistically independent reflectivity and additive noise, the maximum likelihood estimate of the reflectivity can be obtained by L_p-norm deconvolution for a range of multivariate probability density functions of the reflectivity and the noise. The L_∞-norm corresponds to a uniform distribution, the L₂-norm to a Gaussian distribution, the L₁-norm to an exponential distribution and the L₀-norm to a variable that is sparsely distributed. For instance, if we assume sparse and spiky reflectivity and Gaussian noise with zero mean, the L_p-norm deconvolution problem is solved best by minimizing the L₀-norm of the reflectivity and the L₂-norm of the noise. However, the L₀-norm is difficult to implement in an algorithm. From a practical point of view, the frequency-domain mixed-norm method that minimizes the L₁norm of the reflectivity and the L₂-norm of the noise is the best alternative. L_p-norm deconvolution can be stated in both time and frequency-domain. We show that both approaches are only equivalent for the case when the noise is minimized with the L₂-norm. Finally, some L_p-norm deconvolution methods are compared on synthetic and field data. For the practical examples, the wide range of possible L_p-norm deconvolution methods is narrowed down to three methods with p= 1 and/or 2. Given the assumptions of sparsely distributed reflectivity and Gaussian noise, we conclude that the mixed L₁norm (reflectivity) L₂-norm (noise) performs best. However, the problems inherent to single-trace deconvolution techniques, for example the problem of generating spurious events, remain. For practical application, a greater problem is that only the main, well-separated events are properly resolved.     

普洱大寨深井噪声压制效果及井孔附近波场特征研究

基于2011年建立的云南普洱大寨深井台站,开展了噪声压制及附近波场特征研究.通过计算该台站的噪声功率谱概率密度函数,显示该井下台站对1 Hz以上的高频噪声具有明显的压制效果, 最高能降低40 dB,其降噪能力优于其它井下台阵,推断与该台站附近的场地条件有关.基于地表与井下地震记录的差异,应用正则化反卷积干涉方法进一步研究该台站附近的波场特征.以地表记录为参考,对井下记录进行反卷积,获取两台站之间的格林函数,直接识别出了原始记录上无法区分的上行入射波与下行地表反射波,然后利用两震相的到时差建立了一个浅层地震波速度模型,与理论模拟的结果一致.研究结果表明,相对于地表观测,井下台站在压制噪声和近地表地震波传播特征研究等方面具有很大的优势,同时该研究对其它地区开展深井观测具有参考意义.      

SELF-MATCHING DECONVOLUTION IN THE FREQUENCY DOMAIN *

In a previous paper the author showed how, by computing an inverse filter in the frequency domain, an automatic compromise could be made between the conflicting requirements to spike a wavelet and to keep the attendant noise amplification within bounds. This paper extends the technique to take account of errors in the estimated shape of the wavelet defined to the deconvolution process. The drastic effects which such errors can have if they are ignored are demonstrated. A novel form of filter–called the “self-matching filter”–is defined which allows the user to limit not only the noise amplification but also the sensitivity of the filter to random uncertainties in the estimated wavelet. This is achieved by whitening the spectrum only within automatically selected pass bands whilst suppressing other noise-dominated or uncertainly defined frequency components. Conventional Wiener filtering is shown to be a special case of this more general filter, namely one in which the wavelet uncertainty is completely ignored. The type of phase spectrum which the output pulse should be designed to possess (e.g. zero phase or minimum phase) is briefly discussed.     

STABILIZATION OF NORMAL-INCIDENCE SEISMOGRAM INVERSION REMOVING THE NOISE-INDUCED BIAS*

A main problem in computing reflection coefficients from seismograms is the instability of the inversion procedure due to noise. This problem is attacked for two well-known inversion schemes for normal-incidence reflection seismograms. The crustal model consists of a stack of elastic, laterally homogeneous layers between two elastic half-spaces. The first method, which directly computes the reflection coefficients from the seismogram is called “Dynamic Deconvolution”. The second method, here called “Inversion Filtering”, is a two-stage procedure. The first stage is the construction of a causal filter by factorization of the spectral function via Levinson-recursion. Filtering the seismogram is the second stage. The filtered seismogram is a good approximation for the reflection coefficients sequence (unless the coefficients are too large). In the non-linear terms of dynamic deconvolution and Levinson-recursion the noise could play havoc with the computation. In order to stabilize the algorithms, the bias of these terms is estimated and removed. Additionally incorporated is a statistical test for the reflection coefficients in dynamic deconvolution and the partial correlation coefficients in Levinson-recursion, which are set to zero if they are not significantly different from noise. The result of stabilization is demonstrated on synthetic seismograms. For unit spike source pulse and white noise, dynamic deconvolution outperforms inversion filtering due to its exact nature and lesser computational burden. On the other hand, especially in the more realistic bandlimited case, inversion filtering has the great advantage that the second stage acts linearly on the seismogram, which allows the calculation of the effect of the inversion procedure on the wavelet shape and the noise spectrum.     

Results from 44 months of observations with a superconducting gravimeter at Moxa/Germany

Results for more than 42 months of observations with the superconducting gravimeter CD-034 at the Geodynamic Observatory Moxa are discussed. Moxa observatory is one of the newer stations within the ‘Global Geodynamics Project’ (GGP). A special feature of the gravimeter at Moxa is its dual sensor system; differences in the results obtained from the two sensor recordings are generally well within the standard deviations of the tidal analysis. One significant difference concerns the slightly different drift rates of 31 and 49.5 nm/s² per year for upper and lower sensor; both sensor drifts can be fitted by a linear function. We find that the noise levels are close to the ‘New Low Noise Model’ for the seismic-modes and are also low in the tidal bands. Due to this low noise, Moxa is a station well suited to search for small geodynamic signals. The long-period variation in the gravity residuals correlates well with the polar motion.The difference signal between the two sensor recordings has a peak-to-peak amplitude of about 6 nm/s² and shows systematic variations. Its spectrum is characterised by instrumental noise between 0.2 and 0.4 cph. The noise level of the difference and of the sum of the two residual datasets are clearly lower, respectively, higher than the noise contents of the gravity residuals themselves. This is a strong indication for the existence of broadband signals common to the two residual datasets, leading to the conjecture that the reduction of environmental effects is still not sufficient.Our results once more emphasize the necessity to correct the data for barometric pressure effects when analyzing the data for seismic modes. The reduction visibly increases the signal-to-noise ratio in the low frequencies of the mode band and helps to avoid misinterpreations of peaks. Besides the well known barometric pressure influence we can establish hydrological effects in the data which are probably caused by soil moisture and groundwater table variations as well as by batch-wise water movement within the weathering layer. As the major part of the observatory surroundings is above gravimeter level, an anticorrelation between hydrological and gravity changes is observed. In addition, it can be shown that global hydrological effects reach an order of magnitude that makes it necessary to consider these effects when analyzing long-period signals like polar motion. Vice versa these effects are large enough to be detectable in the gravity data. A first joint analysis of five datasets from the GGP network shows no indications for signals related to the Slichter triplet or core modes.     

基于功能试验的加速度传感器性能评价

以二阶系统的动力方程为基础,计算不同固有频率和阻尼比下的传感器功能试验响应,讨论功能试验响应随频率和阻尼比的变化,给出了依据功能试验响应确定传感器固有频率和过冲响应比的计算方法,继而统计回归得到过冲响应比随阻尼比的变化关系式.基于传感器的频率响应分析,讨论阻尼比对传感器的幅频特性和相频特性的影响,认为阻尼比在0.68-0.70范围内传感器具有最佳的幅频与相频特性.最后,以传感器自振特性确定方法及性能评价分析为基础,提出基于功能试验的力平衡式加速度传感器性能评价方法,并通过实例对典型传感器的性能进行评价.     

Analysis of observations with dual sensor superconducting gravimeters

Among the 21 superconducting gravimeters presently operating worldwide four instruments exist that are equipped with two vertically aligned sensor units. Three of the instruments are installed in Germany (Bad Homburg, Moxa, Wettzell) and one in South Africa (Sutherland). Comparisons of the data sets obtained with the dual sensor systems yield information on instrumental effects and sensitivity as well as on the efficiency of reductions of environmental effects applied to the data. The latter is an important constraint when looking for small geodynamic signals like Slichter and core modes or aperiodic variations.From analyses of the two data sets of each instrument a small but significant difference of 1-3% in the response of the sensor units on barometric pressure variations is found. Likewise, the records of lower and upper sensor vary slightly but not systematically with regard to the noise levels in the different frequency ranges. The tidal analyses yield an agreement of the tidal parameters generally well within the standard deviations determined from the least squares adjustment in the tidal analysis. The deviations are in the range between 0×10⁻⁴ and 3×10⁻⁴ for the amplitude factor and the phases differ between 0.0005° and 0.01° for the four main tidal constituents O1, K1, M2, and S2.The comparison of the gravity residuals of the two sensors with each other as well as with their sum and difference in the time and frequency domain shows the existence of identical signals in the records of the two sensors in the whole range of observation. This probably means that either the environmental reductions applied are not sufficient or there are additional disturbing effects in the data which have not been taken care of yet. From the study it emerges that it is not possible to get entirely rid of the tidal signals in the data. This is probably also due to the fact that despite reductions the data sets contain additional signals and slightly different noise at tidal frequencies which affect the result of the tidal analysis.