CN102830432B - Method for identifying weak reflection reservoir under cover of coal series strong earthquake reflection characteristics - Google Patents

Abstract

The invention provides a method for identifying a weak reflection reservoir under the cover of coal series strong earthquake reflection characteristics, belonging to the field of seismic exploration reservoir prediction. The method comprises the steps of firstly analyzing an after-stack amplitude maintaining three-dimensional seismic data body, determining a thin layer spectrum resolving calculation range and a spectrum resolving calculation range of the whole data body to obtain a thin layer tuning three-dimensional data body through thin layer spectrum resolving calculation; secondly obtaining tuning three-dimensional data bodies of all sampling points in the after-stack amplitude maintaining three-dimensional seismic data body through a moving time window; subsequently generating a co-frequency component three-dimensional data body or a single frequency component three-dimensional data body through separation; obtaining a time-maximum amplitude frequency data body through comparison; and finally identifying the weak reflection reservoir under the cover of coal series strong earthquake reflection characteristics by using the frequency difference between the sand and the coal series reservoir. With the adoption of the method, a spatial distribution trend of the weak reflection reservoir under the cover of coal series strong earthquake reflection characteristics is reflected in profile and space-time mode visually.

Description

A kind of coal measures strong earthquakes reflectance signature covers the recognition methods of lower weak reflection reservoir

Technical field

The invention belongs to field of seismic exploration reservoir prediction, be specifically related to the recognition methods that a kind of coal measures strong earthquakes reflectance signature covers lower weak reflection reservoir.

Background technology

Spectral Decomposition Technique is the characteristic Reservoir Description decomposed based on frequency spectrum that development in recent years is got up.Geological data is transformed to frequency field from time domain by Spectral Decomposition Technique, can obtain abundanter seismic wave field dynamics and kinematics information.Current petroclastic rock sandstone reservoir predicts that common FREQUENCY SPECTRUM DECOMPOSITION TECHNIQUE has Fourier transform, wavelet transformation, maximum entropy, Amoco company Spectral Decomposition Technique etc.

The feature of these technology is as follows: the basis function of Fourier transform has of overall importance, without time resolution characteristics in time domain, frequency domain can localize, but can not portray the Seismic reflection character of time domain Local Layer; Maximum entropy spectrum is decomposed can obtain better frequency resolution, but by window limit access time, temporal resolution is not high; Wavelet transformation uses scale parameter to control, the width of time frequency window is with signal adaptive transformation, high frequency constantly window narrows automatically, low frequency constantly window broadens automatically, but can not be directly corresponding with frequency parameter, geological meaning is clear and definite not, and the wavelet transformation after improvement can assigned frequency band number and frequency distribution density, can extract the effective frequency information with geological Significance relevant with the frequency response characteristic of objective body; Amoco company Spectral Decomposition Technique is the FREQUENCY SPECTRUM DECOMPOSITION TECHNIQUE for zone of interest, consider thin layer tuning characteristic, spectrum signature in thin layer can be reflected, current application mainly for the tuning 3-D data volume that thin zone of interest calculates, and shows the frequency characteristic of thin zone of interest by the form of frequency spectrum section.

Above-mentioned several FREQUENCY SPECTRUM DECOMPOSITION TECHNIQUE all considers frequecy characteristic, but respectively has relative merits.The result of their spectral decomposition is the spectrum signature of the research purpose layer with sliced form display mostly.

The Daniudi Gasfield in Ordos Basin of China belongs to low hole, hypotonic, the lithologic deposit of dense form based on river channel sand.Gas field is mainly based on the Taiyuan Forma-tion sandstone reservoir of the lower Shihezi Formation of the Permian system, Shanxi group and the Carboniferous system, and the distribution of trap controls by sand body development degree.Shanxi group and Taiyuan Forma-tion coal seam are grown, and coal seam is large with country rock resistance difference, seismic reflection energy is strong and continuous.Lower Shihezi Formation box counting main stem sand body is DaNiuDi gas field main productive layers, sand body thinner and in length and breadth stacked, phase transformation is frequent, nonuniformity is strong, sand body and country rock wave impedance difference is little, reflected energy is more weak.The strong coal measure strata reflection of DaNiuDi gas field often masks the reflectance signature of the box counting reservoir-sand body in the portion of being located thereon.

In Clastic Stratum of Country Rocks reservoir prediction, extract the information relevant with amplitude and remain the conventional method of current sand body identification.But due to the interference of coal measure strata strong earthquakes reflectance signature and the complicated and changeable of the thin reservoir of the weak reflection of channel sand, make directly to be restricted the precision that the seismic data of limited resolution carries out Sand-body Prediction by conventional method, cannot meticulous depiction thin gas-bearing sandstone reservoir internal reflection feature.Frequency-amplitude response can obtain than time m-amplitude-frequency response abundanter, meticulousr amplitude information.

Although the frequency slice display mode generally adopted at present can obtain zone of interest interference image in the plane, explanation personnel are also needed to identify structure and the pattern of representative geologic sedimentation by rule of thumb.Only comprehensive not with the Spatial Distribution Pattern of the spectrum signature reflection sand body of sliced form.

Thin layer Spectral Decomposition Technique is to study stratum internal reflection changing features for speciality, the single-frequency that the tuning 3-D data volume calculated by single thin layer (representing zone of interest) generates is cut into slices and can be represented the horizontal reflectance signature change of Local Layer, but can not reflect larger depth range, more fully information.

Summary of the invention

The object of the invention is to solve the difficult problem existed in above-mentioned prior art, a kind of coal measures strong earthquakes reflectance signature is provided to cover the recognition methods of lower weak reflection reservoir, utilize thin layer Spectral Decomposition Technique, take into account thin layer tuning characteristic and different lithology, physical property rock shows different features in " tuning " frequency place advantage, on the basis obtaining frequency component data body altogether, consider the frequecy characteristic difference of coal measure strata and sandstone reservoir, the reflectance signature of outstanding thin sand body on section and Distribution Pattern, and then the spatial trend of prediction sand body, improve sandstone reservoir precision of prediction.

The present invention is achieved by the following technical solutions:

A kind of coal measures strong earthquakes reflectance signature covers the recognition methods of lower weak reflection reservoir, first described method is analyzed poststack and is protected width 3-d seismic data set, determine the spectral factorization computer capacity of thin layer spectral factorization computer capacity and whole data volume, utilize thin layer spectral factorization to calculate thin layer tuning 3-D data volume; Then by window time mobile, the tuning 3-D data volume that poststack protects all sampled points in width 3-d seismic data set is obtained; Frequency component 3-D data volume or single-frequency component 3-D data volume is altogether generated again through sorting; Through comparing m-peak swing frequency data body when obtaining; Finally utilize the frequency difference of sand body and coal measure strata identify coal measures strong earthquakes reflectance signature cover under weak reflection reservoir.

Said method comprising the steps of:

(1) analyze poststack and protect width 3-d seismic data set

Width 3-d seismic data set is protected to poststack and carries out spectrum analysis, understand sand body and coal seam spectrum signature, determine the computer capacity of thin layer spectral factorization computer capacity and whole data volume;

(2) single thin layer spectral factorization calculates

The thin layer spectral factorization computer capacity determined by step (1) is protected width 3-d seismic data set from poststack and is extracted thin layer data, carries out spectral factorization calculating, obtains thin layer tuning 3-D data volume;

(3) whole data volume spectral factorization calculates;

Window when being moved by pointwise, repeats step (2), through calculating the corresponding thin layer tuning 3-D data volume of all sampled points to each sampled point that poststack is protected in width 3-d seismic data set;

(4) one group of frequency component data body altogether of hanking is divided

Sorting is carried out to the thin layer tuning 3-D data volume of all sampled points that step (3) obtains, formed one group altogether frequency component time m-amplitude data body, also sorting can generate the time m-amplitude slice and section of single-frequency component simultaneously;

(5) peak swing frequency data body is extracted

The result of calculation that step (4) obtains is compared, extraction time-peak swing frequency data body; Rise time-peak swing frequency data body section and section;

(6) with section or section step display (5) result, highlight strong reflection amplitude feature cover under sandstone reservoir spatial feature, carry out the prediction of sand body spatial; Wherein, thin sand dominant frequency is at about 40Hz, and coal measure strata dominant frequency is at about 20Hz.

Compared with prior art, the invention has the beneficial effects as follows:

(1) the present invention obtains multiple frequency component data body altogether, can show unifrequency amplitude section;

(2) the time m-peak swing frequency data body that the present invention extracts from section and spatially can intuitively be reflected in the spatial trend that coal measures strong reflection amplitude feature covers lower weak reflection sandstone reservoir;

(3) the present invention is that the thin RESERVOIR RECOGNITION of Clastic Stratum of Country Rocks provides a kind of directly perceived, practical method with prediction.

Accompanying drawing explanation

Fig. 1 is the principle schematic of the thin layer spectrum imaging that the inventive method uses.

Fig. 2 is the implementation step block diagram of the inventive method.

Fig. 3-1 is the original section in the embodiment of the present invention.

Fig. 3-2 is the 20Hz amplitude section using the inventive method to obtain in the embodiment of the present invention, and wherein dark strong amplitude areas is coal measure strata reflectance signature.

Fig. 3-3 be use the inventive method to obtain in the embodiment of the present invention 40Hz amplitude section, wherein dark strong amplitude areas is coal measure strata reflectance signature.

Fig. 3-4 is the peak swing frequency sections using the inventive method to obtain in the embodiment of the present invention, and wherein dark-background is low frequency, and light color is high frequency, and white is 40Hz frequency, leucoplast reflection sandbody distribution trend.

Embodiment

Below in conjunction with accompanying drawing, the present invention is described in further detail:

(1) ultimate principle

Thin layer Spectral Decomposition Technique is the technology of the seismic data volume time-frequency convert based on discrete fourier transform.

Discrete fourier transform formula is:

$X_m=\underset{n=0}{\overset{N-1}{\mathrm{\Σ}}}{x}_n{e}^{-\mathrm{inm}\frac{2\mathrm{\π}}{N}};m,n=\mathrm{0,1},...,N-1$

Wherein x _nfor finite discrete seismic signal, X _mfor its frequency spectrum, N is sampling number.

X _mcan be write as the expression formula containing real part and imaginary part:

X _m＝U _m+iV _m

Then spectral amplitude A _mbe expressed as:

$A_m={({U^2}_m+{V^2}_m)}^{\frac{1}{2}}$

Frequency representation is:

$f_m=\frac{m}{\mathrm{N\Δ}}$

Wherein, Δ is sampling rate, and m is frequency number, and N is sampling number.

The theoretical foundation of thin layer Spectral Decomposition Technique is that thin bed reflections system can produce complicated tuned reflective (as shown in Figure 1).Thin strate to be reflected in frequency field unique features express can instruction time variation in thickness.Thin layer spectral factorization calculates thin layer tuning 3-D data volume.Reflect the relation between acoustic wave character that the spectral amplitude obtained can determine to form the single stratum of reflection by thin layer tuning, spectral amplitude falls into curve determination thin strate situation of change frequently by spectrum.It is relevant with the change of local rock mass (as local geology, fluid, sedimentology etc.) that spectrum falls into curve frequently.Spectral amplitude falls into frequently characteristic reflection thin layer time variation in thickness, and complex formation spectral change of caving in can disclose thin strate inner lithology cross direction profiles feature.

After determining based on the tuning 3-D data volume of thin layer, the tuning 3-D data volume of whole calculating data volume can be obtained by window time mobile, then generate frequency component data body and single-frequency component data body altogether through gather.

Because coal seam is different with the dominant frequency of non-coal seam sand body, and all belong to thin layer, can directly distinguish by the different frequency amplitude-frequency response feature that thin layer Spectral Decomposition Technique obtains.When coal measure strata coexists, even if obtain single-frequency section, but single-frequency section highlights remain coal measure strata feature, weak reflection sandstone features can not get fine embodiment (as shown in accompanying drawing 3-2 and accompanying drawing 3-3).Weak reflection sandbody features under utilizing the frequency difference of sand body and coal measure strata that the strong amplitude of coal measure strata can be made to cover is represented (as shown in accompanying drawing 3-4).

(2) step of the inventive method

The realization flow of the inventive method is as shown in Figure 2, specific as follows:

(1) analyze poststack and protect width 3-d seismic data set

Width 3-d seismic data set is protected to poststack and carries out spectrum analysis, understand sand body and coal seam spectrum signature, determine the spectral factorization computer capacity of thin layer spectral factorization computer capacity and whole data volume;

(2) single thin layer spectral factorization calculates

The thin layer spectral factorization computer capacity determined by step (1) is protected width 3-d seismic data set from poststack and is extracted thin layer data, carries out spectral factorization calculating, obtains thin layer tuning 3-D data volume;

(3) whole data volume spectral factorization calculates;

Window when being moved by pointwise, repeats step (2), through calculating the corresponding thin layer tuning 3-D data volume of all sampled points to each sampled point that poststack is protected in width 3-d seismic data set;

(4) one group of frequency component data body altogether of hanking is divided

Sorting is carried out to the thin layer tuning 3-D data volume of all sampled points that step (3) obtains, formed one group altogether frequency component time m-amplitude data body, also sorting can generate the time m-amplitude slice and section of single-frequency component simultaneously;

(5) peak swing frequency data body is extracted

The result of calculation that step (4) obtains is compared, extraction time-peak swing frequency data body; Rise time-peak swing frequency data body section and section;

(6) with section or section step display (5) result, highlight strong reflection amplitude feature cover under sandstone reservoir spatial feature, carry out the prediction of sand body spatial; Wherein, thin sand dominant frequency is at about 40Hz, and coal measure strata dominant frequency is at about 20Hz.

Accompanying drawing 3-1 is that a poststack protects width seismic section, on section, strong reflection amplitude feature is the reflection (the coal measure strata B see in accompanying drawing 3-1) of Shanxi group and layer position, Taiyuan Forma-tion coal measure strata place, on this coal measure strata B, square box 2+3 section main stem sand body is this district's main productive layers (the reservoir A see in accompanying drawing 3-1), and on this sand body accompanying drawing 3-1, reservoir A feature is not obvious.

Accompanying drawing 3-2 and accompanying drawing 3-3 is respectively in 20Hz and 40Hz amplitude section, and strong earthquakes reflectance signature still concentrates on coal measure strata (the coal measure strata B see in accompanying drawing 3-2 and accompanying drawing 3-3), and the feature of reservoir A is not obvious.

On accompanying drawing 3-4, the feature of reservoir A is obvious, and the lateral extension of reservoir A and relative thickness are also high-visible, and actual well drilled confirms that the D place sand body of reservoir A is thicker than E place sand body, and D place obtains high yield gas, and E place obtains comparatively high yield gas.C layer is box 1 section of ultra-thin sand body (not obvious on accompanying drawing 3-1 ~ accompanying drawing 3-3), and C layer is better than D place at E place sand body enrichment degree.

Technique scheme is one embodiment of the present invention, for those skilled in the art, on the basis that the invention discloses application process and principle, be easy to make various types of improvement or distortion, and the method be not limited only to described by the above-mentioned embodiment of the present invention, therefore previously described mode is just preferred, and does not have restrictive meaning.

Claims (1)

1. a coal measures strong earthquakes reflectance signature covers the recognition methods of lower weak reflection reservoir, it is characterized in that: first described method is analyzed poststack and protected width 3-d seismic data set, determine the spectral factorization computer capacity of thin layer spectral factorization computer capacity and whole data volume, utilize thin layer spectral factorization to calculate thin layer tuning 3-D data volume; Then by window time mobile, the tuning 3-D data volume that poststack protects all sampled points in width 3-d seismic data set is obtained; Frequency component 3-D data volume or single-frequency component 3-D data volume is altogether generated again through sorting; Through comparing m-peak swing frequency data body when obtaining; Finally utilize the frequency difference of sand body and coal measure strata identify coal measures strong earthquakes reflectance signature cover under weak reflection reservoir;

Said method comprising the steps of:

(1) analyze poststack and protect width 3-d seismic data set

Width 3-d seismic data set is protected to poststack and carries out spectrum analysis, understand sand body and coal seam spectrum signature, determine the spectral factorization computer capacity of thin layer spectral factorization computer capacity and whole data volume;

(2) single thin layer spectral factorization calculates

The thin layer spectral factorization computer capacity determined by step (1) is protected width 3-d seismic data set from poststack and is extracted thin layer data, carries out spectral factorization calculating, obtains thin layer tuning 3-D data volume;

(3) whole data volume spectral factorization calculates;

Window when being moved by pointwise, repeats step (2), through calculating the corresponding thin layer tuning 3-D data volume of all sampled points to each sampled point that poststack is protected in width 3-d seismic data set;

(4) one group of frequency component data body altogether of hanking is divided

Sorting is carried out to the thin layer tuning 3-D data volume of all sampled points that step (3) obtains, formed one group altogether frequency component time m-amplitude data body, also sorting can generate the time m-amplitude slice and section of single-frequency component simultaneously;

(5) peak swing frequency data body is extracted

The result of calculation that step (4) obtains is compared, extraction time-peak swing frequency data body; Rise time-peak swing frequency data body section and section;

(6) with section or section step display (5) result, highlight strong reflection amplitude feature cover under sandstone reservoir spatial feature, carry out the prediction of sand body spatial; Wherein, thin sand dominant frequency is at about 40Hz, and coal measure strata dominant frequency is at about 20Hz.