CN103728659A - Method for improving underground karst detecting precision - Google Patents

Abstract

The invention provides a method for improving the underground karst detecting precision, and belongs to the field of seismic data explaining in the oil and gas field exploitation. The method comprises the first step of frequency division processing, namely, stacked pure-wave amplitude-preserved seismic data are input, and then the frequency division processing is carried out on the stacked pure-wave amplitude-preserved seismic data to obtain a series of frequency division data bodies with the frequency from high to low; the second step of coherent data processing, namely, different frequency division data bodies are respectively calculated through the coherent algorithm to obtain a series of frequency division coherent data bodies; the third step of weighted data processing, namely, the weighted data processing is carried out on the series of frequency division coherent data bodies through an optimal weighting factor to obtain reconstructed data bodies. By means of the method, the form of the karst geologic body can be more accurately recognized, and the development area beneficial to the karst geologic body can be found out.

Description

A kind of method that improves detection of karst cave precision

Technical field

The invention belongs to the seismic data interpretation field in exploration of oil and gas field, be specifically related to a kind of method that improves detection of karst cave precision.

Background technology

The reservoir space of Karst-type carbonate reservoir is main mainly with corrosion hole, hole, has the features such as abundance is high, production capacity is large, is the favo(u)rable target of exploration high yield hydrocarbon-bearing pool.But the general buried depth of China's Karst-type carbonate reservoir is larger, stratum, top suffers again repeatedly weathering, degrades, and surface relief changes greatly, and underground karst Distribution Pattern complexity has the feature of very strong nonuniformity.Above-mentioned unfavorable factor causes carbonate formation seismic imaging poor generally, and data signal to noise ratio (S/N ratio) and resolution are low, and Karst Features prediction difficulty is larger.

At present, in oil-gas exploration, often adopt method of seismic prospecting.But due to the geological data of field acquisition after treatment, make poststack pure wave data volume there is certain frequency band range.While utilizing frequency division geological data geologize lithosomic body in the < < Frequency dependent seismic stratigraphy > > that therefore Zeng etc. (2009) delivers, find, some single-frequency data volume is clearer than the band-limited seismic data volume of routine to portraying of geological objects boundary, scope, the geology details of reflection is also abundanter, thereby explains good thinking is provided for the seismogeology of frequency field.

In addition, coherent body technique is the practical technique of analyzing for seismic data interpretation of releasing in recent years, comparatively effective to identifying Lateral heterogeneity, fracture characteristic and prediction crack and the development belt thereof of trickle rock stratum.Seismic coherence refers to the seismic properties between adjacent seismic trace, as: the measurement of the similarity degrees such as waveform, amplitude, frequency, phase place.Coherent algorithm mainly contains 3 kinds at present, is respectively first generation coherent algorithm C ₁, second generation coherent algorithm C ₂, third generation coherent algorithm C ₃, its gap at aspects such as effect and calculated amount is obvious, wherein second generation coherent algorithm C ₂the most conventional.

In sum, for this kind of existing geophysical prospecting technology of special geologic body, substantially can adopt coherent approach or dividing method to carry out corresponding research.The patent No. is that the patent of invention < < of 200410058167.X discloses a kind of relevant processing predicting small scale faults and crack for the relevant disposal route > > of dominant frequency band of meticulous fault interpretation, through dominant frequency band coherent technique seismic section after treatment, show the minor fault of the omission making new advances, the breakpoint location of minor fault is more accurate, and direction of extension is continuous; The patent No. is that the patent of invention < < of the CN200910138385.7 multiattribute frequency division imaging method > > based on wavelet transformation carries out wavelet transformation to the geological data within the scope of pretreated effective spectrum, to improve recognition capability and the detectability of lithologic trap, stratigraphic trap, little discontinuum, and then improve the precision of reservoir fine prediction.In recent years visible, frequency splitting technology and coherent technique development are very fast, and comparative maturity.But, perfect not enough for the forecasting techniques of deep karst type reservoir, relatively low to portraying with identification prediction precision of karst border, be mainly reflected in the following aspects:

(1) adopt dominant frequency band to tend to cause the information to a certain frequency range to suppress, and the main reflective information of the geologic body of different scales appear in different frequency band ranges sometimes.

(2) wavelet transform process easily causes geology illusion, thereby disturbs seismic interpretation, and then the identification of impact to target geologic body.

(3) also do not solve and how multiple frequency division bodies are carried out preferably and combination, and then optimum extraction goes out the problem of the earthquake information of target geologic body.

Summary of the invention

The object of the invention is to solve the difficult problem existing in above-mentioned prior art, a kind of method that improves detection of karst cave precision is provided, can identify exactly the karst region plastid of different growth scales.

The present invention is achieved by the following technical solutions:

A method that improves detection of karst cave precision, comprises the following steps:

1. frequency division processing: input poststack pure wave is protected width geological data, then poststack pure wave is protected to width geological data and carries out frequency division processing, obtains a series of frequencies frequency division data volume from low to high;

2. coherence data processing: use coherent algorithm to carry out computing to described different frequency division data volume respectively, obtain a series of frequency division coherence data body;

3. weighted data processing: utilize optimal weighted factor to be weighted data processing to described a series of frequency division coherence data body, obtain reconstruct data body.

Wherein,

1. described step comprises the following steps:

(11) the conduct a research well-log information in district and poststack pure wave protected the Fine calibration of width geological data, being about to karst germinal layer segment mark fixes on poststack pure wave guarantor width geological data section, and this karst is grown to interval as target phase, this target phase is carried out to spectrum analysis, obtain the effective band scope that the poststack pure wave of study area is protected width geological data, then according to the rising of described effective band scope design frequency division, frequency band range only;

(12), if well data meets modeling condition, utilize the karst region plastid of known different scales just drilling calculating, to determine the tuned frequency of study area different scales karst region plastid, within this tuned frequency should be included in effective band scope;

The formula of just drilling calculating is as follows:

f＝V/4H

Wherein, H is the thickness of Karst Geological Landscape body Model, and f is tuned frequency, and V is interval velocity;

If well data does not meet modeling condition, utilize inversion method to ask for tuned frequency.By the coherence data body that described poststack pure wave guarantor width geological data is obtained after coherent algorithm is calculated, along karst developing stratum, extract coherence slice, choose the coherent value abnormity point scope that the karst of different growth scales in slice plane causes, and the poststack pure wave of this scope is protected to width seismic data volume and carry out time frequency analysis, with this, add up the tuned frequency of the karst region plastid that obtains different growth scales;

(13) tuned frequency of the karst region plastid to the described different growth scales that obtain in step (12) carries out statistical study and obtains spectrum analysis achievement, the tuned frequency that marks off the karst region plastid of different growth scales is concentrated the frequency range distributing, according to the parameter of described spectrum analysis achievement design frequency division data, described parameter comprises frequency division number and filtering frequency range again;

(14) application discrete Fourier transformation is protected width geological data described poststack pure wave and is transformed to frequency field, then by the described parameter of step (13) design, utilizes frequency filtering method to generate the frequency division data volume of a series of different frequency ranges.

2. described step comprises the following steps:

(21) choose coherent algorithm (C ₁, C ₂or C ₃), then the apparent dip α in the x in coherent algorithm, y direction and β are carried out to parameter testing, find α in the time of accurately identifying the Karst Features in test zone and the value of β, the part that described test zone is study area;

(22) value of applying this α and β, to the calculating that is concerned with of the frequency division data volume of a series of different frequency ranges described in study area, obtains a series of frequency division coherence data body.

3. described step comprises the following steps:

(31) ask for the weighting factor of described frequency division coherence data body; Described weighting factor calculates by equivalent method or non-equivalent method; Then from all weighting factors, select good weighting factor, described good weighting factor refer to can comprehensively give prominence to predict in, the weighting factor on karst region plastid border on a small scale;

(32) for described good weighting factor, carry out coverage goal geologic body and after interior data reconstruction among a small circle, ask for its corresponding evaluation factor, then find the minimum evaluation factor, the weighting factor that the evaluation factor pair of described minimum is answered is optimal weighted factor;

The computing formula of the described evaluation factor is as follows:

$\begin{array}{c}\stackrel{\&OverBar;}{B}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{M}_i/{M^{\&prime;}}_i\\ n\&GreaterEqual;2\end{array}$

In formula,

for evaluating the factor, M _ibe the coherent value of the target geologic body of i point reconstruct data body, M ' _ibe the coherent value of the target geologic body of the frequency division coherence data body of i point poststack pure wave, n is all sampling point numbers in the partial data body used of measuring and calculation;

(33) optimal weighted factor that adopts step (32) to obtain to described frequency division coherence data body carries out study area data reconstruction and obtains reconstruct data body;

The computing formula of carrying out data reconstruction is as follows:

$Y=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i{C}_i$

Wherein: Y is weighted data reconstruct data body after treatment; G _ibe i weighted frequency coherence data body; Wherein k _ibe the optimal weighted factor of i weighted frequency coherence data body,

and 0≤k _i.

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

(1) utilize the inventive method can identify more exactly the form of karst region plastid, search out the growth region of favourable karst region plastid;

(2) the inventive method implementation step is simple, quick and feasible;

(3) utilize the obtained geological effect of the inventive method more more clear than the border of the described karst region plastid of routine techniques method, karst region plastid imaging effect is better, particularly in, the prediction of small-scale karst region plastid.The explanation as an example of Oil-Gas Exploration in South China block example, chance karst reservoir has been bored in the existing a large amount of drilling well checkings of this block, and test obtains middle and high product industrial gas in this karst reservoir.Known to disclosing by the known drilling well in study area, achievement and drilling data match result that the method is implemented are better, in can doping exactly, small-scale karst region plastid.

Accompanying drawing explanation

Fig. 1 is the step block diagram that the present invention improves detection of karst cave precision methods.

Fig. 2 is the time frequency analysis result of certain well karst point in the embodiment of the present invention.

Fig. 3 is the frequency division coherent body horizon slice that the frequency range in the embodiment of the present invention is 30hz～34hz.

Fig. 4 is the frequency division coherent body horizon slice that the frequency range in the embodiment of the present invention is 45hz～49hz.

Fig. 5 is the frequency division coherent body horizon slice that the frequency range in the embodiment of the present invention is 55hz～59hz.

Fig. 6 is that reconstruct data body in the embodiment of the present invention is along layer coherence slice.

Fig. 7 is that conventional poststack coherence data body in the embodiment of the present invention is along layer coherence slice.

Embodiment

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

The tuned frequency achievement that first the inventive method utilizes poststack pure wave to protect width seismic data spectrum analysis achievement, different scales karst is protected width seismic data to poststack pure wave and is carried out frequency division design and computing, obtains a series of frequency division data volumes; Again this series of frequency division data volume is carried out to coherent computing, obtain frequency division coherence data body; Secondly, by the weighting factor to these frequency division coherence data body devise optimum, be beneficial to the response of outstanding different scales karst region plastid; Next step uses this weighting factor to carry out data fusion to frequency division coherence data body, obtains reconstruct data body.This reconstruct data body is extracted corresponding section or implements three-dimensional visualization along karst section, thereby obtain the details of target geologic body.

The relevant weighted array technology of frequency division that the present invention adopts, utilizes the target geologic body response corresponding to different frequency section of geological data, analyzes and extract these response frequency sections, abandons other frequency band data that affect these frequency bands; Recycling coherent body technique is portrayed careful, feature accurately to target geologic body, and these response frequency sections are carried out to coherent computing, multiple frequency division coherence data bodies of a series of and underground multiple geologic body response the bests that obtain; Secondly these frequency division coherence data bodies are utilized to optimal weighted factor reconstruct data body, thereby reach the object of Accurate Prediction target geologic body.

Weighted data is processed the frequency division coherence data body that can reflect different target geologic body feature exactly by different weighting factors, adopt the mode of ranking operation to realize fusion demonstration, thereby realize reinforcement and the complementation to different target geologic body feature between frequency division coherence data body.This technology has the function that abnormal signal is amplified, and the power that can effectively identify different target geological objects boundary changes.Its computing method are:

$Y=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i{C}_i---\left(1\right)$

In formula (1): Y is that weighted data is processed the data volume after merging; C _ibe i weighted data body; k _ibe the optimal weighted factor of i data,

As shown in Figure 1, the method that the present invention improves detection of karst cave precision comprises the following steps:

1. frequency division processing: adopt the tuned frequency achievement of poststack pure wave guarantor's width seismic data and spectrum analysis, different scales karst, these data are carried out to frequency division design and processing, thereby obtain a series of frequencies frequency division data volume from low to high.

2. coherence data processing: respectively different frequency division data volumes is carried out to computing by coherent algorithm, obtain a series of frequency division coherence data body, the calculating parameter of coherent algorithm is made corresponding adjustment depending on geological data.

3. weighted data processing: utilize optimal weighted factor to carry out corresponding weighted data processing to a series of frequency division coherence data bodies, just obtained a reconstruct data body.

Wherein, 1. described step specifically comprises the steps:

(11) carry out poststack pure wave and protect well-shake data Fine calibration of width geological data, utilize well-log information to carry out composite traces and demarcate geological data, be about to certain zone of interest degree of depth in well and demarcate the particular location on seismic section by the conversion of time and the degree of depth, calibrate in seismic section karst developmental section in corresponding well, and karst is grown to target phase and carry out spectrum analysis, with this, obtain study area poststack pure wave and protect width geological data effective band scope, with this, design the rising of frequency division, frequency band range only;

(12) if well data meets modeling condition, utilize the karst region plastid of known different scales just drilling calculating, to determine the tuned frequency of study area different scales karst region plastid, within this tuned frequency should be included in effective band scope, computing formula is as follows:

f＝V/4H (16)

In formula (16), H is the thickness of Karst Geological Landscape body Model, and f is tuned frequency, and V is interval velocity.

If well data does not meet modeling condition, utilize inversion method to ask for tuned frequency.By the coherence data body that described poststack pure wave guarantor width geological data is obtained after coherent algorithm is calculated, along karst developing stratum, extract coherence slice.Choose the coherent value abnormity point scope that the karst of different growth scales in slice plane causes, and the poststack pure wave of this scope is protected to width seismic data volume and carry out time frequency analysis, with this, add up the tuned frequency of the karst region plastid that obtains different growth scales;

(13) utilize the tuned frequency of the karst of the different scales obtaining in step (12) to carry out step-length and the design of filtering frequency range to poststack pure wave guarantor width seismic data volume, some center frequency value should have certain corresponding relation (identical or approaching) with the tuned frequency statistical value of different scales karst, and specific formula for calculation is as follows respectively:

(a) band limits of design filtering, designs centre frequency in the centre of band limits, and computing formula is:

In formula

for i frequency range after filtering is stopped frequency,

for i frequency range after filtering plays frequency,

be the center frequency value of i frequency range, Δ d, for design band limits, as Δ d=0, is single-frequency body, design

should there be some center frequency value consistent or close with the tuned frequency of different scales karst in study area, the minimum only frequency that plays frequency and maximum should be included within the scope of effective band, between frequency range and frequency range can be overlapping, but overlapping range should be not more than 0.25 Δ d.

(b) utilize Δ d and

two parameters, design frequency division number and step-length, computing formula is as follows:

In formula (17), k is divider ratio, for k the last divider ratio centre frequency of design, for first divider ratio centre frequency of design, n is step-length, and i is i divider ratio, i≤k.

(14) application discrete Fourier transformation (DFT) is protected width geological data poststack pure wave and is transformed to frequency field, by the frequency division parameter of design, utilizes frequency filtering method to generate a series of frequency division data volumes.

In the present embodiment, by geological data is carried out to spectrum analysis, the effective band scope that obtains geological data is 10hz～80hz.In Fig. 2, the karst section time frequency analysis result on Nei Mou well point, study area is shown, medium-scale karst (drilling data shows the thick 21m of karst) tuned frequency (as the arrow indication place in Fig. 2) between 40hz～42hz on this well point, the calculating of karst forward model and time frequency analysis result statistics are carried out in study area, show that different scales karst tuned frequency great majority main in study area are distributed in 32hz～57hz scope left and right, concrete condition is fairly large karst quantity relatively less (23m-30m), its tuned frequency is distributed in 31hz～34hz frequency range more, secondly be medium scale karst quantity relatively many (16m-22m), its tuned frequency is distributed in 46hz～48hz frequency range more, also relatively less (10m-15m) of karst quantity on a small scale, its tuned frequency is distributed in 55hz～58hz frequency range more.In example, design center frequency is that 32hz rises, and to center frequency range, is that 57hz finishes.Namely to play Frequency Design be 30hz to low-frequency range, and it is 59hz that high band stops frequency, and all in effective band scope, design step-length is 5hz, and frequency division number is 6, and each band limits is 4hz.Design totally 6 frequency range data, be respectively 30hz～34hz frequency range, 35hz～39hz frequency range, 40hz～44hz frequency range, 45hz～49hz frequency range, 50hz～54hz frequency range, 55hz～59hz frequency range.

Wherein, 2. described step specifically comprises the following steps:

(21) selected C ₂coherent algorithm (also can be selected C1 or C3 algorithm, the computing formula difference of three kinds of algorithms.) and the apparent dip-α in the parameter x in coherent algorithm, y direction and β are carried out to corresponding test, parameter testing is to reach the interior Karst Features of the clear description test zone of energy (test zone is the fritter in survey region) as best, be completed, apply this parameter to the calculating that is concerned with of all frequency division data volumes.C ₂coherent algorithm is mainly first to define one centered by analysis site and window when the rectangle that contains n road geological data or ellipsometric analysis, and local coordinate axis is used as to the center of analysis site, so similarity coefficient σ (τ, α, β) computing formula is:

$\mathrm{\&Sigma;}(\mathrm{\&tau;},\mathrm{\&alpha;},\mathrm{\&beta;})=\frac{{\left[\underset{n-1}{\overset{J}{\mathrm{\&Sigma;}}}u(\mathrm{\&tau;}-\mathrm{\&alpha;xn}-\mathrm{\&beta;n},\mathrm{xn},\mathrm{yn})\right]}^2+{\left[\underset{n-1}{\overset{J}{\mathrm{\&Sigma;}}}{u}^H(\mathrm{\&tau;}-\mathrm{\&alpha;xn}-\mathrm{\&beta;n},\mathrm{xn},\mathrm{yn})\right]}^2}{J\underset{n-1}{\overset{J}{\mathrm{\&Sigma;}}}\{J{[\mathrm{\&tau;}-\mathrm{\&alpha;xn}-\mathrm{\&beta;n},\mathrm{xn},\mathrm{yn}]}^2+{\left[u(\mathrm{\&tau;}-\mathrm{\&alpha;xn}-\mathrm{\&beta;n},\mathrm{xn},\mathrm{yn})\right]}^2\}}$

In above formula, triple variablees (τ, α, β) have defined a part plan at time τ place, and α and β are respectively x, the apparent dip in y direction, and H represents Hilbert transform.The value that changes α, β is adjusted part plan vergence direction, obtains a similarity coefficient.By finding out maximum similarity coefficient σ, part plan is reached and the best-fit of real reflecting interface, similarity coefficient σ is now just coherence's estimated value of analysis site, α, β are present analysis point at x, the apparent dip in y direction.

In the present embodiment, adopt C ₂coherent algorithm is also got α and β is respectively 3 roads and carries out coherent body computing, obtains a series of frequency division coherence data bodies.In actual test, two parameters are from α, β are 3 roads, and increase progressively with certain step-length 2 roads.Same frequency division data volume is calculated the data volume of a series of different calculating parameters and extracts respectively Attribute Discovery along karst layer position by parameter testing requirement, calculating parameter adopts α and β is respectively 3 roads and 5 road result difference are little, but 3 road parameters slightly well, select α and β to be respectively 9 roads and occur fuzzy in karst border.So this example mainly adopts α and β to be respectively 3 roads to carry out coherent body computing.To the coherence slice Analysis on Results of Fig. 3, Fig. 4 and Fig. 5, the karst of the corresponding different growth scales of visible different frequency range, Karst Features is clear.

Wherein, 3. described step specifically comprises the steps:

(31) optimal weighted factor k _ivalue generally can be designed accordingly and be tested and evaluate by equivalent method or non-equivalent method, in design can by large weighting factor value suitably to in, on a small scale the more consistent frequency division coherence data body of karst region plastid tuned frequency tilts, recycling is to evaluating the weighting factor preferably of asking for and analyze and then evaluate out relatively of the factor, and concrete condition is depending on corresponding geological data and target geologic body.Frequency division data corresponding weighting factor in a data reconstruction, can carry out different weighting factor loop tests, finally will optimize in that optimum factor of each data combination.The detailed computing formula of weighting factor is as follows:

A, equivalent method: adopt each weighting factor equal and opposite in direction or unanimous on the whole carrying out, computing formula is:

$k_1=k_2=.........=k_n=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i/n---\left(23\right)$

$\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i=1---\left(24\right)$

K in above formula _ibe the weighting factor of i weighted data body, the data volume number that n is weighting.

B, non-equivalent method: adopt each frequency division coherence data in combination to calculate in seismic amplitude value or the coherent value of sampling karst point, weighting factor principle of design is that mainly the optimum outstanding border of target of prediction geologic body of wanting is for best, and its computing formula is:

$k_i={A^{\&prime;}}_i/\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{A}_i---\left(25\right)$

$\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{A^{\&prime;}}_i=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{A}_i---\left(26\right)$

$\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i=1---\left(27\right)$

K in above formula _ibe the weighting factor of i frequency division coherence data body, A _ibe actual amplitude value or the coherent value of i weighted data body on this sampling karst point, A ' _ifor actual amplitude value or the coherent value of i weighted data on this sampling karst point of design, n is the number of weighted data body.

(32) asking for the evaluation factor can evaluate this weighting factor.Because the effect of weighting be mainly outstanding in, small-scale Karst Features.During being embodied as respectively of effect assessment extracted by n sampling point reconstruct data volume and poststack pure wave coherence data body (same coherent algorithm and parameter), the coherent value in karst region on a small scale, both perform mathematical calculations, ask for the evaluation factor, evaluate the factor less, reconstruct effect is better.Weighting factor and this evaluation factor have cause-effect relationship, first have one group of weighting factor of designing, then utilize these factors to carry out after data reconstruction, then evaluate asking for of the factor, and like this, one group of corresponding one of weighting factor is evaluated the factor.Weighting factor on the same group is not asked for to the corresponding evaluation factor, is actually this many groups weighting factor is carried out to feedback evaluation, constantly circulate evaluate after so that obtain the process of relatively optimum weighting factor.The computing formula of evaluating the factor is as follows:

$\stackrel{\&OverBar;}{B}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{M}_i/{M^{\&prime;}}_i---\left(28\right)$

n≥2(29)

In above formula

for evaluating the factor, M _ibe that (coherent body is data volume, and the each data in coherent body are corresponding to some points of a geologic body for the coherent value of the target geologic body of i point reconstruct data body.In coherent body, the concrete data of certain point, are exactly coherent value), M ' _iit is the coherent value (getting in the calculating of the frequency division coherence data body of asking for from step (22)) of the target geologic body of i point poststack pure wave coherence data body, n is all sampling point numbers in the partial data body used of measuring and calculation, this partial data body is called body of test data, and its selection principle should comprise the karst of different growth scales as far as possible; .It has been generally acknowledged that

be good; When

shi Zewei is good; When

shi Zewei is excellent.

(33) to frequency division coherence data body, utilize optimal weighted factor to carry out data reconstruction and obtain reconstruct data body, data reconstruction computing formula is as follows:

$Y=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i{C}_i---\left(22\right)$

In formula (22): Y is weighted data reconstruct data body after treatment, utilizing this reconstruct data body can extract and analyze karst and grow interval, is the same with conventional data attribute extraction and analytical approach; C _ibe i weighted frequency coherence data body; Wherein k _ibe the optimal weighted factor of i weighted frequency coherence data body, $\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i=1,$ And 0≤k _i.

In the present embodiment, mainly give prominence to medium scale karst and take into account little, large-scale karst, it is 0.2,0.1,0.1,0.3,0.1,0.2 that 6 frequency division coherence data bodies are designed to optimum corresponding weighting factor by non-geometric ratio method, evaluating the factor is 0.52, for well.

Relatively utilize the obtained achievement of the inventive method (as shown in Figure 6) and utilize the achievement (as shown in Figure 7) of routine techniques gained to find out, local detail on Fig. 7 and sharpness etc. are all not as good as Fig. 6, that is the solution cavity of Fig. 6 middle and small scale has all suffered compacting extremely in Fig. 7, solution cavity extremely on Fig. 7 show not out or details fuzzy, secondly portraying obviously not as Fig. 6 karst band, near karst growth zone particularly between X6 and X12 well, in Fig. 6 show clearer, in Fig. 7, do not show or show fuzzy.

The present invention is with frequency division, the diffusion-weighted data reconstruction technology of phase dry doubling, thereby exactly karst region plastid is described.By frequency division, extract the response frequency data to different scales karst region plastid, and these data are carried out to corresponding coherent body computing; Secondly, frequency division coherence data body is weighted and forms reconstruct data body, thereby extract exactly corresponding earthquake information, increase substantially the exploration economic benefit of deep karst reservoir.

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

Claims (4)

1. a method that improves detection of karst cave precision, is characterized in that: said method comprising the steps of:

1. frequency division processing: input poststack pure wave is protected width geological data, then poststack pure wave is protected to width geological data and carries out frequency division processing, obtains a series of frequencies frequency division data volume from low to high;

2. coherence data processing: use coherent algorithm to carry out computing to described different frequency division data volume respectively, obtain a series of frequency division coherence data body;

3. weighted data processing: utilize optimal weighted factor to be weighted data processing to described a series of frequency division coherence data body, obtain reconstruct data body.

2. the method for raising detection of karst cave precision according to claim 1, is characterized in that: 1. described step comprises the following steps:

(11) the conduct a research well-log information in district and poststack pure wave protected the Fine calibration of width geological data, being about to karst germinal layer segment mark fixes on poststack pure wave guarantor width geological data section, and this karst is grown to interval as target phase, this target phase is carried out to spectrum analysis, obtain the effective band scope that the poststack pure wave of study area is protected width geological data, then according to the rising of described effective band scope design frequency division, frequency band range only;

(12), if well data meets modeling condition, utilize the karst region plastid of known different scales just drilling calculating, to determine the tuned frequency of study area different scales karst region plastid, within this tuned frequency should be included in effective band scope;

The formula of just drilling calculating is as follows:

f＝V/4H

Wherein, H is the thickness of Karst Geological Landscape body Model, and f is tuned frequency, and V is interval velocity;

If well data does not meet modeling condition, utilize inversion method to ask for tuned frequency.By the coherence data body that described poststack pure wave guarantor width geological data is obtained after coherent algorithm is calculated, along karst developing stratum, extract coherence slice, choose the coherent value abnormity point scope that the karst of different growth scales in slice plane causes, and the poststack pure wave of this scope is protected to width seismic data volume and carry out time frequency analysis, with this, add up the tuned frequency of the karst region plastid that obtains different growth scales;

(13) tuned frequency of the karst region plastid to the described different growth scales that obtain in step (12) carries out statistical study and obtains spectrum analysis achievement, the tuned frequency that marks off the karst region plastid of different growth scales is concentrated the frequency range distributing, according to the parameter of described spectrum analysis achievement design frequency division data, described parameter comprises frequency division number and filtering frequency range again;

(14) application discrete Fourier transformation is protected width geological data described poststack pure wave and is transformed to frequency field, then by the described parameter of step (13) design, utilizes frequency filtering method to generate the frequency division data volume of a series of different frequency ranges.

3. the method for raising detection of karst cave precision according to claim 2, is characterized in that: 2. described step comprises the following steps:

(21) choose coherent algorithm, then the apparent dip α in the x in coherent algorithm, y direction and β are carried out to parameter testing, find α in the time of accurately identifying the Karst Features in test zone and the value of β, the part that described test zone is study area;

(22) value of applying this α and β, to the calculating that is concerned with of the frequency division data volume of a series of different frequency ranges described in study area, obtains a series of frequency division coherence data body.

4. the method for raising detection of karst cave precision according to claim 3, is characterized in that: 3. described step comprises the following steps:

(31) ask for the weighting factor of described frequency division coherence data body; Described weighting factor calculates by equivalent method or non-equivalent method; Then from all weighting factors, select good weighting factor, described good weighting factor refer to can comprehensively give prominence to predict in, the weighting factor on karst region plastid border on a small scale;

(32) for described good weighting factor, carry out coverage goal geologic body and after interior data reconstruction among a small circle, ask for its corresponding evaluation factor, then find the minimum evaluation factor, the weighting factor that the evaluation factor pair of described minimum is answered is optimal weighted factor;

The computing formula of the described evaluation factor is as follows:

$\begin{array}{c}\stackrel{\&OverBar;}{B}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{M}_i/{M^{\&prime;}}_i\\ n\&GreaterEqual;2\end{array}$

In formula,

for evaluating the factor, M _ibe the coherent value of the target geologic body of i point reconstruct data body, M ' _ibe the coherent value of the target geologic body of the frequency division coherence data body of i point poststack pure wave, n is all sampling point numbers in the partial data body used of measuring and calculation;

(33) optimal weighted factor that adopts step (32) to obtain to described frequency division coherence data body carries out study area data reconstruction and obtains reconstruct data body;

The computing formula of carrying out data reconstruction is as follows:

$Y=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{k}_i{C}_i$

Wherein: Y is weighted data reconstruct data body after treatment; C _ibe i weighted frequency coherence data body; Wherein k _ibe the optimal weighted factor of i weighted frequency coherence data body, and 0≤k _i.

Images