CN105089652A - Pseudo-acoustic curve rebuilding and sparse pulse joint inversion method - Google Patents

Abstract

The invention provides a pseudo-acoustic curve rebuilding and sparse pulse joint inversion method, and belongs to the field of reservoir prediction research. The method comprises the following steps of: 1, performing cross analysis on an acoustic curve and a natural gamma-ray curve; performing statistics on a mathematical relationship of an acoustic parameter and a natural gamma-ray parameter according to the cross analysis; if the two curves have good consistency, regarding that the oil and gas reservoir prediction by the natural gamma-ray curve is effective; and entering the second step; 2, using a wavelet multi-scale transform and information fusion technology to rebuild a DT curve (acoustic-time difference curve) to obtain a pseudo-acoustic curve, wherein the curve includes both natural gamma-ray features and acoustic features; and 3, performing sparse pulse seismic inversion under the DT curve constraints to obtain a final inversion result.

Description

A kind of pseudo-acoustic curve reconstruct and Sparse Pulse joint inversion method

Technical field

The invention belongs to reservoir prediction research field, be specifically related to the reconstruct of a kind of pseudo-acoustic curve and Sparse Pulse joint inversion method, the reconstruct of associating pseudo-acoustic curve and Sparse Pulse Inversion technology carry out reservoir prediction work, thus improve reservoir prediction precision.

Background technology

Optimum Impedance Inversion Method originates in the end of the seventies, widely popular to the eighties.Early stage mainly common recurrence method (as VELOG, SEISLOG etc.) and Sparse Pulse method (as MED, AR, MLD, BED etc.).This technical characterstic is directly from seismic data inverting, well-log information is mainly used in demarcating, and the resolution ratio of inversion result is directly by the restriction of seismic data frequency bandwidth, and inverting Resolution of Acoustic Impedance Section is relatively low, general between 20m-50m, the needs of thicker reservoir study can only be met.The end of the eighties is to the beginning of the nineties, occur that one combines with earthquake, well logging, geology, based on convolution model and geological model, one that is linear and generalized linear inversion is adopted to be referred to as " model inversion method ", the feature of this technology is, well-log information is not only for demarcating, and join in refutation process, inversion result resolution ratio is by seismic resolution and well logging resolution ratio combined influence, the absolute wave impedance of visual resolution (exceeding seismic resolution) 5m-10m can be finally inversed by, to some extent solve thinner reservoir problem.The domestic software comparatively early introducing wave impedance inversion has DELOG, SEISLOG, SLIM, PARM, BCI, ROVIM etc., introduces the softwares such as ISIS, Jason, StrataVista, Vanguard again afterwards.Domestic development have the softwares such as SOML, CCFY, CRIS, RICH, CONSEIS.

Since Connolly (1999) has delivered the paper of elastic wave impedance (EI), domestic and international geophysics educational circles has started the upsurge of elastic impedance inversion research.Elastic impedance inversion belongs to prestack inversion technology, and it contains abundant lithology and fluid information.The concept of elastic impedance and theoretically also to deepen continuously along with application and the development of seismic technology itself.Whitcombe (2002) etc. revises for Connolly formula, the equation of expansion elastic impedance (Extendedelasticimpedance, EEI) of having derived, and can be directly used in lithology and fluid prediction.That makes contributions to elastic impedance research also has, Duffaut etc. (2000) propose the concept of the shear wave elastic impedance (SEI) for P-SV converted wave problem, the formula of converted wave elastic impedance (PSEI) at any angle of having derived such as Ezequiel (2003), and be directly used in reservoir and predict containing the area distribution of micro-gas (fizzwater) and industrial gas (commercialgas).From 2003, geophysics educational circles of China horse strong wind, sweet sharp lamp, Ni Yi, Wang Baoli etc. also start the research work paying close attention to this respect, propose the concept of ray elastic impedance, broad sense elastic impedance, and a large amount of analysis and research have been done to the application potential etc. of elastic impedance.

Elastic impedance inversion directly can carry out fluid prediction, but the method application desired data is geological data before superposition road collection and needs shear wave data etc., this conditional request limits the method and applies on a large scale, and the seismic inversion application therefore based on poststack seismic data is comparatively extensive.

Along with going deep into of lithologic reservoir forecasting research, reservoir and non-reservoir (country rock) parameter differences not obvious time, common seismic inversion technique is utilized to be difficult to identify reservoir preferably, thus Well logging curve reconstruction technology is suggested and is applied in reservoir prediction, the method effectively can improve the precision of Lithology Discrimination and reservoir prediction, reduces multi-solution.

The horizontal uniformity of seismic data that to utilize seismic inversion to carry out reservoir prediction be integrated application better and the higher advantage of well-log information longitudinal frame carry out the description work of reservoir.But along with deepening continuously of oil-gas exploration, subtle pool, lithologic deposit, unconventionaloil pool Tibetan etc. have become the important oil-gas reservoir that the public pays close attention to, and prospecting prediction difficulty also increases thereupon.At present, the carrying out reservoir prediction work can not satisfy the demand based on model, iteration isoinversion method of conventional is utilized.Particularly Sonic Logging Data is by hole collapse, the impact of the wellhole environment such as mud dipping, simultaneously also by the impact of well logging cycle, mud composition, reservoir degree of consolidation and degree of porosity, correctly can not reflect the difference of reservoir and country rock, cause Lithology Discrimination difficulty, on inverting section, Resolution of Acoustic Impedance Section reduces, even if by carrying out correction process to sound wave curve, in certain areas, due to the impact of the factor such as differential compaction and calcareous content, reservoir and country rock also no significant difference on acoustic logging.

Summary of the invention

The object of the invention is to solve the difficult problem existed in above-mentioned prior art, provide a kind of pseudo-acoustic curve to reconstruct and Sparse Pulse joint inversion method, improve the accuracy of reservoir prediction.

The present invention is achieved by the following technical solutions:

A kind of pseudo-acoustic curve reconstruct and Sparse Pulse joint inversion method, comprising:

The first step, intersection analysis is carried out to sound wave curve and gamma ray curve, and according to the relationship of intersection analytic statistics acoustic wave parameter and natural gamma parameter, if two curve has good uniformity, then think that it is effective for utilizing gamma ray curve to carry out oil and gas reservoir prediction, enters second step;

Second step, utilize multi-scale wavelet transform and information fusion technology reconstruct DT curve (interval transit time curve) to obtain pseudo-acoustic curve, this curve not only comprises natural gamma feature but also comprises acoustic characteristic;

3rd step, carries out the Sparse Pulse seismic inversion under DT curve constraint, obtains final inversion result.

Described second step comprises:

(21) multi-scale wavelet transform is utilized to extract the low-frequency component of sound wave curve, obtain the low frequency log that can reflect stratum background removing high frequency components, and remove the feature that the curve after low frequency does not have sound wave, but (this curve be not utilized to have the feature of GR, and merged by the gamma ray curve and low frequency log that comprise high-frequency information in next step and generate pseudo-acoustic curve, this curve not only comprises sound wave curve feature but also comprise gamma ray curve feature.)；

(22) utilizing information fusion technology to be dissolved into by the low frequency component extracted in step (21) comprises in the gamma ray curve of high-frequency information, is finally not only had the low-frequency information of stratum background but also can reflect the pseudo-acoustic curve that formation lithology changes.

Described 3rd step is achieved in that

Using the pseudo-acoustic curve that obtains in second step as original acoustic wave curve, obtain SVEL by formula v=s/DT, then formula z (i)=υ (i) * ρ (i) is substituted into surge impedance model formula in complete the calculating of surge impedance model, calculate the wave impedance from the wave impedance on first floor stratum to target zone successively, obtain final inversion result;

Wherein, v is SVEL, s is Acoustic Wave Propagation distance, DT is interval transit time, z (i) is the wave impedance of i-th layer of bed boundary, υ (i) is the SVEL (being exactly obtained by formula v=s/DT) in i-th layer of stratum, and ρ (i) is the density on i-th layer of stratum, and R (i) is the reflectance factor of i-th layer: $R\left(i\right)=\frac{\mathrm{\υ}\left(i\right)*\mathrm{\ρ}\left(i\right)-\mathrm{\υ}(i-1)*\mathrm{\ρ}(i-1)}{\mathrm{\υ}\left(i\right)*\mathrm{\ρ}\left(i\right)+\mathrm{\υ}(i-1)*\mathrm{\ρ}(i-1)}.$

Compared with prior art, the invention has the beneficial effects as follows: the forecasting research of reservoir is carried out in integrated application pseudo-acoustic curve reconfiguration technique of the present invention and Sparse Pulse Inversion technical tie-up, first the relation of log intersection statistical analysis sound wave curve and gamma ray curve is passed through in research, show that acoustic wave parameter and natural gamma parameter have good uniformity in constituency, also show simultaneously and natural gamma parameter can be utilized to carry out effective reservoir prediction; Then sound wave curve and natural gamma reconstructed a curve D T not only having comprised natural gamma feature (sign of lithology) but also comprised acoustic characteristic (seismic reflection feature) according to multi-scale wavelet transform and information fusion technology; Finally carry out the Sparse Pulse Inversion method predicting reservoir under DT curve constraint, prediction effect accuracy is higher.Novelty of the present invention pseudo-acoustic curve reconfiguration technique is had advance with the theory of carrying out reservoir prediction that combines of Sparse Pulse Inversion technology effective, the application of the method effectively raises the prediction accuracy of oil and gas reservoir, has constructive, practicality.

Accompanying drawing explanation

Fig. 1 is sound wave curve and gamma ray curve intersection.

Fig. 2-1 is former sound wave curve.

Fig. 2-2 is gamma ray curves.

Fig. 2-3 is reconstruct DT curves.

Fig. 3-1 is conventional wave impedance inversion section.

Fig. 3-2 is that well profile is crossed in curve Reconstruction and Sparse Pulse Inversion joint inversion.

Fig. 4 is the step block diagram of this method.

Detailed description of the invention

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

Select Well logging curve reconstruction technology can utilize the log that resolution ratio is relatively high, reconstruct the curve that can reflect subsurface lithologic feature better, for model constrained seismic inversion provides a rational initial model, decrease the multi-solution of inverting, improve the precision of reservoir prediction.Seismic inversion why can predicting reservoir or even thin reservoir, this is because inversion method effectively can utilize the feature that well logging resolution ratio is high, seismic inversion predicting reservoir accuracy is carried out higher under borehole restraint condition, Sparse Pulse Inversion is exactly the effective ways carrying out reservoir prediction under borehole restraint condition, so utilize the curve that curve Reconstruction technology reengineering one can be responsive especially to lithology in this research, utilize this curve to carry out the joint inversion technical research of Sparse Pulse Inversion technological prediction reservoir, have broad application prospects.

(1) pseudo-acoustic curve reconstruct

1) Method And Principle

The realization of the method realizes according to the well log curve relevant to reservoir characteristic response characteristic and activity curve, and by curve intersection, choose reasonable characterizes the log parameter of reflection reservoir characteristic, the new property parameters of reconstruct reflection reservoir characteristic; The pseudo-acoustic curve rebuilding a reflection reservoir characteristic based on sound wave curve and gamma ray curve in the present invention, then for Application of Logging-constrained Inversion.Reservoir characteristic curve Reconstruction technology can provide a rational initial model for constraint seismic inversion, the low-frequency effects that interval transit time changes along with depth of stratum can be eliminated, improve precision of prediction, decrease the multi-solution of inverting, thus improve the precision of reservoir prediction.Reconfiguration technique has following four kinds:

1. conventional log curve correction.

2. empirical formula or statistical fit.

3. Information Statistics weighting.

4. based on the reconfiguration technique of multi-scale wavelet transform.

The present invention selects the method for Corpus--based Method matching, and specific implementation step is as follows:

1. first utilize sound wave curve and reflection formation lithology to change more sensitive natural gamma curve (to answer: these two curves belong to log, be application logging technique log well in the wild exploration obtain, also be utilize seismic data to carry out indispensable log data that seismic interpretation carries out oil-gas exploration), carry out statistical regression analysis, both discoveries corresponding relation is fine, there is higher correlation, can curve Reconstruction be carried out.And in this work area, natural gamma curve more can reflect the change of formation lithology than sound wave curve, so should reach the object of curve Reconstruction.Log is a kind of form of expression of stratum rock signature, when showing same stratomere with these two curves, has certain uniformity between it; Certainly, other curve also may have this uniformity with sound wave curve, but in the present invention is gamma ray curve.As shown in fig. 1, the loose point of two curves than comparatively dense, and can simulate a curve along loose distribution in intersection chart board, then can judge between them be have better conforming.The distribution of loose point is more concentrated, and more close with the curve of matching, its uniformity is better.

2. multi-scale wavelet transform (please refer to " studying based on the shelf depreciation denoising of multi-scale wavelet transform and recognition methods ") is utilized to extract the low-frequency information of acoustic logging, obtain removing one after low frequency temporary transient pseudo-acoustic curve under this step, this temporary transient curve does not have the feature of sound wave, but has the feature of GR.

3. utilize information fusion technology (please refer to " gravity dam deformable statistical regression analysis model and engineer applied ") can reflect that the low frequency component of stratum background velocity is dissolved in high-frequency information by what extract in second step, finally not only be there is the low-frequency information of stratum background but also the pseudo-acoustic curve that formation lithology changes can be reflected.

(2) Sparse Pulse Inversion

1) reflectance factor inverting

Maximum-likelihood deconvolution is adopted to carry out the inverting of reflectance factor, the hypothesis of maximum-likelihood deconvolution formation is thought: the reflectance factor on stratum is formed by the reflection of larger reflecting interface and the little reflection stack combinations with Gaussian Background, derives a minimum target function:

$J=\frac{1}{R^2}\underset{K=1}{\overset{L}{\mathrm{\Σ}}}{r}^2\left(K\right)+\frac{1}{N^2}\underset{K=1}{\overset{L}{\mathrm{\Σ}}}{n}^2\left(K\right)-2M\ln \left(\mathrm{\λ}\right)-2(L-M)\ln (1-\mathrm{\λ})---\left(1\right)$

In formula, R ²and N ²be respectively the mean-square value of reflectance factor and noise, r (K) and n (K) represents reflectance factor and the noise of K sampled point, and M represents the reflection number of plies, and L represents sampling sum, and λ represents the likelihood value of given reflectance factor.

By successive ignition, ask for reflectance factor.

2) wave impedance calculates

According to the reflectance factor that maximum-likelihood deconvolution calculates, in conjunction with initial impedance model, adopt recursive algorithm, inverting obtains initial surge impedance model:

$z\left(i\right)=z(i-1)\frac{1+R\left(i\right)}{1-R\left(i\right)}---\left(2\right)$

In formula, z (i) is the wave impedance value of i-th layer, and R (i) is the reflectance factor of i-th layer.

3) borehole restraint wave impedance inversion

Restrict network method adjusts the impedance initial value calculated according to object function together to every, comprises the adjustment to reflectance factor.Objective optimization function is:

F＝L _p(r)+λL _q(s-d)+α ^-1L ₁ΔZ(3)

In formula, r is reflection coefficient sequence, and Δ z is the difference sequence with impedance trend, and d is seismic channel sequence, and s is synthetic seismogram sequence, and λ is residual error weight factor, and α is trend weight factor, and p, q are the L mould factor.Concrete, right formula Section 1 reflect reflectance factor absolute value and, Section 2 reflects the difference of synthesis sound wave record and original earthquake data, and Section 3 is trend bound term.

The pseudo-acoustic curve that the present invention utilizes first step curve Reconstruction to obtain in the process of carrying out Sparse Pulse Inversion is as original acoustic wave curve D T, SVEL is obtained by formula v=s/DT, then formula z (i)=υ (i) * ρ (i) is substituted into the calculating that can complete surge impedance model in formula (2), thus carry out effective reservoir inversion.Namely calculate according to surge impedance model above, parameter i=1,2,3 ..., the wave impedance of target zone so just can be calculated from the wave impedance on first floor stratum.

As shown in Figure 4, by sound wave curve and gamma ray curve intersection statistics, first the present invention show that acoustic wave parameter and natural gamma parameter have good uniformity, draw and utilize natural gamma parameter to carry out effective reservoir prediction, statistical fit goes out the relational expression between two parameter; Then carry out the reconstruct of sound wave curve and gamma ray curve according to multi-scale wavelet transform algorithm and information fusion technology, obtain a curve D T not only having comprised natural gamma feature (sign of lithology) but also comprised acoustic characteristic (seismic reflection feature); Finally, under DT curve constraint, utilize Sparse Pulse Inversion method predicting oil/gas reservoir.Higher from the accuracy of oil and gas reservoir prediction effect the method predicting reservoir, reservoir and non-storage interlayer difference are obvious, can be easier to identify, indicate pseudo-acoustic curve reconfiguration technique and Sparse Pulse joint inversion technology has advance, and practicality

Fig. 1 is that target target area well sound wave curve and gamma ray curve cross figure, and analyzed between sound wave curve and gamma ray curve by intersection and have correlation, its correlation can be expressed as: Y=-544.780+114.799*lg (x);

Fig. 2-1 to Fig. 2-3 reconstructs DT curve by sound wave curve and gamma ray curve, and this curve not only comprises seismic reflection feature but also comprises lithologic character, can be used for carrying out lithologic reservoir prediction;

Fig. 3-1 utilizes the conventional inversion result based on the mistake well of the question of seismic wave impedance inversion technological prediction reservoir of model, and this technological prediction effective reservoir is on the low side, between reservoir and non-reservoir the indifference opposite sex or otherness less, laterally scale is bigger than normal continuously for reservoir; Fig. 3-2 is for utilizing the Sparse Pulse Inversion result intended under the DT curve constraint that reconstructs of sonic reconstruction, can find out that predicting reservoir and well logging bore that to meet reservoir uniformity better from crossing the inverting section of certain well, between reservoir and non-reservoir, difference is obvious, can be easier to identify.

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 detailed description of the invention of the present invention, therefore previously described mode is just preferred, and does not have restrictive meaning.

Claims (3)

1. pseudo-acoustic curve reconstruct and a Sparse Pulse joint inversion method, is characterized in that: described method comprises:

The first step, intersection analysis is carried out to sound wave curve and gamma ray curve, and according to the relationship of intersection analytic statistics acoustic wave parameter and natural gamma parameter, if two curve has good uniformity, then think that it is effective for utilizing gamma ray curve to carry out oil and gas reservoir prediction, enters second step;

Second step, utilize multi-scale wavelet transform and information fusion technology reconstruct DT curve to obtain pseudo-acoustic curve, this curve not only comprises natural gamma feature but also comprises acoustic characteristic;

3rd step, carries out the Sparse Pulse seismic inversion under DT curve constraint, obtains final inversion result.

2. pseudo-acoustic curve reconstruct according to claim 1 and Sparse Pulse joint inversion method, is characterized in that: described second step comprises:

(21) utilize multi-scale wavelet transform to extract the low-frequency component of sound wave curve, obtain the low frequency log that can reflect stratum background removing high frequency components;

(22) utilizing information fusion technology to be dissolved into by the low frequency component extracted in step (21) comprises in the gamma ray curve of high-frequency information, is finally not only had the low-frequency information of stratum background but also can reflect the pseudo-acoustic curve that formation lithology changes.

3. pseudo-acoustic curve reconstruct according to claim 1 and Sparse Pulse joint inversion method, is characterized in that: described 3rd step is achieved in that

Using the pseudo-acoustic curve that obtains in second step as original acoustic wave curve, obtain SVEL by formula v=s/DT, then formula z (i)=υ (i) * ρ (i) is substituted into surge impedance model formula in complete the calculating of surge impedance model, calculate the wave impedance from the wave impedance on first floor stratum to target zone successively, obtain final inversion result;

Wherein, v is SVEL, s is Acoustic Wave Propagation distance, DT is interval transit time, z (i) is the wave impedance of i-th layer of bed boundary, υ (i) is the SVEL in i-th layer of stratum, and ρ (i) is the density on i-th layer of stratum, and R (i) is the reflectance factor of i-th layer: $R\left(i\right)=\frac{\mathrm{\υ}\left(i\right)*\mathrm{\ρ}\left(i\right)-\mathrm{\υ}(i-1)*\mathrm{\ρ}(i-1)}{\mathrm{\υ}\left(i\right)*\mathrm{\ρ}\left(i\right)+\mathrm{\υ}(i-1)*\mathrm{\ρ}(i-1)}$