阅读说明：本技术 一种地震驱动的高精度层序格架模型构建方法 (Seismic-driven high-precision sequence grid model construction method ) 是由 何文渊 孙平 于 2020-10-23 设计创作，主要内容包括：本发明公开了一种地震驱动的高精度层序格架模型构建方法,方法在井震标定的基础上,进行地震的层位断层解释,之后通过联合解释的层位、断层、井上分层和地震体,在解释的层位断层约束下,将井上更高级别层序的分层使用地震体的产状驱动进行高精度的层序层面的插值,插值出符合地层分布的精细的层序层位,构建高精度的层序格架模型。本发明综合利用的井上分层层序精度高与地震横向信息丰富的特性,充分利用地震产状作为驱动进行更高精度分层层序的约束插值,进而生成更高精度的层序格架模型,提高了层序格架模型的精度,模型的横向展布更为可靠,且使用高精度层序格架模型为约束的地震反演对比分析,可为高精度储层预测表征提供更有力的技术保障。(The invention discloses a method for constructing a seismic-driven high-precision sequence grid model, which is characterized by carrying out seismic horizon fault interpretation on the basis of well seismic calibration, then carrying out high-precision sequence layer interpolation on the higher-level sequence layers on the well by using attitude drive of seismic bodies under the restraint of interpreted horizon faults through jointly interpreted horizons, faults, uphole layers and seismic bodies, interpolating fine sequence horizons conforming to stratum distribution, and constructing the high-precision sequence grid model. The characteristic of high precision of the on-well layered sequence and rich transverse information of the earthquake are comprehensively utilized, the earthquake occurrence is fully utilized as a drive to carry out constrained interpolation of the higher-precision layered sequence, a higher-precision sequence grid model is further generated, the precision of the sequence grid model is improved, the transverse distribution of the model is more reliable, and the high-precision sequence grid model is used for constrained earthquake inversion contrastive analysis, so that more powerful technical guarantee can be provided for high-precision reservoir prediction and characterization.)

1. A construction method of a seismic-driven high-precision sequence grid model is characterized by comprising the following steps:

acquiring logging and core data, performing high-precision stratigraphic division and comparison on a well-connected stratigraphic sequence, and establishing a well-connected stratigraphic comparison section;

secondly, performing high-precision stratigraphic sequence stratigraphic synthesis seismic record calibration comparison based on logging data and seismic data, identifying interpretable tracing stratigraphic sequence layering according to the corresponding relation between the stratigraphic sequence layering and a seismic event on a well-connected seismic section, and performing interpretation adjustment of the stratigraphic sequence layering comparison by taking the transverse spread of the seismic section as a basis;

calculating a seismic direction field, constructing guiding filtering, identifying interpretable sequence layering on a seismic section, performing horizon interpretation of corresponding sequence layering under the driving of the seismic direction field, and establishing a structural grid model;

and fourthly, using the seismic structure lattice model as top and bottom and interlayer constraints, combining a seismic direction field as a drive, establishing constraint conditions according to the lateral variation trend of seismic data, performing lateral interpolation tracking on the calibrated high-precision sequence layers, and combining the interpolated layers with the seismic structure lattice model to generate the high-precision sequence lattice model.

2. The method for constructing the seismic-driven high-precision sequence trellis model according to claim 1, wherein the process of calculating the seismic direction field in the third step specifically comprises: calculating the seismic direction field by solving the eigenvalue of the structure tensor matrix and the eigenvector corresponding to the first eigenvalue by adopting a structure tensor method, wherein the calculation process of the structure tensor is shown as the following formula:

in the formula, GST represents a structure tensor matrix, G_xRepresenting the gradient value in the x-direction, G_yRepresenting the gradient value in the y-direction, G_xyRepresents G_xAnd G_yThe product of (a).

3. The method for constructing a seismic-driven high-precision sequence trellis model according to claim 1, wherein the step three-structure-oriented filtering process specifically comprises: under the drive of a seismic direction field, calculating an accurate guide horizon point by point, and constructing guide filtering by edge-preserving filtering or fault enhancement, wherein the edge-preserving filtering is performed by adopting the following process:

in the formula, Δ c represents the filtering weight of the horizon at each guide point and the current filtering center point, n represents the filtering radius, κ (Δ t) represents the distance between the guide point and the filtering center point, and t (Δ t) represents the difference between the horizon at the guide point and the filtering center point.

4. The method for constructing the seismic-driven high-precision sequence trellis model according to claim 1, wherein the constraint conditions obtained in the process of establishing the constraint conditions according to the lateral variation trend of the seismic data in the fourth step are represented by the following formula:

wherein Δ_i＝T_i-H₀I, T is the depth of the well in time (converted from the velocity field) in stratification, H₀To constrain the time of the horizon, D is the distance between the predicted point and the well, λ₁As a phase weight factor, λ₂As an amplitude weighting factor, λ₃Is a frequency weighting factor, S is the phase of the predicted point,is the average phase value. A is the amplitude of the predicted point and,is the average phase value, F is the frequency of the predicted point,is the average frequency value.

Technical Field

The invention relates to the technical field of oil and gas exploration, in particular to a method for constructing a seismic-driven high-precision sequence grid model.

Background

Over the last century of oil and gas exploration, the goal of oil and gas exploration has shifted from traditional tectonic to tectonic. Particularly, when oil reservoir development gradually enters the middle and later stages, high-precision prediction of distribution of underground lithologic bodies and high-permeability reservoirs is required more and more. The traditional lithology or reservoir property prediction mainly comprises two categories of reservoir parameter modeling and seismic constraint inversion, and the two categories of technologies do not leave a high-precision structural grid model as reservoir or lithology boundary constraint in the model parameter prediction process.

The construction of the traditional lattice model generally has two types of methods:

one is that the isosurface is generated by directly interpolating from the layering and the break points on the well, and the construction of the grid model is gradually completed by performing the intersection editing of the isosurface and the section in the modeling software. The method is only suitable for layering and breaking points at well points to make constraints, no information constraint is provided for transverse changes of formations and faults between wells, and the problems that the modeling process is repeated and the model is difficult to close are easily caused by the interpretation quality of the layering and breaking points on the well. And the editing workload of the section and the isosurface is large, and the local adjustment of the layering scheme can bring about the repeated adjustment of the whole modeling process. Particularly, the inter-well stratum spread characteristics are not restricted by reference information, so that the method is only suitable for development areas with small well spacing and intensive well point correction in the middle and later stages of development, cannot be suitable for rolling exploration edge expansion of development blocks, and has a narrow application range;

another method is the construction of a structural lattice by well-seismic joint interpretation. The method comprises the steps of firstly, on the basis of well seismic calibration, correcting the layering of a depth domain on the well to a time domain, selecting a corresponding homophase axis on a seismic section according to the layering to perform horizon interpretation, and then performing time-depth conversion on the interpreted seismic horizon and performing well point layering correction to serve as a layered structural layer. In addition, the method can accurately identify the break points on the seismic profile to explain the fault, and can also convert the break points into a depth domain through time-depth conversion to be used as a fault plane and input into a structural model to be used as section constraint. The first method of spread correction of the interwell stratum is more accurate because of the reference of interwell information, and the fracture surface of the fault surface phase correction point explained by the earthquake is more accurate because the information of the transverse information correction point of the earthquake body is richer in sampling number. But due to the resolution limitations of conventional seismic data itself, interpretable event often only identifies three or four levels of sequence on the well, and the identification capability for finer hierarchical sequences (five or six levels) is weaker. The method is often suitable for constructing a structural lattice model in the early exploration stage and a rough sequence lattice model in the middle and early exploration stages.

The existing method has certain limitation on the construction method of the high-precision sequence trellis model, and cannot provide effective technical support for high-precision reservoir lithology characterization.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides a method for constructing a high-precision sequence grid model based on well-seismic combination. In actual work, a stratum sequence grid model is constructed by traditional well layering information, no inter-well information constraint exists, inter-well stratum spread information is unreliable, and the method is only suitable for development blocks with small well spacing and dense well points in the middle and later periods of exploration and development. The traditional well-seismic combined seismic horizon fault interpretation method can only identify sequence of four levels and more than three levels, and can not identify sequence of five levels and six levels with higher precision. In the well-seismic combination method, although the resolution of conventional seismic data can identify the sequence strata of three or more than four levels, the horizon distribution of a higher-level sequence can be fitted by combining with the uphole layering under the constraint of the seismic event attitude (dip angle and strike).

Therefore, the invention firstly carries out seismic horizon fault interpretation on the basis of well seismic calibration, then carries out high-precision sequence layer interpolation on the horizons of the higher-level sequence on the well by using the attitude drive of the seismic body under the restriction of the interpreted horizon fault through the horizon, fault, uphole layering and the seismic body which are jointly interpreted, interpolates the fine sequence horizon which accords with the stratum distribution, and realizes the construction of a high-precision sequence trellis model.

The purpose of the invention is realized by the following technical scheme:

a method for constructing a high-precision sequence grid model based on well-seismic combination comprises the following steps:

step 1, performing high-precision sequence stratigraphic division and comparison through well logging and core data analysis, constructing a well-connected sequence stratigraphic comparison section, and providing a sequence comparison scheme for seismic interpretation;

and 2, combining the logging data and the seismic data to calibrate and compare the high-precision sequence stratigraphic synthesis seismic records. On the well-connected seismic section, identifying and interpretably tracking the sequence layering according to the corresponding relation between the sequence layering and the seismic event, and carrying out interpretation and adjustment of sequence layering comparison by taking the transverse spread of the seismic section as a basis;

and 3, calculating a seismic direction field on the basis of the seismic structure guiding filtering. And identifying interpretable sequence layers on the seismic profile, and performing horizon interpretation of the corresponding sequence layers under the driving of a seismic direction field to construct a structural grid model.

And 4, taking the seismic structure grid model as top and bottom and interlayer constraints, combining a seismic direction field as a drive, and carrying out transverse interpolation tracking on the calibrated high-precision sequence layers. And generating a high-precision sequence trellis model by using the interpolated horizon and the seismic structure trellis model.

The invention has the beneficial effects that: the method comprehensively utilizes the characteristics of high precision of the underground layering sequence and rich transverse information of the earthquake, fully utilizes the earthquake occurrence as a drive to carry out the constrained interpolation of the higher-precision layering sequence, and further generates a higher-precision layering sequence trellis model. The accuracy of the sequence trellis model is improved, the transverse distribution of the model is more reliable, and the high-accuracy sequence trellis model constructed by the method can provide a more powerful technical support for high-accuracy reservoir prediction characterization in the field of oil and gas exploration through the seismic inversion contrastive analysis using the high-accuracy sequence trellis model as constraint.

Drawings

FIG. 1 is a process flow diagram of the method of the present invention.

Fig. 2 is a comparison of core homing and stratigraphic layering of the present invention.

FIG. 3 is a comparison graph of the calibration of the single well high precision sequence stratigraphic synthesis record of the present invention.

FIG. 4 is a well tie sequence stratigraphic comparison and seismic geological interpretation profile of the present invention.

FIG. 5 is a seismic structure tensor analysis and seismic direction field profile of the present invention.

FIG. 6 is a high precision sequence horizon seismic-driven interpolated horizon map of the present invention.

FIG. 7 is a seismic structure trellis-constrained seismic-driven high-precision horizon interpolation profile of the present invention.

FIG. 8 is a comparison graph of inversion under the constraint of the seismic-driven high-precision sequence trellis and the common seismic stratigraphic trellis model.

Detailed Description

In order to more clearly understand the technical features, objects, and effects of the present invention, embodiments of the present invention will now be described with reference to the accompanying drawings.

In this embodiment, as shown in a technical flow chart of fig. 1, a method for constructing a high-precision sequence grid model based on well-seismic combination includes the following steps:

step 1, performing high-precision sequence stratigraphic division and comparison through well logging and core data analysis, constructing a well-connected sequence stratigraphic comparison section, and providing a sequence comparison scheme for seismic interpretation;

and 2, combining the logging data and the seismic data to calibrate and compare the high-precision sequence stratigraphic synthesis seismic records. On the well-connected seismic section, identifying and interpretably tracking the sequence layering according to the corresponding relation between the sequence layering and the seismic event, and carrying out interpretation and adjustment of sequence layering comparison by taking the transverse spread of the seismic section as a basis;

and 3, calculating a seismic direction field on the basis of the seismic structure guiding filtering. And identifying interpretable sequence layers on the seismic profile, and performing horizon interpretation of the corresponding sequence layers under the driving of a seismic direction field to construct a structural grid model.

And 4, taking the seismic structure grid model as top and bottom and interlayer constraints, combining a seismic direction field as a drive, and carrying out transverse interpolation tracking on the calibrated high-precision sequence layers. And generating a high-precision sequence trellis model by using the interpolated horizon and the seismic structure trellis model.

The specific implementation scheme of the seismic-driven high-precision sequence grid model construction method is as follows:

1. logging, core high-precision sequence stratigraphy

And (3) performing high-precision sequence stratigraphic division and comparison by combining rock core description with rock electrical characteristic analysis, wherein the comparison result is shown in figure 2, constructing a sequence comparison section of the well-connected sequence stratigraphic, providing a sequence comparison scheme for seismic interpretation, and the sequence comparison and seismic geological interpretation section are shown in figure 4.

2. Well-to-seismic combined high-precision sequence stratigraphic division comparison

And (3) combining the logging data and the seismic data to perform high-precision layer sequence stratum synthetic seismic record calibration comparison, wherein the comparison result is shown in figure 3. On the well-connected seismic section, according to the corresponding relation between the sequence layering and the seismic event, the interpretable tracing sequence layering is identified, and the interpretation and adjustment of the sequence layering comparison are carried out according to the transverse spread of the seismic section.

3. Seismic tectonic guided filtering and seismic sequence stratigraphic interpretation

The seismic direction field is calculated using a structure tensor method. The structure tensor is a matrix derived from the directional gradient of the function, which contains the principal directions of the gradient in the neighborhood around a point and measures the continuity in those directions.

In the formula, GST represents a structure tensor matrix; g_xRepresenting gradient values in the x-direction; g_yRepresents the gradient values in the y-direction, as shown in FIG. 5; g_xyRepresents G_xAnd G_yThe product of (a). And solving the characteristic value of the GST matrix and the characteristic vector corresponding to the first characteristic value to calculate the corresponding seismic direction field. Respectively by calculatingAnd solving the direction fields in the Line and Trace directions, namely driving the tracking interpolation of the horizon at the point in step 4.

Under the drive of a direction field, an accurate guiding position is calculated point by point, and the real inclination angle guiding is realized. And (3) applying an edge-preserving filtering algorithm or a fault enhancement algorithm of the formula (2) to realize the construction of the guiding filtering. After the structure-oriented filtering, the random noise irrelevant to the structure in the seismic data is suppressed, and then the calculation of the direction field is carried out, so that the structure information contained in the seismic data can be better highlighted.

In the formula, Δ c represents the filtering weight of the horizon at each guide point and the current filtering center point, n represents the filtering radius, κ (Δ t) represents the distance between the guide point and the filtering center point, and t (Δ t) represents the difference between the horizon at the guide point and the filtering center point.

And (3) on the basis of obtaining the seismic direction field and the structural sequence which is identified in the step (2) and can be continuously tracked, performing horizon interpretation of corresponding sequence layers under the driving of the seismic direction field, and constructing a structural grid model.

4. Seismic-driven high-precision sequence layered horizon interpolation and high-precision sequence grid model construction

The seismic structure grid model is used as top and bottom and interlayer constraints, a seismic direction field is used as a drive, the characteristic that seismic information is rich in transverse information is fully utilized, a constraint condition formula (3) which takes seismic data transverse variation trend (information such as amplitude, phase, frequency and the like) as constraint conditions is constructed, transverse interpolation tracking is carried out on calibrated high-precision sequence layers, in the process, the global sequence layer comparison and the minimum integral error of the tracked and interpolated layer are fully considered as the optimal conditions, the layer tracking interpolation of the high-precision sequence layers is completed, and the layer tracking interpolation result shown in the figure 6 and the figure 7 is obtained. And generating a high-precision sequence trellis model by using the interpolated horizon and the seismic structure trellis model.

Wherein Δ_i＝T_i-H₀I, T is the depth of the well in time (converted from the velocity field) in stratification, H₀To constrain the time of the horizon, D is the distance between the predicted point and the well, λ₁As a phase weight factor, λ₂As an amplitude weighting factor, λ₃Is a frequency weighting factor, S is the phase of the predicted point,is the average phase value. A is the amplitude of the predicted point and,is the average phase value, F is the frequency of the predicted point,is the average frequency value.

According to the invention, inversion comparison is carried out under the constraint of the earthquake-driven high-precision sequence grid and a common earthquake stratum grid model, the comparison result is shown in FIG. 8, the left graph is a section and plane distribution of the inversion result constrained by the high-precision sequence grid model, the right graph is a section and plane distribution of the inversion result constrained by the common earthquake stratum grid model, and the comparison shows that the distribution consistency of sand bodies among wells is better, and the plane distribution of the sand bodies is more in accordance with the geological deposition rule.

The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.