An inversion method, system and storage medium based on karst facies control

Abstract

The present invention discloses an inversion method, system and storage medium based on karst facies control, belonging to the technical field of karst exploration. The method includes S1: performing spectral analysis on the seismic data of the work area to be measured, performing frequency division to generate seismic data volumes of different frequency bands; S2: establishing a sequence body using well logging data; performing qualitative and quantitative analysis on the karst facies in the work area to be measured to obtain the distribution range of karst facies; establishing a facies control model using the distribution range of karst facies, the sequence body and well logging data; S3: performing AVF characteristic frequency inversion to obtain the non-linear mapping relationship between the well logging wave impedance curve and the seismic waveform, and inputting the non-linear mapping relationship into the facies control model to obtain the inversion result. By jointly modeling using the distribution range of karst facies, the sequence body and well logging data, the morphological characteristics of karst fissure-cave bodies can be effectively controlled. Through joint inversion of multiple wells, false bead-like reflections can be effectively identified, providing a basis for well location deployment.

Description

Technical Field

[0001] The present invention relates to the technical field of karst exploration, and in particular to an inversion method, system and storage medium based on karst phase control. Background Art

[0002] Carbonate karst fracture-cave reservoirs are an important type of reservoir, widely distributed in the Tarim Basin and Sichuan Basin in Xinjiang, my country. However, due to their strong heterogeneity and complex internal structural characteristics, it is difficult to accurately predict reservoirs and semi-quantitatively or quantitatively characterize their vertical and horizontal distribution characteristics.

[0003] For the prediction of carbonate karst fracture-cave reservoirs, a technical combination of "coherence + trend surface + amplitude change rate" has been formed, which has achieved great success in the application of Tahe Oilfield in Tarim, Xinjiang, and successfully upgraded the production of Tahe Oilfield. The core of the technology is to determine the exploration target based on the idea of dominant terrain + dominant amplitude of trend surface superimposed amplitude change rate on the premise of fully understanding the control of strike-slip faults. However, the disadvantage of this method is that the vertical and horizontal scales are too coarse, it is easy to miss some key targets, and it is impossible to semi-quantitatively analyze the reservoir. It is not suitable for the refined exploration stage.

[0004] With the continuous improvement of the exploration level, the prediction of karst fracture-cave reservoirs has changed to a multi-attribute fusion technology of "amplitude gradient attribute + ant body + chaotic structure reflection" to characterize the karst fracture-cave structure in three dimensions, but these attributes can only be described qualitatively. To quantitatively or semi-quantitatively characterize the reservoir, the commonly used method is wave impedance inversion technology. However, the current inversion methods are mostly aimed at layered media, not massive media. In addition, conventional wave impedance inversion can only reflect cave-type reservoirs, and can only determine the top surface of the cave, and the vertical and horizontal resolution is not high, which is slightly weak for the characterization of the interior of the cave and the characterization of fault and fracture-type reservoirs. Many people in the industry have realized this and have successively adopted waveform indication inversion and sedimentary phase control inversion methods.

[0005] The former solves the problem of fracture-cavity identification to a certain extent, but the identification type is relatively single and it is easy to have false fractures and caves. The latter is mainly based on karst fracture-cavity reservoirs with sedimentary facies as the main controlling factor, and has limitations in predicting fault-controlled karst fracture-cavity reservoirs.

[0006] The traditional seismic inversion process generally includes three parts: wavelet extraction, low-frequency model establishment, and inversion. The karst phase-controlled reservoir inversion technology optimizes and improves the low-frequency model establishment, incorporates geological knowledge such as sequence and karst phase into the initial model establishment process, realizes the establishment of karst phase-controlled model, and involves the model in seismic inversion. At present, there are relatively few studies on reservoir prediction using karst phase-controlled inversion, and it is still in the exploratory stage overall. Summary of the invention

[0007] The object of the present invention is to overcome the deficiencies of the prior art and provide an inversion method, system and storage medium based on karst facies control.

[0008] The object of the present invention is achieved by the following technical solutions: In the first aspect of the present invention, there is provided an inversion method based on karst facies control, including the following steps:

[0009] S1: Perform spectral analysis on the seismic data of the work area to be measured to obtain the dominant frequency information and the effective frequency band range; at the same time, according to the well point data, analyze the characteristic frequencies of the target layers of the wells with different display levels; perform frequency division based on the characteristic frequencies, dominant frequency information and effective frequency band range to generate seismic data volumes of different frequency bands;

[0010] S2: Conduct sequence stratigraphic division on the drilling data and calibrate and trace it on the seismic section, and establish a sequence body by using the traced horizons; conduct qualitative and quantitative analysis on the karst facies in the work area to be measured to obtain the distribution range of karst facies; establish a facies control model by using the distribution range of karst facies, the sequence body and the drilling data;

[0011] S3: Use the support vector machine algorithm to calculate the AVF relationship between the amplitude and frequency of the seismic data volume at different thicknesses; use the AVF relationship to perform AVF characteristic frequency inversion to obtain the non-linear mapping relationship between the well log wave impedance curve and the seismic waveform, and input the non-linear mapping relationship into the facies control model to obtain the inversion result.

[0012] Preferably, the sequence stratigraphic division includes: dividing the drilling data into a sequence group composed of multiple multi-level sequences, and the sequence group is distributed in parallel and throughout the work area to be measured.

[0013] Preferably, the qualitative and quantitative analysis is to analyze the seismic reflection characteristics of the work area to be measured; use the shape, boundary reflection structure and internal reflection structure of the seismic facies unit to divide the seismic facies types of the work area to be measured into bead-shaped, low-frequency - medium continuity abnormal reflection, chaotic - weak reflection; depict the longitudinal and lateral distribution of the karst facies of the target layer according to the type of seismic facies to obtain the distribution range of karst facies.

[0014] Preferably, the AVF characteristic frequency inversion uses a BP neural network model or a support vector machine algorithm or a particle swarm algorithm.

[0015] Preferably, single-well calibration is used to divide the drilling data into a sequence group composed of four fourth-level sequences.

[0016] In the second aspect of the present invention, there is provided an inversion system based on karst facies control for implementing any of the above-mentioned inversion methods based on karst facies control, including:

[0017] Seismic data volume generation module: It is used to perform spectral analysis on the seismic data of the work area to be measured to obtain the main frequency information and the effective frequency band range; at the same time, according to the well point data, analyze and obtain the characteristic frequencies of the target layers of different display-level wells; perform frequency division based on the characteristic frequencies, main frequency information, and effective frequency band range to generate seismic data volumes of different frequency bands;

[0018] Facies-controlled model establishment module: It is used to carry out sequence stratigraphic division on the drilling data and calibrate and trace it on the seismic section, and establish a sequence body using the traced horizons; qualitatively and quantitatively analyze the karst facies in the work area to be measured to obtain the distribution range of the karst facies; establish a facies-controlled model using the distribution range of the karst facies, sequence body, and drilling data;

[0019] Inversion module: It is used to calculate the AVF relationship between the amplitude and frequency of the seismic data volume at different thicknesses using the support vector machine algorithm; perform AVF characteristic frequency inversion using the AVF relationship to obtain the non-linear mapping relationship between the logging wave impedance curve and the seismic waveform, and input the non-linear mapping relationship into the facies-controlled model to obtain the inversion result.

[0020] The third aspect of the present invention provides: A computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, any one of the above-mentioned inversion methods based on karst facies control is realized.

[0021] The beneficial effects of the present invention are:

[0022] 1) Karst facies-controlled inversion improves the vertical and horizontal resolution of the section and can better depict the characteristics of karst reservoirs. Compared with conventional sparse pulse inversion, it can be seen that the sparse pulse inversion has a relatively rough manifestation form and cannot finely depict the boundaries of karst reservoirs.

[0023] 2) Using the distribution range of karst facies, sequence body, and drilling data jointly for modeling can effectively control the morphological characteristics of karst fracture-vug bodies. Through joint inversion of multiple wells, false bead-like reflections can be effectively identified, providing a basis for well location deployment. Description of the Drawings

[0024] Figure 1 It is a flow chart of an inversion method based on karst facies control;

[0025] Figure 2 It is a schematic diagram of sequence stratigraphic division of the Yingshan Formation - Yijianfang Formation in the Ordovician system in the northern Tarim Basin;

[0026] Figure 3 It is a schematic diagram of seismic sections and colored inversion of three seismic facies;

[0027] Figure 4 It is a schematic diagram of AVF characteristic frequency inversion. Detailed Embodiments

[0028] Next, in combination with the embodiments, the technical solutions of the present invention will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0029] Refer to Figures 1 - 4 , the first aspect of the present invention provides: an inversion method based on karst facies control, including the following steps:

[0030] S1: Perform spectral analysis on the seismic data of the work area to be measured to obtain the main frequency information and the effective frequency band range; at the same time, based on the well point data, analyze and obtain the characteristic frequencies of the target layers of different display-level wells; perform frequency division according to the characteristic frequencies, main frequency information and effective frequency band range to generate seismic data volumes of different frequency bands;

[0031] S2: Perform sequence stratigraphic division on the drilling data and calibrate and track it on the seismic section, and establish a sequence body by using the tracked horizons; perform qualitative and quantitative analysis on the karst facies in the work area to be measured to obtain the distribution range of the karst facies; establish a facies control model by using the distribution range of the karst facies, the sequence body and the drilling data;

[0032] S3: Use the support vector machine algorithm to calculate the AVF relationship between the amplitude and frequency of the seismic data volume at different thicknesses; use the AVF relationship to perform AVF characteristic frequency inversion to obtain the non-linear mapping relationship between the well log wave impedance curve and the seismic waveform, and input the non-linear mapping relationship into the facies control model to obtain the inversion result.

[0033] Before performing qualitative and quantitative analysis, obtain the mean absolute amplitude attribute, AFE attribute, waveform clustering and colored inversion of the karst facies. Then perform qualitative and quantitative analysis on the karst facies to determine the vertical and horizontal distribution characteristics of the karst facies. Use the known wells to divide the fourth-level sequences of the target layers in this area, calibrate the seismic data, and track it throughout the area. Establish a sequence body of the target layer. Use the well curves in the drilling data, the planar distribution of the karst facies in the distribution range of the karst facies and the sequence body to jointly build a model. On the premise of determining the characteristic frequencies, use the model to constrain the AVF characteristic frequency inversion. The described drilling data is the drilling data of multiple wells encountering the Yingshan Formation.

[0034] In some embodiments, the sequence stratigraphic division includes: dividing the drilling data into multiple multi-level sequences to form a sequence group, and the sequence group is distributed throughout the work area to be measured in a parallel shape.

[0035] Such as Figure 2As shown in the figure, the work area to be measured in the present invention is selected as the northern Tarim Basin, and the main target layers are the Yijianfang Formation (acoustic impedance code T74) and the Ying Mountain Formation (acoustic impedance code T76) of the Ordovician System. There are 26 wells drilled through the target layers in the work area to be measured, among which 7 wells (S109CH, TP189X, TP276H, TP275H, TY1X, TP193, TP183) encountered the Ying Mountain Formation (T76) during drilling. Through single-well calibration, the Ying Mountain-Yijianfang Formation in this area is divided into four fourth-order sequences (SQ1, SQ2, SQ3, SQ4). The sequence groups are distributed in a parallel pattern and are distributed throughout the work area. The reservoir as a whole presents the characteristics of three layers of caves.

[0036] In some embodiments, the qualitative and quantitative analysis is to analyze the seismic reflection characteristics of the work area to be measured; the seismic facies types of the work area to be measured are divided into beaded, low-frequency-medium continuity abnormal reflection, chaotic-weak reflection by using the shape, boundary reflection structure and internal reflection structure of seismic facies units; the longitudinal and transverse distribution of the karst facies of the target layer is characterized according to the type of seismic facies to obtain the distribution range of the karst facies.

[0037] As Figure 3 shown in the figure, the target layer is the Yijianfang Formation (T74) - Ying Mountain Formation (T76) of the Ordovician System. The beaded type is an isolated beaded type with two black sandwiching one red, its amplitude is greater than the first preset value, the lateral continuity is lower than the second preset value, and the seismic facies frequency is greater than the third preset value; the low-frequency-medium continuity abnormal reflection has continuous strong reflection in the target layer, which is a combination of multiple beaded types or chaotic reflection near the fracture; the chaotic-weak reflection has an amplitude lower than the fourth preset value, a continuity lower than the fifth preset value, and a seismic facies frequency lower than the sixth preset value in the target layer, which is the characteristic of the surrounding rock, with a porosity lower than the seventh preset value and a permeability lower than the eighth preset value. Seismic attributes are a measurement of the geometric, kinematic, dynamic and statistical characteristics of seismic data, and also the response of various stratigraphic information and geological phenomena in seismic data. Therefore, under the guidance of seismic facies quantitative analysis, the karst facies can be better characterized. Through comprehensive analysis of various attributes, Dip, integral energy spectrum and colored inversion, which can best reflect the stratigraphic characteristics of this area, are finally selected for three-frequency linkage characterization to clarify the planar distribution characteristics of the karst facies, and the longitudinal and transverse distribution of the karst facies of the target layer is characterized. The upper and lower layers are affected by karstification in different periods, and the differences between the two layers are relatively large. The upper part is mainly developed with beaded reflections, while the lower part has a large area of abnormal low-frequency reflections, showing different characteristics of karstification in different periods.

[0038] The characteristic frequency is preferably determined mainly through two aspects: 1) the frequency differentiation between wells with hydrocarbon shows and wells without hydrocarbon shows; 2) after frequency division processing of the original seismic data, the response characteristics of different frequencies to the reservoir. The two are combined to select the effective frequency to characterize the reservoir and hydrocarbon-bearing characteristics. By analyzing the frequencies of 26 wells drilled through the target formation in the study area, the frequencies of the wells with shows mainly concentrate between 23 - 28 Hz, with an average frequency of 25 Hz. While 17 wells without shows were statistically analyzed, and the average frequency is around 36 Hz, and the difference between the two is relatively obvious.

[0039] In some embodiments, the AVF characteristic frequency inversion uses a BP neural network model or a support vector machine algorithm or a particle swarm algorithm.

[0040] The method selected for AVF characteristic frequency inversion is the BP neural network. Its advantage is that it can obtain the mapping relationship of data between multiple inputs and outputs through learning, and use this network relationship to obtain the corresponding output results. During the application of this method, the adjustment algorithm and parameters are adjusted, and by checking the correlation between the learning curve and the target curve, the quality of the learning effect is determined, so as to continue to obtain the inversion result, which guides the drilling deployment and reserve calculation work. Method principle: For formations with different thicknesses, their tuning frequencies are different; conversely, using this relationship, the relationship between amplitude and frequency (AVF) at different time thicknesses can be obtained. In other words, for a wedge-shaped model, convolving it with Ricker wavelets of different dominant frequencies to obtain a series of synthetic seismic data volumes, thereby obtaining the tuning curves of amplitude and thickness at different frequencies, and thus obtaining the relationship between amplitude and frequency variation (AVF) at different time thicknesses.

[0041] The relationship between amplitude and frequency (AVF) at different time thicknesses shows an important law: the same formation will show different amplitude characteristics under different dominant frequency wavelets. However, the AVF relationship is very complex and it is difficult to represent it with a display function. The method of non-linear mapping of support vector machine (SVM) is required to find this relationship on the logging and seismic wavelet decomposition profiles, and use the AVF information for inversion.

[0042] The inversion results are as Figure 4 shown. The upper left figure represents the karst fracture-vug reservoir body with low impedance, and the lower left figure represents the dense surrounding rock with high impedance. From the planar characteristics, the lateral resolution is relatively high and it has a high degree of coincidence with the actual drilling. From the profile characteristics, the vertical resolution is also relatively high and it coincides with the characteristics of 5 evaluation wells (ZSPJ_27, ZSPJ_31, ZSPJ_18, ZSPJ_23, ZSPJ_25). It is proved that the AVF characteristic frequency inversion based on karst facies control can better characterize the karst reservoir characteristics. It can be used for the later semi-quantitative to quantitative three-dimensional stereo carving work.

[0043] In some embodiments, single well calibration is used to divide the drilling data into four quadruple-level sequences to form a sequence group.

[0044] Different from the traditional qualitative wave impedance inversion method, the method of the present invention forms a constraint model by characterizing the carbonate karst phase and jointly modeling it with the stratigraphic sequence and drilling data. During the process, the drilling data is used for monitoring, and local micro-adjustments are made in combination with the karst phase distribution characteristics in order to achieve the ideal effect for guiding the well deployment work.

[0045] The second aspect of the present invention provides: an inversion system based on karst phase control, used to implement any of the above-mentioned inversion methods based on karst phase control, comprising:

[0046] Seismic data volume generation module: used to perform spectrum analysis on the seismic data of the work area to be tested, and obtain the main frequency information and effective frequency band range; at the same time, according to the well point data, analyze and obtain the characteristic frequency of the target layer of drilling at different display levels; divide the frequency according to the characteristic frequency, main frequency information and effective frequency band range to generate seismic data volumes of different frequency bands;

[0047] Phase-controlled model building module: used to divide the drilling data into sequence stratigraphic units and calibrate and track them on the seismic profile, and use the tracked layers to establish the sequence body; conduct qualitative and quantitative analysis on the karst phases in the work area to be measured to obtain the distribution range of the karst phase; and use the karst phase distribution range, sequence body and drilling data to establish the phase-controlled model;

[0048] Inversion module: used to calculate the AVF relationship between the amplitude and frequency of seismic data volumes at different thicknesses using the support vector machine algorithm; use the AVF relationship to perform AVF characteristic frequency inversion to obtain the nonlinear mapping relationship between the logging wave impedance curve and the seismic waveform, and input the nonlinear mapping relationship into the phase control model to obtain the inversion result.

[0049] The third aspect of the present invention provides: a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are loaded and executed by a processor, any of the above-mentioned inversion methods based on karst phase control is implemented.

[0050] The above is only a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art do not deviate from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. An inversion method based on karst facies control, characterized in that: It includes the following steps: S1: Conduct spectral analysis on the seismic data of the work area to be measured to obtain the dominant frequency information and the effective frequency band range; at the same time, based on the well point data, analyze and obtain the characteristic frequencies of the target layers of different display-level wells; perform frequency division according to the characteristic frequencies, dominant frequency information, and effective frequency band range to generate seismic data volumes of different frequency bands; S2: Conduct sequence stratigraphic division on the drilling data and calibrate and trace it on the seismic section, and establish a sequence body using the traced horizons; conduct qualitative and quantitative analysis on the karst facies in the work area to be measured to obtain the distribution range of the karst facies; establish a facies-controlled model using the distribution range of the karst facies, the sequence body, and the drilling data; S3: Use the support vector machine algorithm to calculate the AVF relationship between the amplitude and frequency of the seismic data volume at different thicknesses; use the AVF relationship to perform AVF characteristic frequency inversion to obtain the non-linear mapping relationship between the well logging wave impedance curve and the seismic waveform, and input the non-linear mapping relationship into the facies-controlled model to obtain the inversion result; Before conducting qualitative and quantitative analysis, obtain the average absolute amplitude attribute, AFE attribute, waveform clustering, and colored inversion of the karst facies; The above-mentioned qualitative and quantitative analysis is to analyze the seismic reflection characteristics of the work area to be measured; use the shape, boundary reflection structure, and internal reflection structure of the seismic facies unit to divide the seismic facies types of the work area to be measured into bead-shaped, low-frequency-medium continuity abnormal reflection, chaotic-weak reflection; Characterize the vertical and horizontal distribution of the karst facies of the target layer according to the type of seismic facies to obtain the distribution range of the karst facies; The bead-shaped is an isolated bead with two blacks sandwiching one red, its amplitude is greater than the first preset value, its horizontal continuity is lower than the second preset value, and the seismic facies frequency is greater than the third preset value; the target layer of the low-frequency-medium continuity abnormal reflection has continuous strong reflection, which is a combination of multiple bead-shaped or chaotic reflection near the fracture; the target layer of the chaotic-weak reflection has an amplitude lower than the fourth preset value, a continuity lower than the fifth preset value, a seismic facies frequency lower than the sixth preset value, which is the surrounding rock characteristic, with a porosity lower than the seventh preset value and a permeability lower than the eighth preset value.

2. The inversion method based on karst facies control according to claim 1, characterized in that: The above-mentioned sequence stratigraphic division includes: dividing the drilling data into multiple multi-level sequences to form a sequence group, and the sequence group is distributed in parallel and throughout the work area to be measured.

3. The inversion method based on karst facies control according to claim 1, characterized in that: The above-mentioned AVF characteristic frequency inversion uses a BP neural network model or a support vector machine algorithm or a particle swarm algorithm.

4. The inversion method based on karst facies control according to claim 2, wherein: Use single-well calibration to divide the drilling data into a sequence group composed of four fourth-level sequences.

5. An inversion system based on karst facies control, characterized in that: For implementing the inversion method based on karst facies control as described in any one of claims 1-4, it includes: Seismic data volume generation module: used to conduct spectral analysis on the seismic data of the work area to be measured to obtain the dominant frequency information and the effective frequency band range; at the same time, based on the well point data, analyze and obtain the characteristic frequencies of the target layers of different display-level wells; perform frequency division according to the characteristic frequencies, dominant frequency information, and effective frequency band range to generate seismic data volumes of different frequency bands; Phased model building module: used to conduct sequence stratigraphic division on drilling data and calibrate and trace on seismic profiles, and establish a sequence body using the traced horizons; conduct qualitative and quantitative analysis on the karst facies in the area to be measured to obtain the distribution range of karst facies; establish a phased model using the distribution range of karst facies, the sequence body and drilling data; Inversion module: used to calculate the AVF relationship between the amplitude and frequency of the seismic data volume at different thicknesses using the support vector machine algorithm; conduct AVF characteristic frequency inversion using the AVF relationship to obtain the non-linear mapping relationship between the logging wave impedance curve and the seismic waveform, and input the non-linear mapping relationship into the phased model to obtain the inversion result.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are loaded and executed by a processor, the inversion method based on karst facies control described in any one of claims 1-4 is implemented.

Images