Method for quantitatively predicting reservoir thickness interfered by overlying sandstone with unequal thickness

Abstract

The present application relates to a reservoir thickness prediction method for complex and subtle continental reservoirs, in particular to a reservoir thickness quantitative prediction method for interference of overlying unequal-thickness sandstone. The method comprises the following steps: real drilling analysis, including thickness, formation velocity and density of overlying unequal-thickness sandstone and target reservoir; establishing a target reservoir seismic forward model, analyzing seismic interference characteristics of overlying unequal-thickness sandstone on the reservoir, and selecting a suitable formation slice position; determining the dominant tuning frequency of the target reservoir; seismic frequency division processing, extracting the best formation slice of the dominant tuning frequency; establishing a frequency division slice fusion formula for different thickness ranges; obtaining the target reservoir plane prediction map by using the obtained formula, superimposing different thickness prediction results, and realizing quantitative prediction of reservoir thickness. The method significantly improves the prediction accuracy, further enriches the reservoir thickness prediction technology for complex and subtle continental reservoirs, and has good application effect and promotion prospect.

Description

TECHNICAL FIELD

[0001] The present application relates to a reservoir thickness prediction method for complex and subtle continental reservoirs, in particular to a reservoir thickness quantitative prediction method for the interference of overlying unequal-thickness sandstone. BACKGROUND

[0002] In complex and subtle continental reservoirs, the lithology is complex and variable, and the sedimentary rhythm is diverse. The target reservoir is affected by the interference effect of the overlying strata, and the seismic reflection characteristics of the target reservoir are complex, which often causes difficulties in reservoir seismic prediction. Especially in the case of overlying unequal-thickness sandstone, the interference effect on the target reservoir is unstable in the whole area, sometimes present and sometimes absent. The existing technology has poor applicability and low prediction accuracy. Therefore, it is necessary to study a technical method for eliminating the interference effect of overlying unequal-thickness sandstone and realizing quantitative prediction of the thickness of the target reservoir.

[0003] At present, in many studies on reservoir thickness prediction, most of the literature has noticed the tuning interference effect of thin reservoirs themselves, and has better predicted the distribution range of thin sandstone by applying different tuning frequency strata slices or strata slice combinations. For example, Chinese patent application CN106443781A discloses a method for predicting the distribution of thin sand bodies, and Chinese patent CN105319585B discloses a method for identifying oil and gas reservoirs by using thin layer interference amplitude recovery. However, the geological conditions suitable for this method are relatively simple, and the interference effect of overlying unequal-thickness sandstone on thin layers and the influence on the prediction of thin layer thickness are not considered.

[0004] At the same time, in the past research on the interference effect of overlying strata on target strata, most of the literature has noticed the strong reflection shielding effect of the overlying strata. When the thickness of the overlying shielding layer changes relatively stably in the horizontal direction, wavelet decomposition and reconstruction technology is used, such as "High-precision prediction of thin reservoirs under strong shielding: Taking Sanzhai Sag in Songliao Basin as an example" (Gu Wen, 2017); when the thickness of the overlying shielding layer changes greatly in the horizontal direction, "Nail-type" wavelet + compressed sensing technology is used, such as "Research and application of removing strong shielding based on compressed sensing technology" (Zhang Yunyin, 2019). However, in the above methods, the overlying shielding layer produces relatively stable strong seismic reflection in the whole area, and the interference effect on the target reservoir is stable in the whole area. However, for the case of overlying unequal-thickness sandstone, the interference effect on the target reservoir is unstable in the whole area, sometimes present and sometimes absent, which leads to poor applicability and low prediction accuracy of the above technology.

[0005] Therefore, it is currently necessary to provide a reservoir thickness quantitative prediction method suitable for the interference of overlying unequal-thickness sandstone and improve the prediction accuracy. SUMMARY

[0006] The application aims to provide a reservoir thickness quantitative prediction method for overlying unequal-thickness sandstone interference, which can eliminate interference and realize accurate quantitative prediction of target reservoir thickness under the condition of overlying unequal-thickness sandstone.

[0007] The application aims to provide a reservoir thickness quantitative prediction method for overlying unequal-thickness sandstone interference, which can eliminate interference and realize accurate quantitative prediction of target reservoir thickness under the condition of overlying unequal-thickness sandstone.

[0008] The application aims to provide a reservoir thickness quantitative prediction method for overlying unequal-thickness sandstone interference, which can eliminate interference and realize accurate quantitative prediction of target reservoir thickness under the condition of overlying unequal-thickness sandstone.

[0009] Step 1, analysis of actual drilling wells, including thicknesses of overlying unequal-thickness sandstone and target reservoir, formation velocities and densities; establishment of a seismic forward modeling model of the target reservoir, analysis of seismic interference characteristics of the overlying unequal-thickness sandstone on the reservoir, and selection of a suitable formation slice position;

[0010] Step 2, determination of the dominant tuning frequency of the target reservoir;

[0011] Step 3, seismic frequency division processing, and extraction of the best formation slice of the dominant tuning frequency;

[0012] Step 4, establishment of a frequency division slice fusion formula for different thickness ranges:

[0013] Step 5, obtaining of a planar prediction map of the target reservoir with different thicknesses by using the formula obtained in step 4, superposition of prediction results of different thicknesses, and realization of quantitative prediction of the reservoir thickness.

[0014] The application aims to provide a reservoir thickness quantitative prediction method for overlying unequal-thickness sandstone interference, which can eliminate interference and realize accurate quantitative prediction of target reservoir thickness under the condition of overlying unequal-thickness sandstone.

[0015] In step 1, the relationship between the target reservoir thickness and the amplitude variation of the formation slice at different positions is quantitatively analyzed by wave equation seismic forward modeling simulation, and the wave peak or wave trough of the seismic reflection of the target reservoir is selected as the suitable formation slice position.

[0016] Further preferably, when the thickness of the overlying sandstone is large, the wave peak position of the seismic reflection of the target reservoir is selected as the suitable formation slice position; and when the thickness of the overlying sandstone is small, the wave trough position of the seismic reflection of the target reservoir is selected as the suitable formation slice position.

[0017] In step 2,

[0018] f=v / lambda=v / 4h

[0019] Wherein, v is the velocity of the longitudinal wave, f is the main frequency of the seismic data, lambda is the wavelength of the seismic longitudinal wave, and h is the tuning thickness; when the thickness of the sandstone is equal to one-fourth of the wavelength, h is equal to the thickness of the target sand body, and the calculated f is the dominant tuning frequency of the target reservoir.

[0020] In step 3, any time-frequency analysis algorithm such as Fourier transform, wavelet transform, S transform, generalized S transform can be selected to perform frequency division processing on the seismic data to generate an advantageous tuning frequency body, and to extract the amplitude attribute of the target sand body stratal slice to obtain the best stratal slice under the advantageous tuning frequency;

[0021] Further preferably, the generalized S transform algorithm is selected for time-frequency analysis, and the formula is as follows:

[0022]

[0023] Wherein, t represents time; f represents frequency; τ is the position of the Gaussian window on the time axis, and the time window width changes with the change of frequency, and the reciprocal of the frequency determines the size of the time window; p is a parameter for controlling the width of the Gaussian window, and the value range is 0-1, and the smaller the p value, the higher the frequency resolution; x(t) represents the time domain input signal; exp is the natural logarithm function; π is the circular constant; i represents the imaginary unit; and dt represents the integral function. The frequency division algorithm of the generalized S transform can conveniently and effectively divide the seismic data into several frequency bands, has high time-frequency resolution, and can effectively avoid frequency leakage between frequency bands; at the same time, the lossless and reversible nature of the generalized S transform can basically realize error-free reconstruction, meeting the requirements of actual seismic data processing.

[0024] In step 4, the actual drilled well reservoirs are respectively correlated with the respective advantageous tuning frequency stratal slice attributes according to different thickness ranges, the frequency division stratal slice with a higher correlation coefficient is selected for weighted linear regression, and the quantitative characterization formula of the frequency division stratal slice in different thickness ranges is established.

[0025] Further preferably, the target reservoir peak and trough slice amplitudes in the advantageous frequency range are selected, different thicknesses of the actual drilled well reservoirs are respectively statistically fitted with the frequency division slices, the frequency division slice with a higher correlation coefficient is selected for weighted linear regression, and the quantitative characterization formula of the frequency division stratal slice in different thickness ranges is established:

[0026] Thickness=A1k1+A2k2+A3k3+……+A n k n +C

[0027] Wherein, Thickness is the predicted thickness of the thin sandstone; A is the position of the best stratal slice of the advantageous tuning frequency (target reservoir peak and trough slice amplitude), A1 is the first stratal slice combination of the advantageous frequency, An is the n-th stratal slice combination of the advantageous frequency; k1 is the weighted coefficient value of the first stratal slice combination, k n is the weighted coefficient value of the n-th stratal slice combination; and C is a weighted constant. According to the quantitative formula, the thickness of the reservoir interfered by the overlying unequal thickness sandstone can be quantitatively predicted.

[0028] Beneficial effects:

[0029] Based on seismic forward modeling analysis, this invention optimizes the location of stratigraphic slices to eliminate the interference effects of overlying sandstone of unequal thickness. Under the constraint of actual drilling thickness, a quantitative formula for reservoir thickness is established, significantly improving prediction accuracy. This invention enables quantitative prediction of target reservoir thickness, further enriching the reservoir thickness prediction technology for complex and concealed continental oil reservoirs, and has good application effects and promising prospects for wider application. Attached Figure Description

[0030] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0031] Figure 1 This is a flowchart of the method for quantitative prediction of reservoir thickness based on the interference of overlying sandstone of unequal thickness as described in Embodiment 1 of the present invention;

[0032] Figure 2 This is a seismic forward model diagram of the overlying sandstone of varying thickness and the target reservoir;

[0033] Figure 3 It is a cross-sectional analysis diagram of the target reservoir thickness and the amplitude of the target reservoir peak position slice;

[0034] Figure 4 It is a cross-sectional analysis diagram of the target reservoir thickness and the amplitude of the slice at the target reservoir trough location;

[0035] Figure 5 This is a 40Hz well seismic frequency division slice of sandstone thickness prediction map;

[0036] Figure 6 This is a 50Hz well seismic frequency division slice of sandstone thickness prediction map;

[0037] Figure 7 This is a reservoir thickness prediction map after merging two types of frequency-division stratigraphic slices. Detailed Implementation

[0038] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0039] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, and / or combinations thereof.

[0040] To enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be described in detail below with reference to specific embodiments.

[0041] Example 1

[0042] like Figure 1 As shown, the method for quantitatively predicting reservoir thickness in the context of interference from overlying sandstone of unequal thickness includes the following steps:

[0043] Step 1: Through actual drilling analysis, the rock physical characteristics such as the thickness, velocity, and density of the overlying sandstone of varying thickness and the target reservoir are determined, and a seismic forward model of the overlying sandstone reservoir is established. In this embodiment, the lithological combination type is mainly the overlying sandstone of varying thickness and the target sandstone reservoir.

[0044] Statistical analysis shows that the thickness of the overlying sandstone of unequal thickness ranges from 4 to 19 m, with an average of 10 m and a velocity of 3500 m / s; the thickness of the target sandstone reservoir ranges from 3 to 27 m, with an average of 15 m and a velocity of 3400-3800 m / s. Through wave equation seismic forward modeling, the seismic response characteristics when the overlying sandstone of unequal thickness begins to interfere with the target reservoir were determined. The relationship between the target reservoir thickness and the amplitude variation of stratigraphic slices at different locations was quantitatively analyzed. Based on the principle of avoiding reflections from the overlying sandstone of unequal thickness, the stratigraphic slice locations that best reflect the changes in the target reservoir thickness were selected: the peak and trough amplitudes of the target reservoir's seismic reflections. Specifically, when the overlying sandstone thickness is large, the interference from the overlying sandstone is small, and the peak position of the target reservoir's seismic reflection is preferred; when the overlying sandstone thickness is small, the interference from the overlying sandstone is large, and the trough position of the target reservoir's seismic reflection is preferred. Ultimately, through the above statistical analysis, the goal of eliminating the interference effect of the overlying sandstone of unequal thickness was achieved. The process proceeds to step 2.

[0045] Step 2: Based on the thickness tuning principle, assuming the formation velocity is fixed, when the sandstone thickness is equal to one-quarter wavelength, calculate the corresponding frequency using the seismic P-wave propagation velocity formula:

[0046] f = v / λ = v / 4h

[0047] Wherein, v is the longitudinal wave velocity, f is the main frequency of seismic data, λ is the seismic longitudinal wave length, h is the tuning thickness. Let h equal the target sand thickness, using the above formula to calculate f is the dominant tuning frequency of the target sand.

[0048] Through the statistical analysis of the actual drilling wells, it is concluded that the thickness of the target sandstone reservoir mainly distributes between 10-20m and 21-30m. According to the statistical sandstone velocity of 3500m / s, it is calculated that the dominant tuning frequency of the target reservoir is mainly 40Hz and 50Hz. The process enters step 3.

[0049] Step 3, when the seismic data is frequency division processed, the more accurate generalized S transform algorithm is preferred for time-frequency analysis. According to the following formula, the calculation is carried out:

[0050]

[0051] Wherein, t represents time; f represents frequency; τ is the position of the Gaussian window on the time axis, the time window width changes with the change of frequency, and the reciprocal of frequency determines the size of the time window; p controls the parameter of the Gaussian window width, the value range is 0-1, the smaller the p value, the higher the frequency resolution; x(t) represents the time domain input signal; exp is the natural logarithm function; π is the circular constant; i represents the imaginary unit; dt represents the integral function.

[0052] In this embodiment, the seismic data is divided into 35Hz, 40Hz, 45Hz, 50Hz and 55Hz, a total of 5 frequency division data bodies, reflecting the information of sand bodies of different thicknesses. And further, the best stratum slice of the target reservoir of 40Hz and 50Hz two dominant tuning frequencies is extracted: the peak and trough slice amplitude of the target reservoir. The process enters step 4.

[0053] Step 4, the peak and trough slice amplitudes of the target reservoir of 40Hz dominant tuning frequency of the 21-30m sand thickness of 40 actual drilling wells are selected for weighted linear regression, and the frequency division stratum slice fusion formula of the target reservoir thickness (21-30m) is established:

[0054] Thickness1=0.11*A1+0.0047*A2+13.3

[0055] Wherein, Thickness1 is the predicted target reservoir thickness (21-30m); A is the peak and trough slice amplitude of the target reservoir of 40Hz dominant tuning frequency, A1 is the peak slice amplitude of the target reservoir of 40Hz dominant frequency, and A2 is the trough slice amplitude of the target reservoir of 40Hz dominant frequency.

[0056] The 10-20m sand body thickness of 40 drilled wells and the target reservoir wave peak and trough slice amplitude of 50Hz dominant tuning frequency are selected to perform weighted linear regression, and a fusion formula of the target reservoir thickness (10-20m) of the frequency division stratum slice is established:

[0057] Thickness2=0.16*A1+0.19*A2+15.3

[0058] Thickness2 is the predicted target reservoir thickness (10-20m); A is the target reservoir wave peak and trough slice amplitude of 50Hz dominant tuning frequency, A1 is the target reservoir wave peak slice amplitude of 50Hz dominant frequency, and A2 is the target reservoir wave trough slice amplitude of 50Hz dominant frequency. The process goes to step 5.

[0059] Step 5, the established fusion formula of the two types of thickness range (10-20m, 21-30m) frequency division stratum slice is applied, the reservoir plane prediction map of 10-20m and 21-30m thickness is obtained respectively, and the prediction results of different thicknesses are superimposed, realizing the quantitative prediction of reservoir thickness under the interference of overlying unequal thickness sandstone.

[0060] Figure 2 is the seismic forward model diagram of the overlying unequal thickness sandstone and the target reservoir. Figure 3 is the cross plot analysis diagram of the target reservoir thickness and the target reservoir wave peak position slice amplitude, which shows that when the target reservoir thickness is between 21-30m, the target reservoir wave peak position slice amplitude is proportional to the target reservoir thickness, and has a good linear relationship. Figure 4 is the cross plot analysis diagram of the target reservoir thickness and the target reservoir wave trough position slice amplitude, which shows that when the target reservoir thickness is between 10-20m, the target reservoir wave peak position slice amplitude is proportional to the target reservoir thickness, and has a good linear relationship. Figure 5 is the 40Hz well-seismic frequency division slice sandstone thickness seismic prediction map, and the analysis shows that the prediction result is highly consistent with the 21-30m sand body thickness of the drilled well, and the average thickness prediction error is 9.0%. Figure 6 is the 50Hz well-seismic frequency division slice sandstone thickness seismic prediction map, and the analysis shows that the prediction result is highly consistent with the 10-20m sand body thickness of the drilled well, and the average thickness prediction error is 7.2%. Figure 7 is the reservoir thickness prediction map after the fusion of the two types of frequency division stratum slice, and the analysis shows that the prediction result is highly consistent with the sand body thickness of the drilled well, the average thickness prediction error of the participating well is 7.7%, the average thickness prediction error of the verification well is 19.1%, and the quantitative prediction of reservoir thickness under the interference of overlying unequal thickness sandstone is realized.

[0061] The expression method used in the specification is the customary method for those skilled in the art, which is well known to those skilled in the art and will not be explained in more detail.

[0062] The embodiments of the present application are explained in detail as above, and are not intended to limit the scope of the present application. It will be obvious to those skilled in the art that various changes and modifications can be made in the embodiments of the present application without departing from the spirit and scope of the present application. Therefore, such changes and modifications are intended to be included within the scope of the present application.

Claims

1. A method for quantitative prediction of reservoir thickness overlying an interference of sandstones of unequal thicknesses, characterized in that, The method comprises the following steps: Step 1, real well analysis, including the thickness, formation velocity and density of overlying unequal-thickness sandstone and target reservoir; a seismic forward model of the target reservoir is established, the seismic interference characteristics of the overlying unequal-thickness sandstone on the reservoir are analyzed, and a suitable formation slice position is selected; Step 2, determining the dominant tuning frequency of the target reservoir; Step 3, seismic frequency division processing, extracting the best formation slice under the dominant tuning frequency; Step 4, establishing a frequency division slice fusion formula in different thickness ranges; Step 5, obtaining the target reservoir plane prediction map by using the formula obtained in step 4, superimposing the prediction results of different thicknesses, and realizing quantitative prediction of the reservoir thickness; In step 1, the quantitative relationship between the target reservoir thickness and the amplitude change of the formation slice at different positions is analyzed by wave equation seismic forward simulation, and the wave peak or wave trough of the target reservoir seismic reflection is selected as the suitable formation slice position; When the thickness of the overlying sandstone is large, the wave peak position of the target reservoir seismic reflection is selected as the suitable formation slice position; when the thickness of the overlying sandstone is small, the wave trough position of the target reservoir seismic reflection is selected as the suitable formation slice position; In step 2, f=v / λ=v / 4h Wherein, v is the velocity of the longitudinal wave, f is the main frequency of the seismic data, λ is the wavelength of the seismic longitudinal wave, and h is the tuning thickness; when the sandstone thickness value is equal to one-fourth of the wavelength, h is equal to the thickness of the target sand body, and the calculated f is the dominant tuning frequency of the target reservoir; In step 4, the reservoirs of real wells are analyzed according to different thickness ranges, and well-seismic correlation analysis is performed on the respective dominant tuning frequency formation slice attributes, the frequency division formation slice with a higher correlation coefficient is selected for weighted linear regression, and the quantitative characterization formula of the frequency division formation slice in different thickness ranges is established; The wave peak and wave trough slice amplitudes of the target reservoir in the dominant frequency range are selected, different thicknesses of the real well reservoirs are respectively fitted with the frequency division slice, the frequency division slice with a higher correlation coefficient is selected for weighted linear regression, and the quantitative characterization formula of the frequency division formation slice in different thickness ranges is established: Thickness = A1k1+ A2k2+ A3k3+... + A n k n + C Wherein, Thickness is the predicted thin sandstone thickness; A is the best formation slice position of the dominant tuning frequency, i.e. the target reservoir wave peak and trough slice amplitude, A1 is the first dominant frequency formation slice combination, An is the nth dominant frequency formation slice combination; k1 is the weighting coefficient value of the first formation slice combination, n k is the weighting coefficient value of the nth formation slice combination; C is the weighting constant.

2. The method for quantitative prediction of reservoir thickness overlying the interference of unequal thick sandstone according to claim 1, characterized in that, In step 3, any time-frequency analysis algorithm of Fourier transform, wavelet transform, S transform, and generalized S transform is selected to perform frequency division processing on the seismic data, generate the dominant tuning frequency body, and extract the amplitude attribute of the target sand body formation slice to obtain the best formation slice under the dominant tuning frequency; The generalized S transform algorithm is selected for time-frequency analysis, and the formula is as follows: Wherein, t represents time; f represents frequency; τ is the position of the Gaussian window on the time axis, the time window width changes with the frequency, and the inverse of the frequency determines the size of the time window; p controls the parameter of the Gaussian window width, the value range is 0-1, the smaller the p value, the higher the frequency resolution; x(t) represents the time domain input signal; exp is the natural logarithm function; π is the circular constant; i represents the imaginary unit; dt represents the integral function.

Images