Well seismic calibration method based on wave-induced pore fluid dissolution seismic attenuation model

By coarsing the seismic attenuation model and viscoelastic model based on wave-induced pore fluid dissolution, the problem of velocity dispersion error in well seismic matching is solved, and the reservoir characteristics are accurately identified and calibrated, which is suitable for low-pore shales Reservoir.

CN118962809APending Publication Date: 2024-11-15CHINA PETROLEUM & CHEMICAL CORP +1
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202310543911.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-15
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The prior art ignores the physical mechanism behind the velocity dispersion in well-seismic matching, resulting in calibration errors, and the parameters of the rock physical model are difficult to obtain and lack physical significance, making it difficult to accurately predict reservoir characteristics.

Method used

The seismic attenuation model based on wave-induced pore fluid dissolution is used to calculate the attenuation curve and mass factor through rock physical modeling and coarseness of viscoelastic model, which is used for well seismic data calibration and match the combined seismic data and field data.

Benefits of technology

Accurate calibration of well seismic data is achieved, parameter complexity is reduced, and is suitable for low-pore and low-permeability shale reservoirs, improving the accuracy of reservoir characteristics recognition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118962809A_ABST
    Figure CN118962809A_ABST
Patent Text Reader

Abstract

The invention provides a well seismic calibration method based on a wave-induced pore fluid dissolution seismic attenuation model, and the method comprises the steps: 1, carrying out the rock physical modeling according to the geological, geochemical and rock core data; 2, an attenuation curve is calculated according to the rock physical model, and the coarsened attenuation curve is obtained through coarsening of the viscoelastic model; and step 3, well seismic data calibration is carried out, and reasonable matching between the synthetic seismic data and the field data is found. According to the well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model, velocity dispersion is accurately predicted by using a proper rock physical model, and the difference between two kinds of data is calibrated by coarsening the rock physical model.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of oilfield exploration, and in particular to a well seismic calibration method based on a wave-induced pore fluid dissolution seismic attenuation model. Background Art

[0002] Well logging data and seismic data are the two most basic raw data in the process of oilfield exploration and development, and play an important role in the detailed description of reservoirs such as reservoir prediction. The inconsistent reservoir information reflected leads to multiple solutions for reservoir prediction and description. How to match well and seismic data has become an indispensable and important link in predicting reservoir lithology, physical parameters and oil and gas potential in oil and gas exploration and development research. There are three main methods for velocity matching and calibration of well and seismic data:

[0003] The first is the Backus effective average method, which uses effective medium theory and layering method to achieve high-frequency to low-frequency velocity correction. (Sams and Williamson, 1994) discussed in depth the impact of Backus effective average on elastic scattering characteristics in the medium; (Cao Danping, 2015) used Backus effective average to conduct an in-depth study on the upscaling of logging data. The velocities of P and S waves can draw rock physics analysis conclusions comparable to the seismic scale.

[0004] Second, the multi-resolution analysis method stretches the signals between adjacent events of the acoustic logging synthetic seismic record to align them with the corresponding events on the side hole record. (Dong Enqing, 1999) The multi-resolution analysis technology is used to resample the logging data and match the synthetic record with the seismic record, with high matching accuracy.

[0005] The third is the rock physics model calibration method, which combines velocity dispersion theory and absorption attenuation mechanism to achieve velocity calibration at different scales. Based on the resonant Q model (Yang Wencai, 1987), many researchers (Ba Jing et al., 2012) measured the acoustic logging velocity of different target layers respectively. Dispersion calibration was carried out to improve the well-seismic matching effect. (H.Wang, 2011) Combined with the DEM model and the microscopic dispersion theory model, the logging frequency band velocity was inferred to the seismic frequency band, and good seismic matching and inversion calculation were performed, which achieved good results in carbonate reservoirs.

[0006] In practical applications, the Backus effective average method is simple and effective, but does not consider the attenuation effect. Multi-resolution analysis is highly automated, but still requires accurate synthetic seismic records and formation correlation. The rock physics model calibration method can directly match data of different scales. In order to combine well logging and seismic data, it is necessary to select a suitable rock physics model for dispersion calibration to improve the accuracy of well-seismic matching.

[0007] Wave-induced fluid flow (WIFF) is considered to be the main cause of seismic wave diffusion and attenuation in fluid-saturated porous media (Pride et al., 2004; Gurevich et al., 2010). In many theories, meso-heterogeneity and micro-heterogeneity are considered to be the main mechanisms leading to WIFF. In addition, in most rocks, the coexistence of meso-heterogeneity and micro-heterogeneity can lead to significant changes in fast P-wave velocities, which means that the effects of both mechanisms on dispersion and attenuation need to be considered simultaneously. The Squirt micro-model mainly uses the ejection mechanism of solid / fluid interaction to estimate velocity dispersion and attenuation in fully saturated rocks (Jack Dvorkin et al., 1995). Pride et al. proposed a dual-porosity theory to explain the corresponding seismic waves in water- and gas-saturated porous media, which can explain the attenuation of 10-2-10-1 in the seismic frequency band. The White model describes the complex modulus of partially saturated spherical gas-encapsulated media and layered media composed of two inhomogeneous porous media (White, 1975; White, 1975). Johnson generalized the arbitrary shape of gas blocks (Johnson, 2001). In recent years, under the conditions of saturated fluid and bubble phases in rocks, it is believed that there will be obvious attenuation at the low-frequency end of seismic waves, and the corresponding mechanism of seismic wave induced pore fluid dissolution (WIGED, used to explain the obvious attenuation of seismic waves (Tisato et al., 2012; Tisato et al., 2013; Tisato et al., 2015), these models reflect the mesoscopic loss mechanism. Chapman measured the attenuation of two Berea sandstone samples, and the results showed that compared with the WIFF mechanism, the WIGED mechanism can be closer to the measured attenuation data (Chapman, 2017). In addition, as a seismic wave attenuation mechanism, WIGED can describe micron pores. It is more suitable for low-porosity and low-permeability shale reservoirs.

[0008] Since the parameters of the rock physics model are large, difficult to obtain, and have no practical physical meaning, the viscoelastic model is a better way to replace the rock physics model. The Zener model can fit the mesoscale White model (Stefano Picotti, Jose M. Carcione, J. German Rubino, et al., 2010), the Biot-Rayleigh model (Wang E, Carcione JM, Ba J., 2019), and the dispersion and attenuation of two scales (mesoscopic and microscopic) (Zhang B, Yang D, Cheng Y, et al., 2019). Some researchers believe that the Cole-Cole model can more accurately simulate the acoustic properties of porous media than the rough approximation of Zener's mechanical model (Stefano Picotti and Jose M. Carcione, 2017; Lu and Hanyga, 2004), but there are 5 parameters, which is relatively many, while the Power Law model only requires two parameters (Maksim Bano, 2004), which is easier to use. (Chapman, 2017) studied the seismic attenuation of bidirectional fluid-saturated Berea sandstone, compared the attenuation curves of the WIFF model and the WIGED model at different saturations and pressures, and used the SLS model for fitting, and found that the fitting effect with WIGED was very good. Some scholars have also compared multiple viscoelastic models. Paul compared the attenuation and frequency relationship curves of the Maxwell model, Voight solid, standard linear, Burgers solid and Power Law models. The results show that the attenuation curve and elastic stiffness curve vary greatly with frequency, and each model has a different change law (Paul, 2004). Toverud et al. compared 8 models on the zero-offset vertical seismic profile (VSP) data set: Kolsky-futterman model, Power Law model, Kjartansson model, Muller model, Azimi second model, Azimi third model, Cole-Cole model and standard linear-solid model (SLS). The results show that in the same depth area, the SLS model has the best results in simulating attenuation (Toverud Tand Ursin B, 2005).

[0009] Seismic wave attenuation is affected by wave-induced pore fluids, and the extension between synthetic data needs to be calibrated using well logging data and seismic data for more accurate interpretation. The core of well-seismic calibration is to understand the velocity dispersion between the well logging band and the seismic band. Traditional methods ignore the physical mechanism behind velocity dispersion, resulting in calibration errors.

[0010] In the Chinese patent application with application number: CN201510300798.6, a method for estimating seismic wave attenuation gradient based on variable modal decomposition is involved. The method includes: calibrating the target layer using well logging, mud logging and synthetic seismic records; performing variable modal decomposition on the seismic data channel by channel, extracting the attenuation gradient of the modal function in combination with the least squares method, and performing weighted summation on the attenuation gradient of the modal function to obtain the attenuation gradient value of each seismic channel; and determining the lithology and hydrocarbon properties of the target layer using the seismic attenuation gradient feature body. The invention can be more effectively applied to the attenuation gradient estimation of seismic signals with low signal-to-noise ratio and large spectrum fluctuations. Therefore, it can improve the effectiveness of absorbing attenuation gradients to indicate oil and gas, and can be accurately used to identify lithology, hydrocarbon properties, etc.

[0011] In the Chinese patent application with application number: CN201010521370.1, a method for reservoir analysis using seismic wave absorption and attenuation characteristics is involved, which belongs to the field of oil, gas and coalbed methane seismic exploration and development. This method is based on the theory of seismic wave absorption and attenuation and seismic spectrum decomposition technology. For different types of thin reservoirs, the seismic forward modeling technology is used to perform amplitude correction by analyzing the correlation between reservoir thickness and seismic amplitude and frequency. Then, under the constraints of geological formations, seismic absorption and attenuation attributes are used to perform reservoir evaluation and analysis based on stratigraphic units. This method achieves the purpose of evaluating the gas content of reservoirs using seismic wave absorption and attenuation attributes, and has a wide range of application value in the evaluation and description of heterogeneous reservoirs, especially in the evaluation and detailed description of unconventional reservoirs such as coalbed methane.

[0012] In the Chinese patent application with application number: CN201710242793.1, a method and device for simulating and analyzing the velocity and attenuation of seismic waves are involved. The method includes the following steps: establishing a saturated porous rock model and a saturated fractured rock model of the medium in the work area; under set conditions, embedding the saturated fractured rock model into the saturated porous rock model to obtain the elastic coefficient matrix of the medium; the elastic coefficient matrix includes the elastic coefficients of the medium in all directions; determining the complex velocity of the seismic wave propagating in the medium according to the elastic coefficient matrix of the medium; determining the inverse quality factor of the medium and the velocity of the seismic wave propagating in the medium according to the complex velocity. The analysis results obtained by this application embodiment can more objectively reflect the influence of the underground medium on the velocity and energy attenuation of seismic waves.

[0013] In the Chinese patent application with application number: CN201711276470.0, a method, device, system and computer-readable medium for calibrating synthetic seismic records are involved, and the method includes the following steps: using the density and acoustic wave curves corrected by the logging environment to calculate the reflection coefficient of the corresponding formation; calculating the initial synthetic seismic record based on the wavelet main frequency; locking the position of the standard layer relative to the time scale, and when the wavelet main frequency is constant, adjusting the velocity attenuation coefficient to calibrate the synthetic seismic record. According to the method, device, system and computer-readable medium for calibrating synthetic seismic records of the present invention, by obtaining the velocity attenuation coefficient without changing the reflection coefficient, the "non-uniform" matching of the wellbore seismic travel time can be achieved, ensuring the corresponding relationship between the synthetic record wave group energy and the seismic wave wave group energy.

[0014] The above existing technologies are greatly different from the present invention and fail to solve the technical problem we want to solve. For this reason, we have invented a new well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model. Summary of the invention

[0015] The purpose of the present invention is to provide a well seismic calibration method based on a wave-induced pore fluid dissolution seismic attenuation model, which uses an appropriate rock physics model to accurately predict velocity dispersion and calibrates the difference between the two data by coarsening the rock physics model.

[0016] The purpose of the present invention can be achieved by the following technical measures: a well seismic calibration method based on a wave-induced pore fluid dissolution seismic attenuation model, the well seismic calibration method based on a wave-induced pore fluid dissolution seismic attenuation model comprises:

[0017] Step 1: rock physics modeling based on geological, geochemical and core data;

[0018] Step 2, calculating the attenuation curve according to the rock physics model, and obtaining the roughened attenuation curve by roughening the viscoelastic model;

[0019] Step 3: calibrate the well seismic data to find a reasonable match between the synthetic seismic data and the field data.

[0020] The purpose of the present invention can also be achieved by the following technical measures:

[0021] In step 1, according to geological, geochemical and core data, a rock physics model is selected and the parameters in the model are coarsened. Three rock physics models, namely the WIGED model, the SLS model and the Pride model, are used to calculate the dispersion of the P-wave velocity.

[0022] In step 1, the bubble dissolution rate for:

[0023]

[0024] Where r is the radius of the bubble, D w For CO 2 -H 2 Diffusion coefficient of O, P f is the fluid pressure, γ is H 2 O-CO 2 The surface tension, K HPC For CO 2 -H 2 Henry's law constant for O, in the liquid shell around the bubble, C w is the concentration of the gas;

[0025] The rate of change of bubble radius for:

[0026]

[0027] Where R is the gas constant, T is the temperature, and the complex effective bulk modulus K of the fluid can be obtained based on the numerical calculation of the cyclic stress volume of micron bubbles of carbon dioxide. f , using K f The elastic modulus of Berea sandstone and the effective bulk modulus K of the saturated sample are calculated using the Gassmann fluid substitution method, and the complex Young's modulus E' of the saturated rock is calculated as:

[0028] E′=3K(1-2v) (3)

[0030] Where, v is Poisson’s ratio;

[0031] Equation (2) consists of two terms: and

[0032] These two terms are respectively proportional to the diffusion caused by bubble dissolution and bubble elastic deformation. c is defined as the frequency contained between the low frequency limit and the high frequency limit, where d I Equivalent to d R , for f c The estimate is:

[0033]

[0034] In step 1, the viscoelastic medium model is used to describe the propagation of waves in complex media; the force-displacement relationship can be expressed as:

[0035]

[0036] Where E1 and E 2 is the additional elastic modulus, η represents the viscosity coefficient of the fluid, σ represents the stress tensor of the solid skeleton, is the first-order derivative of strain, ε is the strain, and η is the material constant representing viscosity. Assuming sinusoidal motion is ε=ε 0 e iωt , σ=σ 0 e iωt And replace equation (5),

[0037] σ 0 =M(ω)ε 0 (6)

[0039] Has a complex frequency-dependent modulus

[0040]

[0041] Under the extreme conditions of low frequency and high frequency, the limiting modulus is obtained and M ∞ =E 2 ,ω→∞;

[0042] The frequency-dependent complex modulus is rewritten according to the following constraints:

[0043]

[0044] when When , the Q quality factor can be written as follows:

[0045]

[0046] If and only if ω=ω r The attenuation reaches its maximum value when:

[0047]

[0048]

[0049] ρ is the density, M ∞ and M 0 are the elastic moduli when ω→∞ and ω→0, respectively, and then V p Substituting (ω) into the Troplitz equation yields the complex reflection coefficient.

[0050] In step 1, the critical porosity Pride model is:

[0051]

[0052]

[0053] In the formula, K d and μ d are the bulk modulus and shear modulus of the dry frame, K 0 and μ 0 are the bulk modulus and shear modulus of the mineral, φ is the porosity, α is the integration parameter, φ c is the critical porosity, γ is CO 2 -H 2 The surface tension of O.

[0054] In step 2, the attenuation curve is calculated, and the attenuation Q(f) curve is calculated according to the rock physics model. The accurate Q(f) expression is obtained by coarsening the viscoelastic model.

[0055] In step 2, Backus proposed that a layered medium consisting of multiple layers of isotropic media is effectively anisotropic in the long-wave limit; the anisotropic Backus average gives a transversely isotropic medium described by five effective stiffnesses:

[0056]

[0057]

[0058]

[0059]

[0060]

[0062] In the formula, C ij and are the elastic stiffness coefficients of the equivalent medium before and after Backus averaging, respectively. <> represents the weighted average of the elastic parameters of different types of mineral components in brackets.

[0063] In step 2, constant-Q models provide a parameterization of rock seismic attenuation for petroleum exploration and seismology; by reducing the number of parameters, they can improve seismic inversion.

[0064] In step 2, the stress and strain in a one-dimensional linear inelastic medium are convolved as:

[0065]

[0066] Among them, ψ is the relaxation function, t is the time variable, and the symbol * represents time convolution. A dot above the variable represents the time derivative;

[0067] Define the relaxation function:

[0068]

[0069] Among them, M 0 is the bulk modulus, Γ is the Euler gamma function, t 0 is the reference time, γ is a dimensionless parameter, H is a unit step function, and Fourier transform of equation (19) yields:

[0070]

[0071] in, is the Fourier transform operator, M(ω) is the complex modulus, ∈(ω) is the strain, and ~ represents the Fourier transform. After calculation, we can get:

[0072]

[0073] Where M0 is the bulk modulus, is the reference frequency.

[0074] Calculate the phase velocity and attenuation factor, the complex velocity is:

[0075]

[0076] where ρ is the density, the phase velocity c is the frequency ω divided by the real part of the complex wave number, and then

[0077]

[0078] Substituting equations (22) and (23) into equation (24), we obtain:

[0079]

[0080]

[0081] C 0 It is at ω=ω 0 The phase velocity at 0 is the bulk modulus and γ is a dimensionless parameter.

[0082] Attenuation Factor:

[0083]

[0084] α is the attenuation factor, V is the complex velocity, and γ is a dimensionless parameter.

[0085] The quality factor is defined as the peak energy stored in the cycle divided by the energy lost in the cycle, using the equation:

[0086]

[0087] First, derive C from equation (25)0 It is at ω=ω 0 The phase velocity at , the reference frequency, and

[0088]

[0089] Secondly, from equation (28), it can be concluded that Q is independent of frequency, so

[0090]

[0091] Parameterizing the attenuation level, we can thus obtain that Q>0 is equivalent to 0<γ<1 / 2, when ω→0, c→0; when ω→∞, c→∞; it can be seen that the high frequencies of the signal propagate at almost infinite speed, and the differential equation describing the wave motion is parabolic.

[0092] In step 2, we compute the wave equation in differential form. Consider a two-dimensional wave equation of the form

[0093]

[0094] Where ω = (x, z, t) is a field variable, β is the order of the time derivative, D is a positive parameter, and Δ is the two-dimensional Laplace operator

[0095]

[0096] f is a forcing term, considering a plane wave

[0097] exp=[i(ωt-k x xk z z)] (33)

[0098] (k x , k z ) is the complex wave vector, and replacing algebra (33) in the wave equation (31) with f = 0 yields the dispersion equation:

[0099] (iω) β +Dk 2 =0 (34)

[0100] Where i represents the imaginary part, β is the order of the time derivative, and D is a positive parameter. is the complex wave number, equation (34) is the Fourier transform of equation (31), since k2 1 / 4 qx2 = M, and comparison of equations (34) and (22) yields

[0101] β=2-2γ and

[0102] Equations (31) and (35) are wave equations corresponding to Kjartansson's stress-strain relation. In order to obtain the true value of the quality factor corresponding to the propagation of rock waves, γ < < 1 and the time derivative in equation (31) have fractional order;

[0103] Using equations (22) and (35), we can obtain

[0104]

[0105] σ is stress, ρ is density, D is a positive parameter, ∈ is strain, and β is the order of the time derivative.

[0106] Combined with equation (36) is the momentum equation

[0107]

[0108]

[0109] Among them, u x and u z is the displacement component, redefining

[0110]

[0111] Differentiating and adding equations (37) and (38), substitution of equation (36) yields:

[0112]

[0113] in

[0114]

[0115] Use (iω) β-2 Multiplying by the Fourier transform of equation (40) and after inverse Fourier transform, we get the inhomogeneous wave equation:

[0116]

[0117] Where s is the earthquake source. If the medium is homogeneous, the type of the equation is equation (31).

[0118] The rock physics model gives the velocity dispersion between seismic data and well logging data for specific rock properties, the attenuation curve Q rock (f) can be used to generate synthetic seismic records:

[0119]

[0120] where u is the particle displacement, ω ris the reference angular frequency, ω is the angular frequency, the third term is the attenuation of the amplitude intensity caused by the viscoelastic effect, and the fourth term is the phase distortion caused by the viscoelastic effect;

[0121] Considering that the Q(f) expression in the rock physics model is unavailable, the viscoelastic model is coarsened using non-physical parameters, and then the attenuation characteristics are well modeled by minimizing the problem:

[0122] min∑ f |Q rock (f)-Q visco (f)| 2 (Equation 2)

[0124] The SLS model can simulate most rock physics models (Chapman, 2017), Q sLS (f) can be expressed as follows:

[0125]

[0126] Among them, τ σ is the stress relaxation time, τ is the strain relaxation time, and by combining equations 2 and 3, the coarsened Q(f) of the SLS model can be derived:

[0127]

[0128] Where L is the relaxation mechanism, Q rock (f) is the numerical simulation attenuation curve based on the rock physics model.

[0129] In step 3, the well seismic data is calibrated. According to the reflection coefficient and the attenuation curve Q(f), the attenuated synthetic seismic data can be obtained. By adjusting the Q-related parameters in the viscoelastic model, a reasonable match between the synthetic seismic data and the field data can be found.

[0130] The well-seismic calibration method based on the wave-induced porous fluid dissolution seismic attenuation model in the present invention uses an appropriate rock physics model to accurately predict velocity dispersion, and calibrates the difference between the two data by coarsening the rock physics model; the WIGED model can describe micron pores and is suitable for low-porosity, low-permeability shale reservoirs. For tight reservoir oil and gas reservoirs and unconventional shale reservoirs, due to the dense characteristics of their reservoirs, conventional models such as jet models are not applicable. Therefore, a rock physics model that considers dissolution and melting is established, which is more suitable for tight reservoirs. .The scale coarsening method of the multi-parameter rock physics model uses a viscoelastic model with fewer parameters to characterize the rock physics model with more parameters through the least squares method or fitting method, which reduces the complexity of the parameters and solves the problem of difficult to obtain parameters, thereby achieving the purpose of using the rock physics model to guide well-seismic matching. BRIEF DESCRIPTION OF THE DRAWINGS

[0131] Figure 1 It is a flow chart of a specific embodiment of the well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model of the present invention;

[0132] Figure 2 is a schematic diagram of an SLS structure in a specific embodiment of the present invention;

[0133] Figure 3 is a velocity dispersion and attenuation diagram calculated by the WIGED model in a specific embodiment of the present invention;

[0134] Figure 4 It is a velocity variation diagram of a specific embodiment of the present invention with a porosity of 0.04 and a gas saturation of 0.25%;

[0135] Figure 5 It is a velocity variation diagram of a specific embodiment of the present invention with a porosity of 0.06 and a gas saturation of 0.25%;

[0136] Figure 6 It is a velocity variation diagram of a specific embodiment of the present invention with a porosity of 0.08 and a gas saturation of 0.15%;

[0137] Figure 7 It is a velocity variation diagram of a specific embodiment of the present invention with a porosity of 0.08 and a gas saturation of 0.2%;

[0138] Figure 8 is a comparison diagram of 1 / Q curves of an IGED model and an SLS model in a specific embodiment of the present invention;

[0139] Fig. 9 This is a comparison chart of the application effect of the Backus averaging method when the time window is 16;

[0140] Fig.10This is a comparison chart of the application effect of the Backus averaging method when the time window is 4;

[0141] Fig.11 It is a graph showing the change of velocity of the WIGED model and the constant Q model in a porosity range of 0.04 to 0.08 and a gas saturation of 0.25% in a specific embodiment of the present invention;

[0142] Fig.12 It is a velocity variation diagram of the WIGED model and the constant Q model in a specific embodiment of the present invention when the porosity is 0.08 and the gas saturation range is 0.15% to 0.25%;

[0143] Fig.13 In one embodiment of the present invention, the Backus averaging method is used to compare synthetic seismic data maps from three different sized sample windows using accelerated speed;

[0144] Fig.14 In one specific embodiment of the present invention, Fig.11 Comparison of the accelerated maximum and minimum velocities of the constant Q model in synthetic seismic data;

[0145] Fig.15 In one specific embodiment of the present invention, Fig.11 Comparison of the accelerated velocities in the maximum and minimum velocities of the SLS model against the synthetic seismic data;

[0146] Fig.16 It is a well-seismic calibration diagram before and after the WIGED model in a specific embodiment of the present invention uses logging data and core measurement results to compare Q-1 results and uses the Q-1 results of the WIGED method as guidance. DETAILED DESCRIPTION

[0147] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0148] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations and / or combinations thereof.

[0149] like Figure 1 As shown, Figure 1The flowchart of the well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model of the present invention is as follows. The well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model comprises:

[0150] Step 1: Rock physics modeling: According to geological, geochemical and core data, a rock physics model is selected and the parameters in the model are coarsened. The dispersion calculation of P-wave velocity is performed using three rock physics models: WIGED model, SLS model and Pri de model.

[0151] Step 2: Calculate the attenuation curve and calculate the Q(f) curve based on the rock physics model. By roughening the viscoelastic model, an accurate Q(f) expression can be obtained.

[0152] Step 3: Well seismic data calibration: Based on the reflection coefficient and Q(f), attenuated synthetic seismic data can be obtained. By adjusting the Q-related parameters in the viscoelastic model, a reasonable match between the synthetic seismic data and the field data can be found.

[0153] The following are several specific embodiments of the present invention.

[0154] Example 1

[0155] In a specific embodiment 1 of the present invention, the well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model of the present invention includes:

[0156] Step 1: Rock physics modeling: According to geological, geochemical and core data, a rock physics model is selected and the parameters in the model are coarsened. The dispersion calculation of P-wave velocity is performed using three rock physics models: WIGED model, SLS model and Pride model.

[0157] (1) Bubble dissolution rate for:

[0158]

[0159] Where r is the radius of the bubble, D w For CO 2 -H 2 Diffusion coefficient of O, P f is the fluid pressure, γ is H 2 O-CO 2 The surface tension, K HPC For CO 2 -H 2 Henry's law constant for O, in the liquid shell around the bubble, C w is the concentration of the gas;

[0160] The rate of change of bubble radius for:

[0161]

[0162] Where R is the gas constant, T is the temperature, and the complex effective bulk modulus K of the fluid can be obtained based on the numerical calculation of the cyclic stress volume of carbon dioxide containing micrometer bubbles (μGB): f , using K f The elastic modulus of Berea sandstone and the effective bulk modulus K of the saturated sample are calculated using the Gassmann fluid substitution method, and the complex Young's modulus E' of the saturated rock is calculated as:

[0163] E′=3K(1-2v) (3)

[0165] Where, v is Poisson’s ratio;

[0166] Equation (2) consists of two terms: and

[0167] These two terms are respectively proportional to the diffusion caused by bubble dissolution and bubble elastic deformation. c is defined as the frequency contained between the low frequency limit and the high frequency limit, where d I Equivalent to d R , for f c The estimate is:

[0168]

[0170] (2) The viscoelastic medium model is used to describe the propagation of waves in complex media; the force-displacement relationship can be expressed as:

[0171]

[0172] Where E 1 and E 2 is the additional elastic modulus, η represents the viscosity coefficient of the fluid, σ represents the stress tensor of the solid skeleton, is the first-order derivative of strain, ε is the strain, and η is the material constant representing viscosity. Assuming sinusoidal motion is ε=ε 0 e iωt , σ=σ 0 e iωt And replace equation (5),

[0173] σ 0 =M(ω)ε 0 (6)

[0175] Has a complex frequency-dependent modulus

[0176]

[0177] Under the extreme conditions of low frequency and high frequency, the limiting modulus is obtained and M ∞ =E 2 ,ω→∞;

[0178] The frequency-dependent complex modulus is rewritten according to the following constraints:

[0179]

[0181] when When , the Q quality factor can be written as follows:

[0182]

[0183] If and only if ω=ω r The attenuation reaches its maximum value when:

[0184]

[0185]

[0186] ρ is the density, M ∞ and M 0 are the elastic moduli when ω→∞ and ω→0, respectively, and then V p Substituting (ω) into the Zoplitz equation, we get the complex reflection coefficient;

[0187] (3) The critical porosity Pride model is:

[0188]

[0189]

[0190] In the formula, K d and μ d are the bulk modulus and shear modulus of the dry frame, K 0 and μ 0 are the bulk modulus and shear modulus of the mineral, φ is the porosity, α is the integration parameter, φ c is the critical porosity;

[0191] Step 2: Calculate the attenuation curve and calculate the Q(f) curve based on the rock physics model. By roughening the viscoelastic model, an accurate Q(f) expression can be obtained.

[0192] (1) Backus proposed that a layered medium composed of multiple layers of isotropic media is equivalently anisotropic in the long-wave limit; the anisotropic Backus average gives a transversely isotropic medium described by five effective stiffnesses:

[0193]

[0194]

[0195]

[0196]

[0197]

[0198] In the formula, C ij and are the elastic stiffness coefficients of the equivalent medium before and after Backus averaging, respectively. <> represents the weighted average of the elastic parameters of different types of mineral components in brackets;

[0199] (2) Constant Q models provide a parameterization of rock seismic attenuation for petroleum exploration and seismology; by reducing the number of parameters, they can improve seismic inversion;

[0200] The stress and strain in a one-dimensional linear inelastic medium are convoluted:

[0201]

[0202] Among them, ψ is the relaxation function, t is the time variable, and the symbol * represents time convolution. A dot above the variable represents the time derivative;

[0203] Define the relaxation function:

[0204]

[0205] Among them, M 0 is the bulk modulus, Γ is the Euler gamma function, t 0 is the reference time, γ is a dimensionless parameter, H is a unit step function, and Fourier transform of equation (19) yields:

[0206]

[0207] in, is the Fourier transform operator, M(ω) is the complex modulus, ∈(ω) is the strain, and ~ represents the Fourier transform. After calculation, we can get:

[0208]

[0209] Among them, M 0 is the bulk modulus, is the reference frequency;

[0210] Phase velocity and attenuation factor: The complex velocity is

[0211]

[0212] where ρ is the density, the phase velocity c is the frequency ω divided by the real part of the complex wave number, and then

[0213]

[0214] Substituting equations (22) and (23) into equation (24), we obtain:

[0215]

[0216]

[0217] C 0 It is at ω=ω 0 The phase velocity at 0 is the bulk modulus and γ is a dimensionless parameter.

[0218] Attenuation Factor:

[0219]

[0220] α is the attenuation factor, V is the complex velocity, and γ is a dimensionless parameter.

[0221] The quality factor is defined as the peak energy stored in the cycle divided by the energy lost in the cycle, using the equation:

[0222]

[0223] First, derive C from equation (25) 0 It is at ω=ω 0 The phase velocity at , the reference frequency, and

[0224]

[0225] Secondly, from equation (28), it can be concluded that Q is independent of frequency, so

[0226]

[0227] Parameterizing the attenuation level, we can thus obtain that Q>0 is equivalent to 0<Y<1 / 2, when ω→0, c→0; when ω→∞, C→∞; it can be seen that the high frequencies of the signal propagate at almost infinite speed, and the differential equation describing the wave motion is parabolic;

[0228] Wave equation in differential form: Consider a two-dimensional wave equation of the form

[0229]

[0230] Where ω = (x, z, t) is a field variable, β is the order of the time derivative, D is a positive parameter, and Δ is the two-dimensional Laplace operator

[0231]

[0232] f is a forcing term, considering a plane wave

[0233] exp=[i(ωt-k x xk z z)] (33)

[0234] There are ω real numbers and (k x , k z ) complex wave vector, and replace algebra (33) in the wave equation (31) with f = 0 to obtain the dispersion equation:

[0235] (iω) β +Dk 2 =0 (34)

[0236] in, is the complex wave number, equation (34) is the Fourier transform of equation (31), since k2 1 / 4 qx2 = M, and comparison of equations (34) and (22) yields

[0237] β=2-2γ and

[0238] Equations (31) and (35) are wave equations corresponding to Kjartansson's stress-strain relation. In order to obtain the true value of the quality factor corresponding to the propagation of rock waves, γ < < 1 and the time derivative in equation (31) have fractional order;

[0239] Using equations (22) and (35), we can obtain

[0240]

[0241] Combined with equation (36) is the momentum equation

[0242]

[0243]

[0244] Among them, u x and u zis the displacement component, redefining

[0245]

[0247] Differentiating and adding equations (37) and (38), substitution of equation (36) yields:

[0248]

[0249] in

[0250]

[0251] Use (iω) β-2 Multiplying by the Fourier transform of equation (40) and after inverse Fourier transform, we get the inhomogeneous wave equation:

[0252]

[0253] Where s is the earthquake source. If the medium is homogeneous, the type of equation is equation (31);

[0254] The rock physics model gives the velocity dispersion between seismic data and well logging data for specific rock properties, the attenuation curve Q rock (f) can be used to generate synthetic seismic records:

[0255]

[0256] where u is the particle displacement, ω r is the reference angular frequency, ω is the angular frequency, the third term is the attenuation of the amplitude intensity caused by the viscoelastic effect, and the fourth term is the phase distortion caused by the viscoelastic effect;

[0257] Considering that the Q(f) expression in the rock physics model is unavailable, the viscoelastic model is coarsened using non-physical parameters, and then the attenuation characteristics are well modeled by minimizing the problem:

[0258] min∑ f |Q rock (f)-Q visco (f)| 2 (Equation 2)

[0260] The SLS model can simulate most rock physics models (Chapman, 2017). Figure 2Schematic diagram of the SLS structure in a specific embodiment of the present invention. E1 and E2 are springs representing elastic elements, η is the damping representing the viscous element, and the SLS model is a spring connected in parallel with a damping and then connected in series with a spring; Q SLS (f) can be expressed as follows:

[0261]

[0262] Among them, τ σ is the stress relaxation time, τ is the strain relaxation time, and by combining equations 2 and 3, the coarsened Q(f) of the SLS model can be derived:

[0263]

[0264] Where L is the relaxation mechanism, Q rock (f) is the numerical simulation attenuation curve based on the rock physics model;

[0265] Step 3: Well seismic data calibration: Based on the reflection coefficient and Q(f), attenuated synthetic seismic data can be obtained. By adjusting the Q-related parameters in the viscoelastic model, a reasonable match between the synthetic seismic data and the field data can be found.

[0266] Example 2

[0267] In a specific embodiment 2 of the present invention, the method uses three rock physics models, namely the WIGED model, the SLS model and the Pride model, to calculate the P-wave velocity dispersion, such as Figure 3 As shown in the figure, the solid line is the P-wave velocity and the dashed line is the attenuation. Here, we consider a rock with micron pores. There are gas microbubbles in the pores with a radius of 1 μm. Other parameters in the model are 1.8×10 9 Bulk modulus of oil (Ko), gas surface tension (Sigma) of 0.012 Pa·m, 8.3144621 m 3 The gas constant (R) of Pa / K mol and Henry's law constant of gas (KHpc) are 4535 Pa·m 3 / mol, temperature T is 350K, and the gas diffusion coefficient in water (DW) is 1.00×10 -8 m 2 / s, gas solubility (Sl) is 4535Pa·m 3 / mol, initial pressure (Pi) is 2.00×10 7 Pa, dP is 100Pa, and water viscosity (visc_f) is 1.00×10 -3 , the viscosity of carbon dioxide (visc_g) is 4.80×10 -5, water density (rho_w) is 840kg / m 3 、Gas density (rho_g) 119kg / m 3 , the viscosity of methane (visc_g) is 6.00×10 -6 , the viscosity of the oil (visc_f) is 8.00×10 -4 , high pressure (Ph) is 2.00×10 7 Pa, bubble initial radius (r) is 5.00×10 -6 m, water saturation (Sat) is 1.50×10 -3 , porosity (phi) is 6.00×10 -2 , the mineral content of sand (Sand) is 0.35, the clay mineral content (Clay) is 0.65, the Poisson's ratio (pois) is 0.14, and the bulk modulus of gas (Kg) is 1.00×10 4 , perm is 1.00×10 -12 m 2 and AR_c is 1.00×10 -3 .

[0268] The corner frequency of the Squirt model is about 10 5 Hz, Pride model is about 10 9 Hz. Compared with the other two models, the WIGED model is suitable for describing rocks in the seismic frequency band, and its angular frequency is between 0.1 and 100 Hz. The maximum values ​​of the attenuation curves of the three models are similar, and the attenuation predicted by the Squirt model and the Pride model is slightly larger. For velocity dispersion, the WIGED model and the Pride model have similar velocity ranges, with a lower limit of 2880 m / s and an upper limit of 3230 m / s. Although the lower limits are similar, the Squirt model gives a larger upper limit of 3400 m / s.

[0269] The WIGED model gives a frequency-dependent Q curve, considering that the constant Q assumption is always accepted for real data applications. The conventional constant Q theory is then compared to show the velocity dispersion between the logging frequency and the seismic frequency. The results of the velocity dispersion are shown in the figure, with a porosity range of 0.04 to 0.08 and a gas saturation of 0.15% to 0.25%. The results obtained by the SLS model are almost the same as those of the WIGED model. The conventional Q model gives a monotonic velocity curve, which is quite different from the WIGED model. From Figure 4 to Figure 5 , with the increase of porosity, the velocity range of the WIGED model decreases. The upper limit decreases from 3500m / s to 3000m / s, and the lower limit decreases from 3300m / s to 2550m / s. Figure 6 to Figure 7 , with the increase of gas saturation, the velocity range is not much different. Figures 4 to 7The solid black line represents the results of the WIGED model, the white dots represent the results of the SLS model, and the dashed line represents the results of the constant-Q model.

[0270] Comparison of velocity dispersion calculated using the WIGED model, SLS model, and constant Q model. The solid line is the result of the WIGED model, where the bubble radius is 4×10 -6 and 6×10 -6 The respective dashed curves are from the SLS model. The other dashed curves are from the constant-Q model, with the Q value selected from the WIGED model at 100 Hz.

[0271] The corresponding 1 / Q curve is shown in the figure ( Figure 8 ) are compared. The two curves in each sub-figure represent different bubble radii. As the porosity increases, the attenuation increases. However, the increase in gas saturation has little effect on the attenuation. Among them, the solid line is the WIGED model and the white dots are the SLS model.

[0272] Velocity changes from well logging data to seismic data based on rock physics model and Backus averaging method

[0273] Well logging data is used to demonstrate the velocity difference between the WIGED model and the Backus average method. First, using the Backus average method, the velocity changes are as follows Fig. 9 , 10 As shown, Fig. 9 This is a comparison chart of the application effect of the Backus averaging method when the time window is 16. The solid line is the measured speed data, and the dotted line is the data after Backus coarsening. The 16 sampling points in the time window are converted into one sampling point through the Backus averaging method; Fig.10 This is a comparison chart of the application effect of the Backus averaging method when the time window is 4. The solid line is the measured speed data, the dotted line is the data after Backus coarsening, and the 4 sampling points in the time window are converted into one sampling point by the Backus averaging method. Using windows of different sizes, the speed results obtained from the Backus averaging method show different properties. As the size of the window increases, the speed curve differs more from the actual data. However, the speed in these three cases has the same range of variation as the actual data.

[0274] The WIGED model and the constant Q model are applied to the real data for velocity correction. The results are as follows: Fig.11 and Fig.12 As shown in Figure 2. The speed improvement of the SLS model and the constant Q model compared to the Backus averaging method has obvious changes with the read data. The speed values ​​of both models are smaller than the real data. In these cases, the SLS model always gives a smaller speed than the constant Q model. Fig.11 and Fig.12Velocity variation diagrams from the WIGED model and the constant Q model. The white dots are the velocity range of the constant Q model, and the black dots are the velocity range of the WIGED model.

[0275] Velocity dispersion correction for synthetic seismic data

[0276] The SLS model, the constant Q model and the accelerated Backus averaging method are all applied to synthetic seismic data for velocity dispersion correction. The synthetic seismic data is obtained by convolving the reflection coefficient with the minimum phase wavelet. The velocity used initially is in the well logging band. The amplitude of the data does not change, only the phase changes with different velocities. First, the Backus averaging method is used, such as Fig.13 As shown in Figure 2, for three different window sizes, the synthetic data show little change from the real data. The event of about 0.7 seconds arrives earlier than the real data.

[0277] The constant Q model is used to correct the velocity dispersion. Fig.14 As shown in Figure 2, for the maximum velocity of the constant Q model, the synthetic data are similar to the actual data. For the minimum velocity, since its value is smaller than the logging velocity, the stretching effect of the synthetic data is obvious, especially for the data after 0.4s.

[0278] Apply the SLS model, such as Fig.15 As shown. Fig.11 The values ​​of the medium speed are much smaller, so the synthetic data shows much greater stretching than both the constant Q model and the Backus averaging method. You can see that the stretching starts at 0.2 seconds and gets larger at 0.4 seconds. The maximum stretching time is 0.9 seconds. The entire length of the data becomes 1.2 seconds.

[0279] Example 3

[0280] In a specific embodiment 3 of the present invention, the method is applied to field data collected in northern China. -1 The results are calculated using well logging data from the WIGED model. Fig.16 The left side shows the Q of the model at 60 Hz at depths from 1500m to 2500m. -1 Results. Measurement results and Q in core data -1 The measurements agree well. The speed is then increased through the SLS model. Fig.16 The borehole calibration is shown before and after using the velocity dispersion results of the SLS method as a guide for stretching the event. Fixed at 1900ms, 1820ms was stretched by 5ms. The event at 1740ms was stretched by 8ms and the event at 1820ms was fixed. The corresponding velocity dispersions were calculated to be about 7% and 4%, which are consistent with the results of the SLS model. The accelerated velocity is suitable for borehole calibration and makes the correction more reasonable.

[0281] When selecting a rock physics model for well-seismic calibration, it is necessary to consider the geological, geochemical and core data analysis reports, then select an appropriate rock physics model and update the parameters in the model. The Squirt model and the Pride model are not suitable for predicting the attenuation of microporous rocks in the seismic band. The energy loss caused by fluid flow can be ignored, and the dissolution and seepage of gas become the main attenuation mechanism.

[0282] Although the traditional constant Q model has been widely accepted, it cannot predict accurate velocity dispersion when the angular frequency occurs in the seismic band. For the same rock properties, its velocity range is different from the WIGED model. The velocity range of the Backus averaging method is almost similar to the actual data. The size of the selected window does not improve the accuracy of the velocity. Both methods are not suitable for coarsening. The SLS model is suitable for coarsening the WIGED model. Under different porosity and gas saturation conditions, the SLS model can simulate the velocity dispersion and attenuation curves well.

[0283] There are clear differences between the synthetic seismic data obtained from the velocities calculated using these methods. Since the velocity range of the Backus averaging method is similar, there is almost no stretching of the events in the synthetic seismic data. The constant Q model underestimates the velocity dispersion, so the synthetic seismic data show little stretching until 0.4s. Due to the accumulation of attenuation effects, attenuation can be observed. The SLS model predicts the attenuation reasonably, and the synthetic seismic data are clearly different from the records produced by the well log velocities.

[0284] The well-seismic calibration method uses the WIGED model to predict the attenuation of the seismic band by comparing it with the WIFF model, the Squirt model, and the Whi te model, because the WIGED model can describe micron pores and is suitable for low-porosity and low-permeability shale reservoirs. Compared with the traditional constant model and the Backus average method, the SLS model can be selected as a suitable model for the coarsening of the WIGED model because the SLS model is better in simulating attenuation.

[0285] The SLS model, constant Q model and Backus averaging method are used to correct the velocity dispersion of synthetic seismic data. The synthetic data of the Backus averaging method show little change from the actual data. For the maximum velocity of the constant Q model, the synthetic data is similar to the actual data. For the minimum velocity, since its value is less than the logging velocity, the stretching effect of the synthetic data is obvious, especially for the data after 0.4s. The velocity value in the SLS model is small, so the stretching of the synthetic data is greater than that of the constant Q model and the Backus averaging method.

[0286] This method of well-seismic calibration technology based on the wave-induced pore fluid dissolution seismic attenuation model is applied to field data collected in northern China to obtain Q in the measurement results and core data. -1 The measured results are consistent with the calculated velocity dispersion and the results of the SLS model, thus verifying the feasibility and accuracy of the method.

[0287] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein with equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0288] Except for the technical features described in the specification, all other technical features are known technologies to those skilled in the art.

Claims

1. A well seismic calibration method based on a wave-induced pore fluid dissolution seismic attenuation model, characterized in that: The well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model includes: Step 1: rock physics modeling based on geological, geochemical and core data; Step 2, calculating the attenuation curve according to the rock physics model, and obtaining the roughened attenuation curve by roughening the viscoelastic model; Step 3: calibrate the well seismic data to find a reasonable match between the synthetic seismic data and the field data.

2. The well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 1 is characterized in that: In step 1, according to geological, geochemical and core data, a rock physics model is selected and the parameters in the model are coarsened. Three rock physics models, namely the WIGED model, the SLS model and the Pride model, are used to calculate the dispersion of the P-wave velocity.

3. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 2 is characterized in that: In step 1, the bubble dissolution rate for: Where r is the radius of the bubble, D w is the diffusion coefficient of CO2-H2O, P f is the fluid pressure, γ is the surface tension of H2O-CO2, K HPC is the Henry's law constant of CO2-H2O. In the liquid shell around the bubble, C w is the concentration of the gas; The rate of change of bubble radius for: Where R is the gas constant, T is the temperature, and the complex effective bulk modulus K of the fluid can be obtained based on the numerical calculation of the cyclic stress volume of micron bubbles of carbon dioxide. f , using K f The elastic modulus of Berea sandstone and the effective bulk modulus K of the saturated sample are calculated using the Gassmann fluid substitution method, and the complex Young's modulus E' of the saturated rock is calculated as: E′=3K(1-2v) (3) Where, v is Poisson’s ratio; Equation (2) consists of two terms: and These two terms are respectively proportional to the diffusion caused by bubble dissolution and bubble elastic deformation. c is defined as the frequency contained between the low frequency limit and the high frequency limit, where d I Equivalent to d R , for f c The estimate is:

4. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 2 is characterized in that: In step 1, the viscoelastic medium model is used to describe the propagation of waves in complex media; the force-displacement relationship can be expressed as: Where E1 and E2 are additional elastic moduli, η represents the viscosity coefficient of the fluid, and σ represents the stress tensor of the solid skeleton. is the first-order derivative of strain, ε is strain, and η is the material constant representing viscosity; assuming sinusoidal motion is ε=ε0e iωt ,σ=σ0e iωt And replace equation (5), σ0=M(ω)ε0 (6) Has a complex frequency-dependent modulus Under the extreme conditions of low frequency and high frequency, the limiting modulus is obtained and M ∞ =E2,0→∞; The frequency-dependent complex modulus is rewritten according to the following constraints: when When , the Q quality factor can be written as follows: If and only if ω=ω r The attenuation reaches its maximum value when: ρ is the density, M ∞ and M0 are the elastic moduli when ω→∞ and ω→0, respectively. Then V p Substituting (ω) into the Troplitz equation yields the complex reflection coefficient.

5. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 2, characterized in that: In step 1, the critical porosity Pride model is: In the formula, K d and μ d are the bulk modulus and shear modulus of the dry framework, K0 and μ0 are the bulk modulus and shear modulus of the mineral, φ is the porosity, α is the integration parameter, φ c is the critical porosity, and γ is the surface tension of CO2-H2O.

6. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 1, characterized in that: In step 2, the attenuation curve is calculated, and the attenuation Q(f) curve is calculated according to the rock physics model. The accurate Q(f) expression is obtained by coarsening the viscoelastic model.

7. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 6 is characterized in that: In step 2, Backus proposed that a layered medium consisting of multiple layers of isotropic media is effectively anisotropic in the long-wave limit; the anisotropic Backus average gives a transversely isotropic medium described by five effective stiffnesses: In the formula, C ij and are the elastic stiffness coefficients of the equivalent medium before and after Backus averaging, respectively. 〈〉 represents the weighted average of the elastic parameters of different types of mineral components in brackets.

8. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 6 is characterized in that: In step 2, constant-Q models provide a parameterization of rock seismic attenuation for petroleum exploration and seismology; by reducing the number of parameters, they can improve seismic inversion.

9. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 8, characterized in that: In step 2, the stress and strain in a one-dimensional linear inelastic medium are convolved as: Among them, ψ is the relaxation function, t is the time variable, and the symbol * represents time convolution. A dot above the variable represents the time derivative; Define the relaxation function: Where M0 is the bulk modulus, Γ is the Euler gamma function, t0 is the reference time, γ is a dimensionless parameter, and H is the unit step function. Fourier transform of equation (19) yields: in, is the Fourier transform operator, M(ω) is the complex modulus of the volume, ∈(ω) is the strain, and ~ represents the Fourier transform. After calculation, we can get: Where M0 is the bulk modulus, is the reference frequency.

10. The well seismic calibration method based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 9, characterized in that: In step 2, the phase velocity and attenuation factor are calculated, and the complex velocity is: where ρ is the density, the phase velocity c is the frequency ω divided by the real part of the complex wave number, and then Substituting equations (22) and (23) into equation (24), we obtain: C0 is the phase velocity at ω = ω0, M0 is the bulk modulus, and γ is a dimensionless parameter; Attenuation Factor: α is the attenuation factor, V is the complex velocity, and γ is a dimensionless parameter; The quality factor is defined as the peak energy stored in the cycle divided by the energy lost in the cycle, using the equation: First, from equation (25) it is deduced that C0 is the phase velocity at ω = ω0, the reference frequency, and Secondly, from equation (28), it can be concluded that Q is independent of frequency, so Parameterizing the attenuation level, we can thus obtain that Q>0 is equivalent to 0<γ<1 / 2, when ω→0, c→0; when ω→∞, c→∞; it can be seen that the high frequencies of the signal propagate at almost infinite speed, and the differential equation describing the wave motion is parabolic.

11. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 10, characterized in that: In step 2, we compute the wave equation in differential form. Consider a two-dimensional wave equation of the form Where ω = (x, z, t) is a field variable, β is the order of the time derivative, D is a positive parameter, and Δ is the two-dimensional Laplace operator f is a forcing term, considering a plane wave exp=[i(ωt-k x x-k z z)] (33) (k x , k z ) is the complex wave vector, and replacing algebra (33) in the wave equation (31) with f = 0 yields the dispersion equation: (iω) β +Dk 2 =0 (34) Where i represents the imaginary part, β is the order of the time derivative, and D is a positive parameter. is the complex wave number, equation (34) is the Fourier transform of equation (31), since k2 1 / 4qx2=M, comparison of equations (34) and (22) yields Equations (31) and (35) are wave equations corresponding to Kjartansson's stress-strain relation. In order to obtain the true value of the quality factor corresponding to the propagation of rock waves, γ < < 1 and the time derivative in equation (31) have fractional order; Using equations (22) and (35), we can obtain σ is stress, ρ is density, D is a positive parameter, ∈ is strain, and β is the order of the time derivative; Combined with equation (36) is the momentum equation Among them, u x and u Z is the displacement component, redefining Differentiating and adding equations (37) and (38), substitution of equation (36) results in: in Use (iω) β-2 Multiplying by the Fourier transform of equation (40) and after inverse Fourier transform, we get the inhomogeneous wave equation: Where s is the earthquake source. If the medium is homogeneous, the type of equation is equation (31); The rock physics model gives the velocity dispersion between seismic data and well logging data for specific rock properties, the attenuation curve Q rock (f) can be used to generate synthetic seismic records: where u is the particle displacement, ω r is the reference angular frequency, ω is the angular frequency, the third term is the attenuation of the amplitude intensity caused by the viscoelastic effect, and the fourth term is the phase distortion caused by the viscoelastic effect; Considering that the Q(f) expression in the rock physics model is unavailable, the viscoelastic model is coarsened using non-physical parameters, and then the attenuation characteristics are well modeled by minimizing the problem: min∑ f |Q rock (f)-Q visco (f)| 2 (Equation 2) The SLS model can simulate most rock physics models, Q SLS (f) can be expressed as follows: Among them, τ σ is the stress relaxation time, τ is the strain relaxation time, and by combining equations 2 and 3, the coarsened Q(f) of the SLS model can be derived: Where L is the relaxation mechanism, Q rock (f) is the numerical simulation attenuation curve based on the rock physics model.

12. The method for well seismic calibration based on the wave-induced pore fluid dissolution seismic attenuation model according to claim 1, characterized in that: In step 3, the well seismic data is calibrated. According to the reflection coefficient and the attenuation curve Q(f), the attenuated synthetic seismic data can be obtained. By adjusting the Q-related parameters in the viscoelastic model, a reasonable match between the synthetic seismic data and the field data can be found.

Citation Information

Patent Citations

  • Method of using absorption and attenuation characteristics of seismic wave for reservoir analysis

    CN102707317A

  • Method for estimating attenuation gradient of seismic waves based on variable mode decomposition

    CN105093294A

  • Seismic wave speed and attenuation simulation analysis method and device

    CN106959469A

  • Method for calibrating synthetic seismic records, apparatus, system, and computer-readable medium

    CN108037532A