WO2020215170A1 - Seismic petrophysical experiment analysis-based logging and seismic speed matching method - Google Patents

Abstract

A logging and seismic speed matching method based on seismic petrophysical experiment analysis, comprising: step 1, establishing, by means of seismic frequency band petrophysical experiment analysis, a curve indicating the change of speed with frequency, i.e. the change rule of velocity frequency dispersion; step 2, constructing a petrophysical model suitable for characterizing dispersion features of a seismic frequency band; step 3, using a petrophysical model modulus calculation formula to perform point-by-point mapping dispersion correction on a sonic curve; and step 4, performing synthetic seismogram and calibration on the sonic curve subjected to dispersion correction, comparing the near-well seismic trace with the synthetic seismogram for analysis, and outputting a correction curve result of all depths of the whole sonic logging curve. Said method solves the problem of matching between seismic speed and logging speed in terms of scale, and corrects the speed dispersion caused by a large difference between seismic speed and logging speed in terms of a frequency band, having clear physical content and a petrophysical foundation, greatly improving the precision of well-to-seismic matching and calibration.

Description

Velocity matching method of logging and seismic based on seismic rock physics experiment analysis Technical field

The invention relates to the technical field of exploration geophysics, in particular to a method for matching logging and seismic velocity based on seismic rock physics experiment analysis.

Background technique

In the process of inversion and interpretation, in order to achieve the combination and comparative analysis of seismic data and logging data, the usual method is to make synthetic records of logging acoustic data, and use synthetic records to calibrate the seismic data so that the time horizon information of the earthquake is The depth of logging and lithological characteristics are combined. However, due to the differences in scale, frequency and propagation path between surface seismic and sonic logging in acquisition and processing, the formation velocity obtained by sonic logging is inconsistent with the velocities obtained by surface seismic and other observation programs such as VSP. In the process of well-seismic matching, it is necessary to edit and correct the logging data so that the synthetic record and the seismic record have a good correspondence. At present, the most widely used well-seismic matching methods are the closure error method and the ratio method. In actual work, the closure error method and the ratio method are the overall "translation" of the synthetic record corresponding to the seismic trace through the well, or the partial “stretching” and “compression”, in order to make the synthetic record and the seismic trace beside the well The relevance is optimal. However, the disadvantage of this approach is that the physical meaning is not clear.

In terms of the dispersion correction of the acoustic wave velocity curve, the conventional acoustic wave dispersion correction method based on the viscoelastic solid model and the resonant Q model, although the quantitative influence caused by the wave dispersion effect can be considered from the physical mechanism, so as to reasonably correct the acoustic wave The velocity dispersion ratio makes it match the observation velocity in the seismic wave frequency band, which has a certain effect on improving the accuracy of well-seismic matching based on the design of petrophysical theoretical models. But these methods are given a specific scale factor to correct the logging speed to the seismic frequency speed, that is, it is assumed that the dispersion characteristics of the reservoir rocks are consistent. However, due to changes in the physical properties (porosity, permeability, pore structure) of the reservoir rock in the vertical and horizontal directions, its dispersion characteristics are different at each point. It is more reasonable and more effective to improve the accuracy of well-seismic matching according to the changes of reservoir lithology, physical properties and fluid properties. Therefore, the above methods have certain defects and limitations.

In actual engineering, the observation frequency of acoustic logging is generally in the range of 2K～20KHz, which is much higher than the frequency of seismic waves in conventional seismic exploration (10-125Hz). The significant frequency difference makes the acoustic waves of logging and the seismic waves excited by ground seismic exploration There are obvious differences in the propagation velocity in fluid-laden heterogeneous formations. The large frequency difference between the logging frequency band and the seismic frequency band causes a large difference in the natural frequency dispersion and the dispersion frequency of the two, which has become a key factor affecting the matching degree of logging speed and seismic speed, and further affects the seismic calibration. The accuracy affects the effect of seismic interpretation and reservoir inversion. Therefore, the key to sonic logging data to solve the well-seismic matching problem is to adjust the sonic logging speed reasonably to eliminate the influence of velocity dispersion, and to make it more compatible with the seismic velocity through the iterative correction method of synthetic records. The conventional sound wave curve dispersion correction method based on viscoelastic solid model and resonance Q model, although the quantitative influence caused by the wave dispersion effect can be considered from the physical mechanism, so as to reasonably correct the dispersion ratio of the sound wave speed to make it and The observation speed in the seismic wave frequency band is matched, so that the well-seismic matching can be realized on the basis of the design of the rock physics theoretical model. But these methods are given a specific scale factor to correct the logging speed to the seismic frequency speed, that is, it is assumed that the dispersion characteristics of the reservoir rocks are consistent. However, due to changes in the physical properties (porosity, permeability, pore structure) of the reservoir rock in the vertical and horizontal directions, its dispersion characteristics are different at each point. It is more reasonable and more effective to improve the accuracy of well-seismic matching according to the changes of reservoir lithology, physical properties and fluid properties.

In order to improve the matching accuracy of sonic logging velocity and seismic velocity, the accuracy of synthetic record calibration is further improved. We invented a new method for matching logging and seismic velocity based on seismic rock physics experiment analysis, which solved the above technical problems.

Summary of the invention

The purpose of the present invention is to provide a point-to-point dispersion correction of the acoustic curve based on the actual seismic frequency band petrophysical experiment results and petrophysical theoretical model, and then make a synthetic record and compare it with the well side channel, and further improve the well seismic through iterative correction A method of matching accuracy between logging and seismic velocity based on seismic rock physics experimental analysis.

The purpose of the present invention can be achieved by the following technical measures: a method for matching logging and seismic velocity based on seismic rock physics experimental analysis, and the method for matching logging and seismic velocity based on seismic rock physics experimental analysis includes: Step 1, through seismic frequency band Analyze the rock physics experiment to establish the velocity change curve with frequency, that is, the velocity dispersion change law; step 2, build a rock physics model suitable for characterizing the frequency dispersion characteristics of the seismic frequency; step 3, use the rock physics model modulus calculation formula to perform acoustic waves Point-by-point mapping of the curve dispersion correction; step 4, perform synthetic record calibration on the sonic curve after dispersion correction, compare and analyze the side channel of the seismic well with the synthetic seismic record, and output the correction curve for all depths of the entire sonic logging curve result.

The purpose of the present invention can also be achieved through the following technical measures:

In step 1, select the target formation core, carry out the full-band petrophysical laboratory test, measure the elastic parameters of different lithology, physical properties, and fluid-containing properties, including the longitudinal wave speed and shear wave speed at different frequency points; according to the laboratory test results , To form the relationship curve between velocity and frequency change, quantitatively analyze the difference between the velocity corresponding to the seismic frequency band and the velocity corresponding to the logging frequency band, and estimate the correction amount for matching the logging and seismic velocity.

In step 2, the rock with only dry and hard pores is used as the new equivalent matrix, its bulk modulus is K ^stiff , replaced by K ^h , soft pores are added and the effect of the fluid jet in the soft and hard pores is considered. The medium equivalent modulus K _mf and μ _mf are calculated by the following formula:

Where ω is the circular frequency, η is the dynamic viscosity of the pore fluid, p is the effective pressure of the rock, α _c is the pore aspect ratio, φ _c (P) is the porosity of the soft pore under a certain effective pressure p, μ _d (p) are the dry volume and shear modulus when the rock medium contains soft pores with an aspect ratio of α _c and porosity φ _c (p) under a certain effective pressure; the second item on the right is the addition of a specific aspect ratio Soft pores change the bulk modulus K ^stiff under consideration of jetting action; after considering the effect of soft pores, the remaining hard pores still satisfy the Gassmann equation after fluid saturation due to their incompressibility. At this time, the volume of the hard pores is fully saturated The modulus K _sat and shear modulus μ _sat are calculated by the following formula:

μ _sat (p, ω) = μ _mf (p, ω). (2)

In the process of adding soft pores iteratively, in addition to adding soft pores for the first time, the value calculated by adding soft pores for the kth time will be regarded as the K ^h and μ ^h added to the soft pores for the (k+1)th time; On the basis of K ^h , soft pores are added to obtain the (k+1)th K _d and μ _d ; the entire process of adding soft pores is expressed as:

According to the Betti reciprocity theorem, the modulus of the non-"relaxed" part is obtained when the plastic porosity is known, and the results are as follows:

In formulas (4) and (5), K _uf and μ _uf are the bulk modulus and shear modulus of the high-frequency non-relaxed rock skeleton respectively. K _dry-hP is the bulk modulus of dry rock under higher pressure, φ _soft Is the plastic porosity at a given pressure; using formula (6), the shear modulus and bulk modulus of saturated rock under different pressures have the following relationship:

In step 3, the petrophysical model modulus calculation formula is used to perform the point-by-point mapping dispersion correction of the acoustic wave curve according to the porosity, permeability and GR (gamma) curve division of the lithology and logging oil-water interpretation conclusions on the well ; According to the petrophysical model of micro-jet mechanism with multiple pore distribution constructed in step 2, the main model parameters that need to be input are the lithology, porosity, permeability and pore structure distribution of the reservoir rock.

In step 3, the lithology of the reservoir rock is judged by the GR (gamma) curve. The corresponding mudstone baseline in the gamma curve represents mudstone, and the sandstone with a large variation range can be seen from the law obtained from low-frequency petrophysical testing. : The pore structure of mudstone is simple, and dry mudstone has basically no dispersion. There is no need to calibrate from logging frequency to seismic frequency. The pore structure of tight reservoir sandstone is complex. When the soft pores are relatively developed, the dispersion amplitude is large. There is a big difference between the well frequency band velocity and the seismic frequency band velocity, and the amount of dispersion correction is large.

In step 3, the permeability of the reservoir rock is calculated using the Kozeny-Carman relationship; the Kozeny-Carman formula is expressed as follows:

In the formula, k is the permeability of the reservoir rock, τ is the pore curvature, d is the particle size of the constituent rock, and Φ is the porosity of the reservoir rock.

In step 3, the determination of the distribution of pore results is to obtain the pore distribution characteristics of the reservoir sandstone through logging data. Because the reservoir rocks mainly undergo the same sedimentation and diagenesis process, it is assumed that the pore structure has the same distribution characteristics, and the research is assumed The target interval of the reservoir has the same pore structure distribution characteristics as the test sample, and only a proportional correction is made according to the relative size of the porosity; the log data is used to obtain the pore distribution characteristics of the reservoir sandstone; it can be seen from the results of the petrophysical experiment, The characteristic frequencies of tight sandstones in the study area are mainly located in seismic frequency bands, and fluid-saturated rocks at logging frequencies represent high-frequency velocities; the velocity of reservoir rocks is determined by porosity and pore shape.

In step 3, after the pore distribution characteristics of the reservoir section are obtained, the petrophysical model in step 2 is used to correct the pore fluid-related velocity dispersion effect; the correction amount is higher than that in the reservoir with more fractures. The few reservoir sections show the effect of jet flow.

In step 4, the synthetic record calibration is performed on the sonic curve after dispersion correction, and the side channel of the seismic well is compared with the synthetic seismic record. If the matching degree is high and the correlation coefficient is more than 80%, the sonic correction of the well is directly output If the matching degree is not high, return to the petrophysical model parameters for parameter correction, and then perform the point-by-point dispersion correction of the acoustic curve until the matching correlation coefficient between the side channel of the seismic well and the synthetic seismic record reaches 80%. .

In step 4, the modified parameters include the aspect ratio of hard pores and the distribution range of aspect ratio of soft pores.

The logging and seismic velocity matching method based on seismic petrophysical experiment analysis in the present invention further improves the accuracy of well-seismic matching through the idea of iterative matching. This method is based on the actual seismic frequency band petrophysical experiment results and petrophysical theoretical model. According to the changes of reservoir lithology, physical properties and fluid properties, the acoustic wave curve is mapped to the dispersion correction point by point, which has higher dispersion correction accuracy. , And then make synthetic records and compare them with side channels, and further iteratively modify the model parameters to make them more compatible with seismic records, which greatly improves the accuracy of well-seismic matching. The conventional well-seismic matching method often stretches and compresses the acoustic synthetic record reflection horizon during the synthetic record calibration process to make it correspond to the actual seismic record reflection, which is called the closure error method and the ratio method. The synthetic record corresponds to the overall “translation” of the seismic trace across the well, or the local “stretch” and “compression”, in order to optimize the correlation between the synthetic record and the seismic trace beside the well. However, the disadvantage of this approach is that the physical meaning is not clear, and the larger extension or compression will change the time-depth relationship on the well, causing the seismic horizon and geological stratification to not correspond in depth. The well-seismic matching correction method formed by this project not only eliminates the matching problem of seismic velocity and logging velocity on the scale, but also corrects the velocity dispersion problem caused by the large difference in frequency band between the two. It has a clear The physical connotation and rock physics foundation have greatly improved the accuracy of well-seismic matching and calibration.

Description of the drawings

FIG. 1 is a flowchart of a specific embodiment of a method for matching logging and seismic velocity based on seismic rock physics experimental analysis of the present invention;

Fig. 2 is a schematic diagram of a speed versus frequency curve obtained by a laboratory test in a specific embodiment of the present invention;

Fig. 3 is a comparison diagram of the dispersion correction result of the multiple pore distribution petrophysical model of HG102 well and the correction result of a conventional well curve in a specific embodiment of the present invention;

Figure 4 is a schematic diagram of synthetic seismic records from the original well curve of Well HG102 in a specific embodiment of the present invention;

Fig. 5 is a schematic diagram of the synthetic seismic record of the HG102 well calibration well curve in a specific embodiment of the present invention.

Detailed ways

In order to make the above-mentioned and other objects, features and advantages of the present invention more obvious and understandable, preferred embodiments are listed below, which are illustrated in conjunction with the drawings, and are described in detail as follows.

As shown in Fig. 1, Fig. 1 is a flow chart of the method for matching logging and seismic velocity based on seismic petrophysical experiment analysis of the present invention.

Step 101: Establish a curve of velocity versus frequency, that is, the law of velocity dispersion change, by analyzing the rock physics experiment in the seismic frequency band. Specifically, the target formation core is selected to carry out full-frequency (2-2000HZ, 1Mhz) petrophysical laboratory tests to measure the elastic parameters of different lithology, physical properties, and fluid-containing properties, including longitudinal wave velocity and transverse wave velocity at different frequency points. According to the laboratory test results, the relationship curve between the speed and the frequency change is formed, and the difference between the speed corresponding to the seismic frequency band and the speed corresponding to the logging frequency band is quantitatively analyzed, and the correction amount for matching the logging and seismic speed is estimated. The measurement and analysis show obvious changes in speed with frequency in the frequency range of 2 to 1000 Hz, and can better reflect the high and low frequency speed limit values.

According to the dispersion model given by the multi-porous jet rock physics theory, the velocity change characteristics of the sample in the frequency range of 1 to 1000 Hz can be more accurately characterized, and the law of seismic dispersion can be accurately characterized.

Step 102: Construct a petrophysical model suitable for characterizing the dispersion characteristics of seismic frequency bands, that is, a microscopic jet petrophysical model with multiple pore distributions, and consider the influence of fluid relaxation in soft pores on its flexibility, which can reflect seismic wave-induced pores In particular, the jet flow formed by the compression and closure of soft pores.

Classical Biot theoretical model, Gassmann equation, etc. believe that the dispersion of seismic frequency band is much less than 2%, which can basically be ignored. The micro-jet petrophysical model with multiple pore distribution takes into account the influence of fluid relaxation in soft pores on its flexibility. The dispersion effect mainly comes from the fluid flow in the pores induced by seismic waves, especially the compression and closure of soft pores. Jet stream effect.

The modulus calculation formula derivation process of this model is embodied as: taking the rock with only dry hard pores as the new equivalent matrix, its bulk modulus is K ^stiff (usually replaced by K ^h ), adding soft pores and considering soft pores and The medium equivalent modulus K _mf and μ _mf can be calculated by the following formula under the influence of the fluid jet in the hard pores:

Where ω is the circular frequency, η is the dynamic viscosity of the pore fluid, φ _c (P) is the porosity value of the soft pore under a certain effective pressure p, and μ _d (p) is the aspect ratio of the rock medium under a certain effective pressure. Is the dry volume and shear modulus of soft pores with α _c and porosity φ _c (p). The second term at the right end can also be understood as adding soft pores with a specific aspect ratio and changing the bulk modulus K ^stiff under consideration of jetting. After considering the effect of soft pores, the remaining hard pores still satisfy the Gassmann equation after fluid saturation due to their incompressibility. At this time, the bulk modulus K _sat and shear modulus μ _sat when the hard pores are fully saturated can be calculated by the following formula:

μ _sat (p, ω) = μ _mf (p, ω). (2)

The aspect ratio of the soft pores in the actual rock cannot be a fixed value, but is continuously distributed within a certain range. After obtaining the distribution data of the soft pore aspect ratio in the rock with respect to the porosity, if it is discretized, the influence of the soft pore jet flow can still be calculated by formulas for each discretized pore aspect ratio and corresponding porosity. Then the K _mf and μ _mf of the rock based on the pore distribution are calculated by iteratively adding the soft pores of each aspect ratio. In the process of adding soft pores iteratively, in addition to adding soft pores for the first time, the value calculated by adding soft pores for the kth time will be regarded as the K ^h and μ ^h added to the soft pores for the (k+1)th time; On the basis of K ^h , soft pores are added to obtain the (k+1)th K _d and μ _d . The entire process of adding soft pores can be expressed as:

The total dispersion of elastic waves in rocks can be regarded as the comprehensive cumulative effect of the dispersion of soft pores in various aspect ratios. It is generally believed that when the saturated fluid has a low viscosity, the jet flow acts at the logging frequency or higher ultrasonic frequency. Through the analysis, it can be seen that considering the actual distribution characteristics of soft pores in rock media, the velocity value may have a non-negligible difference from the result of Gassmann equation even in the seismic frequency band. At the same time, there is a significant difference in velocity between the logging frequency band and the seismic frequency band.

Therefore, the velocity dispersion based on the jet stream mechanism is the main cause of velocity dispersion due to uneven pore stiffness (caused by pores or cracks with different surface rates and differences in the arrangement of pores or cracks). According to this model, the viscous non-"relaxed" part of the pore space (the unbalanced part of pore pressure) is equal to the plastic pores closed under high pressure, and the elastic modulus of dry rock under high pressure is approximately equal to that when the rock is saturated with fluid and undrained. Modulus. According to Betti's reciprocity theorem, the modulus of the non-"relaxed" part can be obtained when the plastic porosity is known, and the results are as follows:

In formulas (4) and (5), K _uf and μ _uf are the bulk modulus and shear modulus of high-frequency non-"relaxed" rock skeleton, respectively, and K _dry-hP is the bulk modulus of dry rock under higher pressure, φ _soft is the plastic porosity under a given pressure. Using formula (6), it can be obtained that the shear modulus and bulk modulus of saturated rock under different pressures have the following relationship:

Step 103: Using the petrophysical model modulus calculation formula, according to the porosity, permeability and gamma curve on the well, the lithology divided by the gamma curve and the logging oil-water interpretation conclusions, carry out the point-by-point mapping dispersion correction of the acoustic wave curve;

According to the petrophysical model of the micro-jet mechanism with multiple pore distributions constructed in step 102, the main model parameters that need to be input are the lithology, porosity, permeability, and pore structure distribution of the reservoir rock.

The lithology of the reservoir rock is mainly judged by the GR (gamma) curve. The mudstone baseline in the gamma curve represents mudstone, and the sandstone with a large variation range. The law obtained from the low-frequency petrophysical test shows: mudstone pore structure Single, dry mudstone has basically no frequency dispersion, and there is no need to calibrate the speed from logging frequency band to seismic frequency band. However, the pore structure of tight reservoir sandstone is complicated. When soft pores (micro-fractures) are relatively developed, the dispersion amplitude is relatively large. There is a big difference between the well frequency band velocity and the seismic frequency band velocity, and the amount of dispersion correction is large.

The permeability of reservoir rocks is mainly calculated using the Kozeny-Carman relationship. The Kozeny-Carman formula is expressed as follows:

In the formula, τ is the pore curvature (the ratio of the flow distance between the two points in the pore and the straight line distance, usually taken as the fixed value τ = 3), and d is the particle size of the rock composition (the particle diameter needs to be determined as 80μm based on the CT scan image results ).

The determination of the distribution of pore results is to obtain the pore distribution characteristics of the reservoir sandstone through logging data. Since the reservoir rocks mainly undergo the same sedimentation and diagenesis process, it can be assumed that the pore structure has the same distribution characteristics, and the target layer of the studied reservoir is assumed The section and the test sample have the same pore structure distribution characteristics, and only a proportional correction is made according to the relative size of the porosity. Use logging data to obtain the pore distribution characteristics of the reservoir sandstone. It can be seen from the results of petrophysical experiments that the characteristic frequencies of tight sandstones in the study area are mainly located in seismic frequency bands, and fluid (water, oil) saturated rocks at logging frequencies represent high-frequency velocities. The velocity of the reservoir rock depends on the porosity and pore shape.

After obtaining the pore distribution characteristics of the reservoir section, the petrophysical model in step 2 can be used to correct the velocity dispersion of the pore fluid. The correction amount in the reservoir with more fractures is higher than that in the reservoir with less fractures, indicating the effect of jet flow.

Step 104: Perform synthetic record calibration on the sonic curve after dispersion correction, compare and analyze the side channel of the seismic well with the synthetic seismic record, if the matching degree is high (correlation coefficient above 80%), directly output the sonic correction speed of the well If the matching degree is not high, return to the petrophysical model parameters for parameter correction, and then carry out the point-by-point dispersion correction of the acoustic curve until the matching correlation coefficient between the side channel of the seismic well and the synthetic seismic record reaches 80%. Finally, the calibration curve results for all depths of the entire sonic logging curve are output.

The iterative matching process of the calibration of the sonic synthetic record after dispersion correction is: the side channel of the seismic well and the synthetic seismic record are compared and analyzed. If the matching degree is high (correlation coefficient above 80%), the sonic correction speed of the well is directly output. If the matching degree is not high, return to the petrophysical model parameters for parameter correction. The corrected parameters include the aspect ratio of hard pores, the distribution range of soft pore aspect ratio, etc. After parameter correction, the point-by-point dispersion correction of the acoustic curve is performed until The iterative correction is terminated after the matching correlation coefficient between the side channel of the seismic well and the synthetic seismic record reaches 80%. Finally, the calibration curve results for all depths of the entire sonic logging curve are output.

After speed correction, the forward synthetic seismic record improves the consistency with the side channel of the well, and improves the quality of fine horizon calibration and small layer comparison. The correlation coefficient of the seismic calibration increases from 0.55 of the original well curve to 0.83 of the ES model; However, in terms of the deep-hour correspondence, compared with the original longitudinal wave well curve, the deep-hour correspondence has been improved to a certain extent after speed correction.

In a specific embodiment applying the present invention, it includes:

(1) Establishment of the law of speed and frequency change

Select target formation cores and carry out full-frequency (2-2000HZ, 1Mhz) petrophysical laboratory tests to measure elastic parameters of different lithology, physical properties, and fluid-bearing properties, including longitudinal wave velocity and transverse wave velocity at different frequency points. According to the laboratory test results, the relationship curve between the speed and the frequency change is formed, and the difference between the speed corresponding to the seismic frequency band and the speed corresponding to the logging frequency band can be quantitatively analyzed, and the correction amount for matching the logging and seismic speed can be estimated. The measurement and analysis show obvious changes in speed with frequency in the frequency range of 2 to 1000 Hz, and can better reflect the high and low frequency speed limit values.

As shown in Figure 2, the dispersion model given by the multi-porous jet rock physics theory can more accurately characterize the velocity change characteristics of the sample in the frequency range of 1 to 1000 Hz, and accurately characterize the law of seismic dispersion.

(2) Acoustic curve dispersion correction rock physics model

Classical Biot theoretical model, Gassmann equation, etc. believe that the dispersion of seismic frequency band is much less than 2%, which can basically be ignored. The tight sandstone reservoirs tested by petrophysical tests in the actual seismic frequency band often have very obvious dispersion characteristics in the seismic frequency band. The dispersion effect mainly comes from the fluid flow in the pores induced by seismic waves, especially the jet stream formed by the compression and closure of soft pores. Therefore, it is necessary to construct a petrophysical model suitable for characterizing the dispersion characteristics of seismic frequency bands, and the micro-jet petrophysical model with multiple pore distribution takes into account the influence of fluid relaxation in soft pores on its flexibility.

The modulus calculation formula derivation process of this model is embodied as: taking the rock with only dry hard pores as the new equivalent matrix, its bulk modulus is K ^stiff (usually replaced by K ^h ), adding soft pores and considering soft pores and The medium equivalent modulus K _mf and μ _mf can be calculated by the following formula under the influence of the fluid jet in the hard pores:

Where ω is the circular frequency, η is the dynamic viscosity of the pore fluid, φ _c (P) is the soft pore porosity under a certain effective pressure p, and μ _d (p) is the aspect ratio of the rock medium under a certain effective pressure. The dry volume and shear modulus of soft pores with _c and porosity φ _c (p). The second term at the right end can also be understood as adding soft pores with a specific aspect ratio and changing the bulk modulus K ^stiff under consideration of jetting. After considering the effect of soft pores, the remaining hard pores still satisfy the Gassmann equation after fluid saturation due to their incompressibility. At this time, the bulk modulus K _sat and shear modulus μ _sat when the hard pores are fully saturated can be calculated by the following formula:

μ _sat (p, ω) = μ _mf (p, ω).

The aspect ratio of the soft pores in the actual rock cannot be a fixed value, but is continuously distributed within a certain range. After obtaining the distribution data of the soft pore aspect ratio in the rock with respect to the porosity, if it is discretized, the influence of the soft pore jet flow can still be calculated by formulas for each discretized pore aspect ratio and corresponding porosity. Then the K _mf and μ _mf of the rock based on the pore distribution are calculated by iteratively adding the soft pores of each aspect ratio. In the process of adding soft pores iteratively, in addition to adding soft pores for the first time, the value calculated by adding soft pores for the kth time will be regarded as the K ^h and μ ^h added to the soft pores for the (k+1)th time; On the basis of K ^h , soft pores are added to obtain the (k+1)th K _d and μ _d . The entire process of adding soft pores can be expressed as:

The total dispersion of elastic waves in rocks can be regarded as the comprehensive cumulative effect of the dispersion of soft pores in various aspect ratios. It is generally believed that when the saturated fluid has a low viscosity, the jet flow acts at the logging frequency or higher ultrasonic frequency. Through the analysis of this article, it can be seen that considering the actual distribution characteristics of soft pores in rock media, the velocity value may have a non-negligible difference from the result of Gassmann equation even in the seismic frequency band. At the same time, there is a significant difference in velocity between the logging frequency band and seismic frequency .

Therefore, the velocity dispersion based on the jet stream mechanism is the main cause of velocity dispersion due to uneven pore stiffness (caused by pores or cracks with different surface rates and differences in the arrangement of pores or cracks). According to this model, the viscous non-"relaxed" part of the pore space (the unbalanced part of pore pressure) is equal to the plastic pores closed under high pressure, and the elastic modulus of dry rock under high pressure is approximately equal to that when the rock is saturated with fluid and undrained. Modulus. According to Betti's reciprocity theorem, the modulus of the non-"relaxed" part can be obtained when the plastic porosity is known, and the results are as follows:

In formulas (4) and (5), K _uf and μ _uf are the bulk modulus and shear modulus of high-frequency non-"relaxed" rock skeleton, respectively, and K _dry-hP is the bulk modulus of dry rock under higher pressure, φ _soft is the plastic porosity under a given pressure. Using formula (6), it can be obtained that the shear modulus and bulk modulus of saturated rock under different pressures have the following relationship:

(3) Acoustic curve dispersion correction based on rock physics model

According to the petrophysical model of micro-jet mechanism with multiple pore distribution constructed in step 2, the main model parameters that need to be input are the lithology, porosity, permeability and pore structure distribution of the reservoir rock.

The lithology of the reservoir rock is mainly judged by the GR (gamma) curve. The mudstone baseline in the gamma curve represents mudstone, and the sandstone with a large variation range. The law obtained from the low-frequency petrophysical test shows: mudstone pore structure Single, dry mudstone has basically no frequency dispersion, and there is no need to calibrate the speed from logging frequency band to seismic frequency band. However, the pore structure of tight reservoir sandstone is complicated. When soft pores (micro-fractures) are relatively developed, the dispersion amplitude is relatively large. There is a big difference between the well frequency band velocity and the seismic frequency band velocity, and the amount of dispersion correction is large.

The permeability of reservoir rocks is mainly calculated using the Kozeny-Carman relationship. The Kozeny-Carman formula is expressed as follows:

In the formula, τ is the pore curvature (the ratio of the flow distance between the two points in the pore and the straight line distance, usually taken as the fixed value τ = 3), and d is the particle size of the rock composition (the particle diameter needs to be determined as 80μm based on the CT scan image results ).

The method for determining the distribution of pore results is that because the reservoir rocks mainly undergo the same sedimentation and diagenesis process, it can be assumed that the pore structure has the same distribution characteristics, and the target interval of the studied reservoir and the test sample have the same pore structure distribution characteristics. Only make a proportional correction based on the relative porosity. Use logging data to obtain the pore distribution characteristics of the reservoir sandstone. It can be seen from the results of petrophysical experiments that the characteristic frequencies of tight sandstones in the study area are mainly located in seismic frequency bands, and fluid (water, oil) saturated rocks at logging frequencies represent high-frequency velocities. The velocity of the reservoir rock depends on the porosity and pore shape.

After the pore distribution characteristics of the reservoir section are obtained, the pore fluid-related velocity dispersion can be corrected according to the experimental dispersion law and the calculation formula of the petrophysical model modulus of multiple pore distribution.

The sonic logging curves of Well HG102 and Well HG Deviated 101 were corrected and matched with seismic data using this method. Due to the influence of velocity dispersion, the original sonic logging velocity curve is difficult to match the seismic velocity. The velocity dispersion correction method is used to reasonably correct the velocity of the logging frequency band to the seismic frequency band to achieve logging velocity and seismic velocity. As shown in Figure 3, after velocity dispersion correction, the original acoustic wave velocity curve generally moves to the low frequency band velocity on the left, and the velocity is reduced. Compared with the velocity matching correction method of the resonance Q model, the ES model of the present invention According to the different lithology and fluid properties, the correction amount is different. The mudstone correction amount is small, the velocity dispersion of the sandstone section of the reservoir is the largest, the correction amount is correspondingly larger, and the correction is more reasonable. The speed correction amount of the present invention and the resonance Q value speed correction amount have similar values in the reservoir section, both of which make the logging speed correction to low speed, but the correction amount is higher than the fracture content in the reservoir position with more fractures. Fewer reservoir sections show the effect of jet flow. The calibrated curve was calibrated by synthetic records to test the effect of well-seismic matching. The seismic wavelet adopts zero-phase wavelet. Figures 4 and 5 show the comparison of the synthetic record calibration of the original sonic log curve and the sonic curve after matching and correction of this patent. It can be seen that the synthetic record of the original sonic curve is compared with the seismic profile. Some reflection events in seismic records can correspond to the actual seismic profile, but there are also several sets of reflections that cannot correspond to the actual seismic profile, and the correlation coefficient of seismic calibration is not high. Using the result of calibration of the synthetic seismic record of the acoustic curve after the matching correction method of this patent, it can be seen that the synthetic seismic record after the speed dispersion correction of the acoustic curve improves the consistency with the well side channel, and improves the fine horizon calibration and The quality of the sub-layer comparison, the correlation coefficient of well-seismic calibration increased from 0.55 of the original well curve to 0.83 of the petrophysical model correction of multiple pore distribution, and the accuracy of well-seismic matching has been significantly improved. Compared with the original P-wave well curve, the velocity correction is deeper. -Time correspondence has been improved to a certain extent.

The logging and seismic velocity matching method based on seismic petrophysical experiment analysis of the present invention aims at improving the matching accuracy of sonic logging velocity and seismic velocity, and further improving the accuracy of synthetic record calibration, and proposes a rock physics based on seismic frequency band The sound wave velocity dispersion correction method of experimental analysis and petrophysical model, and the idea of iterative matching has further improved the accuracy of well-seismic matching. This method is based on the actual seismic frequency band petrophysical experiment results and petrophysical theoretical model. According to the changes of reservoir lithology, physical properties and fluid properties, the acoustic wave curve is mapped to the dispersion correction point by point, which has higher dispersion correction accuracy. , And then make synthetic records and compare them with side channels, and further iteratively modify the model parameters to make them more compatible with seismic records, which greatly improves the accuracy of well-seismic matching.

Claims (10)

1. The logging and seismic velocity matching method based on seismic rock physics experiment analysis is characterized in that the logging and seismic velocity matching method based on seismic rock physics experiment analysis includes:

   Step 1. Establish the velocity variation curve with frequency through the analysis of the rock physics experiment in the seismic frequency band, that is, the velocity dispersion variation law;

   Step 2: Construct a petrophysical model suitable for characterizing the dispersion characteristics of seismic frequency bands;

   Step 3. Use the rock physics model modulus calculation formula to perform the point-by-point mapping dispersion correction of the acoustic wave curve;

   Step 4: Perform synthetic record calibration on the sonic curve after dispersion correction, compare and analyze the side channel of the seismic well with the synthetic seismic record, and output the correction curve results for all depths of the entire sonic logging curve.
2. The method for matching logging and seismic velocity based on seismic petrophysical experiment analysis according to claim 1, characterized in that, in step 1, a target formation core is selected and a full-band petrophysical laboratory test is carried out to measure different lithology, The elastic parameters of physical properties and fluid-containing properties, including longitudinal wave velocity and shear wave velocity at different frequency points; based on laboratory test results, a relationship curve between velocity and frequency change is formed, and the velocity corresponding to the seismic frequency band and the velocity corresponding to the logging frequency band are quantitatively analyzed The difference between the logging and seismic velocities is estimated to obtain the correction amount.
3. The method for matching logging and seismic velocities based on seismic petrophysical experiment analysis according to claim 1, wherein in step 2, the rock with only dry hard pores is used as the new equivalent matrix, and its bulk modulus Is K ^stiff , replaced by K ^h , adding soft pores and considering the effect of the fluid jet in the soft pores and hard pores, the medium equivalent modulus K _mf and μ _mf are calculated by the following formula:

   

   

   Where ω is the circular frequency, η is the dynamic viscosity of the pore fluid, p is the effective pressure of the rock, α _c is the pore aspect ratio, φ _c (P) is the porosity of the soft pore under a certain effective pressure p, μ _d (p) are the dry volume and shear modulus when the rock medium contains soft pores with an aspect ratio of α _c and porosity φ _c (p) under a certain effective pressure; the second item on the right is the addition of a specific aspect ratio Soft pores change the bulk modulus K ^stiff under consideration of jetting action; after considering the effect of soft pores, the remaining hard pores still satisfy the Gassmann equation after fluid saturation due to their incompressibility. At this time, the volume of the hard pores is fully saturated The modulus K _sat and shear modulus μ _sat are calculated by the following formula:

   

   μ _sat (p, ω) = μ _mf (p, ω). (2)

   In the process of adding soft pores iteratively, in addition to adding soft pores for the first time, the value calculated by adding soft pores for the kth time will be regarded as the K ^h and μ ^h added to the soft pores for the (k+1)th time; On the basis of K ^h , soft pores are added to obtain the (k+1)th K _d and μ _d ; the entire process of adding soft pores is expressed as:

   

   

   According to the Betti reciprocity theorem, the modulus of the non-"relaxed" part is obtained when the plastic porosity is known, and the results are as follows:

   

   

   In formulas (4) and (5), K _uf and μ _uf are the bulk modulus and shear modulus of the high-frequency non-relaxed rock skeleton respectively. K _dry-hP is the bulk modulus of dry rock under higher pressure, φ _soft Is the plastic porosity at a given pressure; using formula (6), the shear modulus and bulk modulus of saturated rock under different pressures have the following relationship:

   
4. The method for matching logging and seismic velocity based on seismic petrophysical experiment analysis according to claim 1, characterized in that, in step 3, the petrophysical model modulus calculation formula is used, according to the porosity, permeability and galvanic uphole. The lithology and logging oil-water interpretation conclusions divided by the Ma curve, and the point-by-point mapping dispersion correction of the acoustic curve; according to the petrophysical model of the micro-jet mechanism with multiple pore distribution constructed in step 2, the main model parameters that need to be input It is the lithology, porosity, permeability and pore structure distribution of reservoir rocks.
5. The method for matching logging and seismic velocity based on seismic petrophysical experiment analysis according to claim 4, characterized in that, in step 3, the lithology of the reservoir rock is judged by a gamma curve, and the corresponding gamma curve The mudstone baseline represents mudstone, while sandstone has a large variation range. The law obtained from low-frequency petrophysical testing shows that the pore structure of mudstone is simple, and dry mudstone has basically no frequency dispersion, and there is no need to calibrate the logging frequency to seismic frequency. Tight reservoir sandstone has a complex pore structure. When the soft pores are relatively developed, the dispersion amplitude is relatively large, and the velocity in the logging frequency band is different from that in the seismic frequency band, and the dispersion correction amount is large.
6. The method for matching logging and seismic velocity based on seismic petrophysical experiment analysis according to claim 5, wherein in step 3, the permeability of the reservoir rock is calculated using the Kozeny-Carman relationship; the Kozeny-Carman formula is expressed as follows :

   

   In the formula, k is the permeability of the reservoir rock, τ is the pore curvature, d is the particle size of the constituent rock, and Φ is the porosity of the reservoir rock.
7. The method for matching logging and seismic velocity based on seismic petrophysical experiment analysis according to claim 6, characterized in that, in step 3, the pore result distribution is determined by obtaining the pore distribution characteristics of the reservoir sandstone through logging data Because the reservoir rocks mainly undergo the same sedimentation and diagenesis process, it is assumed that the pore structure has the same distribution characteristics, and it is assumed that the target interval of the studied reservoir and the test sample have the same pore structure distribution characteristics, and only the relative size of the porosity is made. Proportion correction; use logging data to obtain the pore distribution characteristics of reservoir sandstone; from the petrophysical experiment results, it can be seen that the characteristic frequency of tight sandstone in the study area is mainly located in the seismic frequency band, and fluid-saturated rock at logging frequency represents high-frequency velocity ; The velocity of reservoir rocks is determined by porosity and pore shape.
8. The method for matching logging and seismic velocity based on seismic petrophysical experiment analysis according to claim 7, characterized in that, in step 3, after obtaining the pore distribution characteristics of the reservoir section, the petrophysical model in step 2 is used to compare the pores The fluid-related velocity dispersion is corrected; the correction amount in the reservoir with more fractures is higher than that in the reservoir with less fractures, showing the effect of jet flow.
9. The method of logging and seismic velocity matching based on seismic petrophysical experiment analysis according to claim 1, wherein in step 4, synthetic recording calibration is performed on the acoustic curve after dispersion correction, and the side channel of the seismic well is Synthetic seismic records are compared and analyzed. If the matching degree is high and the correlation coefficient is more than 80%, the sound wave correction speed of the well will be directly output. If the matching degree is not high, return to the petrophysical model parameters for parameter correction, and then perform the sound wave curve Point-by-point dispersion correction, until the matching correlation coefficient between the side channel of the seismic well and the synthetic seismic record reaches 80%, iterative correction is terminated.
10. The method of logging and seismic velocity matching based on seismic petrophysical experiment analysis according to claim 9, characterized in that, in step 4, the modified parameters include the aspect ratio of hard pores and the distribution range of aspect ratio of soft pores.

Images