Nuclear magnetic resonance logging double-TE observation mode data processing and saturation model construction method

Through the data processing method of nuclear magnetic resonance dual TE observation mode, the particle swarm inversion and least squares method are used to derive fluid information and build a saturation model, which solves the problem of inaccurate analysis of small differences in fluid diffusion and relaxation in the prior art. Accurate calculation of the reservoir hydrocarbon saturation saturation is achieved.

CN119001877APending Publication Date: 2024-11-22CHINA UNIV OF GEOSCIENCES (BEIJING)
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411077499.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

The existing nuclear magnetic resonance double TE logging has a small qualitative analysis of the fluid diffusion and relaxation difference, and the hydrocarbon volume of the reservoir flushing belt cannot be accurately calculated when the temperature, viscosity and diffusion coefficient of the reservoir fluid are unknown.

Method used

Using the data processing method of nuclear magnetic resonance dual TE observation mode, the equations of the amplitude difference of the echo string, the fluid type, fluid content and relaxation parameters are derived, and the particle swarm inversion method is used to obtain the fluid. The diffusion coefficient and relaxation parameters are further obtained by linear inversion of the least squares method to obtain hydrocarbon-containing information, and a saturation model is constructed to calculate the hydrocarbon-containing saturation of the reservoir flushing belt.

Benefits of technology

Accurate identification and quantitative calculations with small differences in fluid diffusion and relaxation are achieved, and the calculation shortcomings of the prior art under unknown conditions are overcome, and the accuracy of reservoir fluid recognition and satisfaction of production applications are improved.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119001877A_ABST
    Figure CN119001877A_ABST
Patent Text Reader

Abstract

The invention relates to a nuclear magnetic resonance logging dual-TE observation mode data processing and saturation model construction method, and relates to the technical field of oil-gas exploration, and the method comprises the steps: extracting an echo string through a nuclear magnetic resonance dual-TE logging mode; according to a nuclear magnetic resonance double-TE logging relaxation mechanism, deducing an equation of an echo string amplitude difference, a fluid type, a fluid content and a fluid relaxation parameter; performing nonlinear inversion on the constructed target function based on a particle swarm inversion method to obtain a diffusion coefficient and a relaxation parameter of the fluid, and generating an echo string difference reflecting fluid information; performing linear inversion on the generated echo string difference through a least square method to obtain a difference spectrum reflecting hydrocarbon-containing information; and constructing a saturation model, and calculating the hydrocarbon-containing volume and saturation of the reservoir flushing zone according to the saturation model. According to the method, hydrocarbon-containing information can be focused through a double-TE differential spectrum method, and the hydrocarbon-containing saturation of the reservoir flushing zone is calculated, so that qualitative identification and quantitative calculation of reservoir fluid are realized.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploration, and in particular to a method for processing data in a nuclear magnetic resonance logging dual TE observation mode and constructing a saturation model. Background Art

[0002] In the field of oil and gas exploration and development, nuclear magnetic resonance (NMR) logging is an important geophysical logging technology that can provide parameters such as total porosity, effective porosity, movable fluid and bound fluid in the formation, and plays an important role in the exploration and development of oil fields. Based on the difference in the longitudinal relaxation time T1 between reservoir water and hydrocarbons, the dual TW method was developed, which mainly includes the difference spectrum method and the time domain analysis method. Based on the difference in the diffusion coefficient D of each phase fluid, the dual TE method, that is, the quantitative diffusion analysis method, was developed. In the current oil and gas exploration, the use of nuclear magnetic resonance dual TE logging is increasing, especially in the identification of gas layers, which has unique advantages. In the nuclear magnetic resonance dual TE logging observation and data analysis, when the rock is saturated with fluids with different diffusion coefficients, due to the influence of diffusion relaxation, the T2 distribution corresponding to the long echo interval is more forward than the T2 distribution of the short echo interval. Among them, the diffusion coefficient of gas is the largest. When saturated with gas, the forward shift of T2 distribution is more obvious; the diffusion coefficients of water and light oil are relatively small, and the forward shift of T2 distribution is small. The analysis methods of dual TE observation data in the prior art include: the spectrum shift qualitative analysis method (SSM) based on dual TE logging, which has a better recognition effect when the difference in fluid diffusion relaxation is large; the qualitative analysis method of constructing water spectrum based on dual TE logging, which has a better effect when the difference in fluid diffusion relaxation is large; the quantitative diffusion analysis method (DIFAN) based on dual TE logging, which calculates the gas saturation of the flushing zone of the reservoir using the intersection diagram constructed by 1 / T2int-Da / Dw, and has a better effect when the temperature, viscosity and diffusion coefficient of the reservoir fluid are known.

[0003] With the existing fluid identification technology in oil and gas exploration, the nuclear magnetic resonance dual TE logging mode is increasingly used, such as the single TW dual TE mode and the dual TW dual TE mode. The original qualitative and quantitative interpretation methods of nuclear magnetic resonance dual TE logging, such as the spectrum shift and diffusion analysis methods, can no longer meet the production needs. The above methods still have the following technical defects in practical applications: when the difference in fluid diffusion relaxation is small, the difference in T2 distribution of dual TE is not obvious, and its qualitative analysis method is not accurate enough; and, when the temperature, viscosity and diffusion coefficient of the reservoir fluid are unknown, the DIFAN method cannot accurately and quantitatively calculate the hydrocarbon-containing volume of the reservoir flushing zone.

[0004] Therefore, it is necessary to develop new data processing methods and reservoir parameter calculation methods for the dual TE mode of nuclear magnetic resonance logging, and to develop a data processing and saturation model construction method for the dual TE observation mode of nuclear magnetic resonance logging to solve the above technical problems. Summary of the invention

[0005] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for processing data in the dual TE observation mode of nuclear magnetic resonance logging and building a saturation model. The method can focus on hydrocarbon information through the dual TE difference spectrum method and calculate the hydrocarbon saturation of the reservoir flushing zone, thereby realizing qualitative identification and quantitative calculation of reservoir fluids.

[0006] In order to solve the above technical problems, the basic concept of the technical solution adopted by the present invention is:

[0007] In a first aspect, a method for processing data in a dual TE observation mode of nuclear magnetic resonance logging and constructing a saturation model comprises:

[0008] Extract echo trains through NMR dual TE logging mode;

[0009] According to the relaxation mechanism of NMR dual TE logging, the equations of echo train amplitude difference, fluid type, fluid content and fluid relaxation parameters are derived.

[0010] Based on the particle swarm inversion method, the constructed objective function is nonlinearly inverted to obtain the diffusion coefficient and relaxation parameter of the fluid, and the echo string difference reflecting the fluid information is generated;

[0011] The generated echo train difference is linearly inverted by the least square method to obtain the difference spectrum reflecting the hydrocarbon-containing information;

[0012] A saturation model is constructed, and the hydrocarbon-containing volume and saturation of the reservoir flushing zone are calculated based on the saturation model.

[0013] In a preferred embodiment of any of the above schemes, during the waiting time, a long echo interval and a short echo interval are respectively used for measurement to obtain two groups of echo trains.

[0014] In a preferred embodiment of any of the above schemes, in the NMR dual TE logging relaxation mechanism, assuming that the reservoir contains only gas and water phases, the echoes measured in the short echo interval mode include: surface relaxation and volume relaxation; the magnetization intensity at time t under long waiting time and short echo interval:

[0015]

[0016] The echoes measured in the long echo interval mode include: surface relaxation, volume relaxation and diffusion relaxation; the magnetization intensity at time t under long waiting time and long echo interval is:

[0017]

[0018] t=T EL *i,i=1,2,3,...,m S ;

[0019] Where, T WL is the long waiting time; T ES is the short echo interval; T EL is the long echo interval; is the magnetization intensity at time t under long waiting time and short echo interval, A / m; is the magnetization intensity at time t under long waiting time and long echo interval, A / m; M 0water,j is the magnetization intensity of the jth component water, A / m; M gas is the magnetic intensity of the gas, A / m; T 2water,j is the transverse relaxation time of water of the jth component, ms; T 2,gas is the transverse relaxation time of gas, ms; T 1,gas is the longitudinal relaxation time of gas, ms; T 1,water is the longitudinal relaxation time of water, ms. is the transverse relaxation time of water diffusion, ms; T 2D,gas is the transverse relaxation time of gas diffusion, ms.

[0020] In any of the above preferred embodiments, the transverse relaxation time of water diffusion is and the transverse relaxation time T of gas diffusion 2D,gas The calculation formulas are:

[0021]

[0022] Where D water is the diffusion coefficient of water, cm 2 / ms;D gas is the diffusion coefficient of gas, cm 2 / ms; G is the magnetic field gradient value, Gs / cm; is the gyromagnetic ratio of hydrogen protons.

[0023] In a preferred embodiment of any of the above schemes, two sets of echo string data with different echo intervals are normalized, and the echo string with long echo interval is used as the standard. At the same time t, the formula for thinning out the echo string with short echo interval is:

[0024] When the influence of water is eliminated, the above formula is transformed into:

[0025]

[0026] In a preferred embodiment of any of the above schemes, the calculation formula after eliminating the influence of water signal is:

[0027]

[0028] Where, ΔM is the echo string difference;

[0029] when T 1,gas , T 2,gas When it is known that the calculation formula after eliminating the influence of the water signal is linear, the calculation formula after eliminating the influence of the water signal is linearly inverted according to the least square method or the SVD method to obtain a double TE difference spectrum containing reservoir fluid information;

[0030] when T 1,gas , T 2,gas When it is unknown, the calculation formula after eliminating the influence of water signal is nonlinear, and nonlinear inversion is performed on the calculation formula after eliminating the influence of water signal, and the particle swarm algorithm is used to search for the optimal gas and water diffusion coefficients and relaxation parameters.

[0031] In a preferred embodiment of any of the above schemes, when a particle swarm algorithm is used to search for the optimal gas and water diffusion coefficients and relaxation parameters, the fitness function inverted by the particle swarm algorithm is:

[0032]

[0033] The calculation formula for gas saturation in the reservoir flushing zone is:

[0034]

[0035] Where f is the residual sum of squares; Δφ(0) is the porosity difference calculated by the CPMG sequence acquisition of double TE echo trains; φ gas is the gas-bearing porosity of the formation flushing zone; φ is the formation porosity; HI g It is the hydrogen content index of the gas flushed into the formation.

[0036] In the second aspect, a nuclear magnetic resonance logging dual TE observation mode data processing and saturation model building system includes:

[0037] An acquisition module, used for extracting echo trains through a nuclear magnetic resonance dual TE logging mode;

[0038] The processing module is used to derive the equation of echo train amplitude difference and fluid type, fluid content and fluid relaxation parameter according to the relaxation mechanism of nuclear magnetic resonance dual TE logging; perform nonlinear inversion on the constructed objective function based on the particle swarm inversion method to obtain the diffusion coefficient and relaxation parameter of the fluid, and generate echo train difference reflecting the fluid information; perform linear inversion on the generated echo train difference by the least square method to obtain the difference spectrum reflecting the hydrocarbon information; construct a saturation model, and calculate the hydrocarbon volume and saturation of the reservoir flushing zone according to the saturation model.

[0039] According to a third aspect, a computing device is provided, comprising:

[0040] one or more processors;

[0041] The storage device is used to store one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described above.

[0042] In a fourth aspect, a computer-readable storage medium is provided, wherein a program is stored in the computer-readable storage medium, and when the program is executed by a processor, the method described above is implemented.

[0043] After adopting the above technical solution, the present invention has the following beneficial effects compared with the prior art:

[0044] The present invention extracts the following according to the nuclear magnetic resonance dual TE logging mode: a long waiting time short echo interval group and a long waiting time long echo interval group as echo trains, and according to the nuclear magnetic resonance dual TE logging relaxation mechanism, derives the equation of echo train amplitude difference and fluid type, fluid content and fluid relaxation parameter, and performs nonlinear inversion on the constructed objective function based on the particle swarm inversion method to obtain the diffusion coefficient and relaxation parameter of the fluid, and generates echo train differences reflecting fluid information. Secondly, the present invention further uses the least squares method to perform linear inversion on the generated echo train differences to obtain a difference spectrum reflecting hydrocarbon information. Finally, the present invention calculates the hydrocarbon volume and saturation of the reservoir flushing zone according to the constructed saturation model.

[0045] The specific implementation modes of the present invention are further described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. Some specific embodiments of the present application will be described in detail in an illustrative and non-restrictive manner with reference to the drawings. The same reference numerals in the drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the drawings:

[0047] Figure 1 It is a schematic diagram of the data processing and saturation model building method of the nuclear magnetic resonance logging dual TE observation mode of the present invention.

[0048] Figure 2 It is a schematic diagram of the steps of data processing and saturation model construction in the dual TE observation mode of nuclear magnetic resonance logging in Example 2 of the present invention.

[0049] Figure 3 This is a numerical simulation diagram of rock models with different gas saturations in Example 2 of the present invention;

[0050] Figure 4 is a comparison diagram of the inversion results of the nuclear magnetic resonance double TE difference spectrum in Example 2 of the present invention;

[0051] Figure 5 is a comparison diagram between the gas saturation result calculated in Example 2 of the present invention and the model;

[0052] Figure 6 This is a diagram of the nuclear magnetic resonance logging interpretation results of Well H21 in Example 3 of the present invention;

[0053] Figure 7 It is a schematic diagram of the data processing and saturation model building system of the dual TE observation mode of nuclear magnetic resonance logging in Example 4 of the present invention.

[0054] It should be noted that these drawings and text descriptions are not intended to limit the conceptual scope of the present invention in any way, but to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. The elements in the drawings are schematic and not drawn to scale. DETAILED DESCRIPTION

[0055] In order to enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present application.

[0056] The following embodiment of the present application takes the data processing and saturation model construction method of the dual TE observation mode of nuclear magnetic resonance logging as an example to explain the scheme of the present application in detail, but this embodiment cannot limit the protection scope of the present application.

[0057] Embodiment 1:

[0058] The physical basis of nuclear magnetic resonance logging technology is the magnetism of the atomic nucleus and its interaction with the external magnetic field. The original data measured is a relaxation decay curve composed of hundreds of spin echoes. The spin echo train obtained by measuring the transverse relaxation time of the CPMG sequence does not decay in a single exponential manner, but is the sum of multiple exponential decays.

[0059] like Figure 1 As shown, the present invention provides a method for processing data in a dual TE observation mode of nuclear magnetic resonance logging and building a saturation model, comprising:

[0060] Step 11, extracting echo trains through nuclear magnetic resonance dual TE logging mode;

[0061] Step 12, according to the relaxation mechanism of nuclear magnetic resonance dual TE logging, derive the equation of echo train amplitude difference and fluid type, fluid content and fluid relaxation parameter;

[0062] Step 13, performing nonlinear inversion on the constructed objective function based on the particle swarm inversion method to obtain the diffusion coefficient and relaxation parameter of the fluid, and generating an echo train difference reflecting the fluid information;

[0063] Step 14, performing linear inversion on the generated echo train difference by the least square method to obtain a difference spectrum reflecting hydrocarbon-containing information;

[0064] Step 15: construct a saturation model, and calculate the hydrocarbon-containing volume and saturation of the reservoir flushing zone according to the saturation model.

[0065] In an embodiment of the present invention, according to the nuclear magnetic resonance dual TE logging mode, a long waiting time short echo interval group and a long waiting time long echo interval group are extracted as echo trains. According to the nuclear magnetic resonance dual TE logging relaxation mechanism, the equation of the echo train amplitude difference and the fluid type, fluid content and fluid relaxation parameter is derived. Based on the particle swarm inversion method, the constructed objective function is nonlinearly inverted to obtain the diffusion coefficient and relaxation parameter of the fluid, and an echo train difference reflecting the fluid information is generated. Secondly, the present invention further uses the least squares method to perform linear inversion on the generated echo train difference to obtain a difference spectrum reflecting the hydrocarbon information. Finally, the present invention calculates the hydrocarbon volume and saturation of the reservoir flushing zone according to the constructed saturation model.

[0066] Embodiment 2:

[0067] like Figure 2 As shown, based on the technical solution of Example 1, this embodiment constructs the data processing and saturation model of the NMR logging dual TE observation mode according to the following steps:

[0068] Step 1: Generate an echo train (a long waiting time and short echo interval group, a long waiting time and long echo interval group) according to the nuclear magnetic resonance dual TE logging mode.

[0069] Step 2: Based on the relaxation mechanism of NMR dual TE logging, derive the equations of echo train amplitude difference and fluid type, fluid content and fluid relaxation parameters.

[0070] Step 3: Based on the particle swarm inversion method, the constructed objective function is nonlinearly inverted to obtain the diffusion coefficient and relaxation parameter of the fluid, and the echo train difference reflecting the fluid information is generated. The generated echo train difference is linearly inverted using the least squares method to obtain the difference spectrum reflecting the hydrocarbon information.

[0071] Step 4: Calculate the gas volume and gas saturation of the reservoir flushing zone based on the constructed saturation model and apply it in the study area.

[0072] The specific operation process is as follows:

[0073] In the dual TE observation mode of NMR logging, at long polarization time (T WL ), long and short echo intervals (T EL , T ES ) to obtain two sets of echo trains. According to the relaxation mechanism of nuclear magnetic resonance, under long TE conditions, that is, when TE is 3.6ms, the influence of diffusion relaxation is large and cannot be ignored. In general, when the reservoir is hydrophilic and the reservoir contains only gas and water phases, the echoes measured in the short echo interval mode include: surface relaxation and volume relaxation; the echoes measured in the long echo interval mode include:

[0074] Surface relaxation, bulk relaxation and diffusion relaxation; long waiting time and short echo interval (T ES , T EL )The magnetization intensity at time t:

[0075]

[0076]

[0077] In the formula, is the magnetization intensity at time t under long waiting time and short echo interval, A / m; is the magnetization intensity at time t under long waiting time and long echo interval, A / m; M 0water,j is the magnetization intensity of the jth component water, A / m; M gas is the magnetic intensity of the gas, A / m; T 2water,j is the transverse relaxation time of water of the jth component, ms; T 2,gas is the transverse relaxation time of gas, ms; T 1,gas is the longitudinal relaxation time of gas, ms; T 1,water is the longitudinal relaxation time of water, ms. is the transverse relaxation time of water diffusion, ms; T 2D,gas is the transverse relaxation time of gas diffusion, ms.

[0078] T 2D,water , T 2D,gas It can be expressed as:

[0079]

[0080] Where D water is the diffusion coefficient of water, cm 2 / ms;D gas is the diffusion coefficient of gas, cm 2 / ms; G is the magnetic field gradient value, Gs / cm; is the gyromagnetic ratio of hydrogen protons.

[0081] The two sets of echo string data with long and short echo intervals are normalized. In order to perform subtraction, the echo string with long echo interval is taken as the standard. At the same time t, the echo string with short echo interval is thinned out. Then formula (1) can be transformed into:

[0082]

[0083] In order to eliminate the influence of water, equation (4) can be transformed into:

[0084]

[0085] Subtracting equation (4) from equation (2) can eliminate the influence of water signal:

[0086]

[0087] Formula (5) can be transformed into:

[0088]

[0089] when T 1,gas , T 2,gas When known, the least squares or SVD method is used to perform linear inversion on equation (7) to obtain the double TE difference spectrum, which contains reservoir fluid information. However, the diffusion coefficients of oil and water in the reservoir are unknown, and equation (7) is nonlinear. It is necessary to perform nonlinear inversion to search for the optimal diffusion coefficients of gas and water and relaxation parameters.

[0090] In the embodiment of the present invention, the particle swarm optimization (PSO) is an optimization algorithm based on swarm intelligence, which was proposed by Kennedy and Eberhart in 1995. The PSO algorithm simulates the collective behavior of swarms of organisms such as flocks of birds or schools of fish, and searches for the optimal solution through group collaboration and information exchange. In PSO, each solution in the solution space is called a particle, and the particle searches for the optimal solution by moving in the solution space.

[0091] First, randomly generate the initial position x of each particle id and speed v id , and then set the historical best position p of each particle id Its initial position, the global optimal position p d is the initial position with the best fitness value in the particle swarm. According to formula (8), the fitness value f(x id ), and update the particle's historical best position p id and the global optimal position p d .

[0092] According to formula (6), the fitness function of particle swarm algorithm inversion can be obtained:

[0093]

[0094] In the formula, f is the residual sum of squares.

[0095] In PSO, each particle represents a solution in the solution space and searches for the optimal solution by adjusting its position and velocity. In each iteration, the velocity and position of the particle are updated by the following formula:

[0096]

[0097] Where i is the particle number, d is the particle number dimension, k is the number of iterations, ω is the inertia weight, c1 is the individual learning factor, c2 is the group learning factor, is the velocity vector of particle i in the kth iteration, is the position vector of particle i in the dth dimension in the kth iteration, is the historical optimal position of particle i in the dth dimension in the kth iteration, that is, the optimal solution obtained by the i-th particle after k iterations. is the historical optimal position of the swarm in the dth dimension in the kth iteration, that is, the optimal solution obtained by the entire particle swarm search after k iterations; r1, r2 are random numbers between [0,1].

[0098] If the current position of a particle is better than its historical best position, the historical best position of the particle is updated. If the current position of a particle is better than the global best position, the global best position is updated. When the predetermined number of iterations is reached or the change value of the fitness function is less than the predetermined threshold, the algorithm terminates, and the particle swarm algorithm searches for the optimal gas and water diffusion coefficients and relaxation parameters. Substitute it into formula (7) to obtain the echo string difference ΔM, and use the least squares method to perform linear inversion to obtain the double TE difference spectrum, which contains reservoir fluid information.

[0099] In the specific implementation process of the present invention, in nuclear magnetic resonance logging, the magnetization intensity can be converted into porosity. Considering the hydrogen index (HI) correction, equation (6) can be converted into:

[0100]

[0101] Where Δφ(t) is the porosity difference calculated by acquiring double TE echo trains using CPMG sequence; gas HI is the gas-bearing porosity of the formation flushing zone; g It is the hydrogen content index of the gas flushed into the formation.

[0102] The above formula can be converted to:

[0103]

[0104] Finally, the gas saturation of the reservoir flushing zone is obtained:

[0105]

[0106] In order to verify the accuracy of the above method and technology, a gas-bearing reservoir volume model was set up and tested using numerical simulation methods. Its porosity is 9% (such as Figure 3 As shown), the gas saturation is 0, 20%, 40%, 60%, and 80% respectively. According to rock physics experiments, the transverse relaxation time of gas is set to 45ms, the longitudinal relaxation time of gas is set to 4500ms, the longitudinal relaxation time of water is set to 500ms, the diffusion coefficient of gas is set to 80×10-8cm2 / ms, the diffusion coefficient of water is set to 1.5×10-8cm2 / ms, the hydrogen index of gas is set to 0.4, and the hydrogen index of water is set to 1. The long waiting time is set to 12.98s, the short waiting time is set to 1.0s, the short echo interval is set to 0.9ms, and the long echo interval is set to 3.6ms. Using the preset model, given parameters, and gas-water content, according to the dual TE response mechanism of nuclear magnetic logging, the dual TE observation echo string is generated by forward modeling through equations (1) and (2).

[0107] According to the particle swarm algorithm, for the generated double TE observation echo train, the diffusion coefficient of gas is searched to be 83×10-8cm2 / ms, the diffusion coefficient of water is 2×10-8cm2 / ms, the transverse relaxation time of gas is 49ms, and the longitudinal relaxation time of gas is 4800ms, with high search accuracy. Based on the searched optimal fluid relaxation parameters and diffusion coefficients, the echo train difference ΔM is generated. Figure 3 and Figure 4 In order to compare the difference spectrum results of the constructed rock model inversion, the generated echo string difference is inverted using the least squares method to obtain the difference spectrum (such as Figure 4 As shown in Figure 2, with the increase of gas saturation, the distribution amplitude of T2 becomes larger and moves in the direction of T2 decreasing.

[0108] At the same time, the corresponding gas saturation is calculated and compared with the model saturation. Figure 5To compare the calculated gas saturation results with the model, under the dual TE observation mode, the gas saturation calculated by the diffusion analysis method (DIFAN) is quite different from the model, and the gas saturation calculated by the new method is consistent with the model. The gas saturation calculated by the new method is better than the DIFAN method. Table 1 shows the absolute error comparison results of different saturation models. The average absolute error of the gas saturation calculated by the DIFAN method is 7.8%, which does not meet the production needs. The average absolute error of the gas saturation calculated by the new method is 4.2%, and the error between the calculated gas volume and the model is the smallest, which verifies that the method is correct and meets the production needs.

[0109]

[0110] Table 1 Comparison of absolute errors of different saturation models

[0111] Embodiment 3:

[0112] Taking a tight sandstone formation in the East China Sea Shelf Basin as an example, the effect of the data processing and saturation model of the NMR logging dual TE observation mode of the present invention is further verified:

[0113] The target layer is a dense sandstone with a burial depth of 3000-5000m. The reservoir is mainly a gas-bearing layer. The formation water salinity varies widely. The resistivity reflects the weak and unclear gas content, and the logging interpretation conformity rate is low. Therefore, the MRT instrument of China Petroleum Logging Group Co., Ltd. is used for nuclear magnetic resonance logging, and the observation mode is D9TWE3. The above method is used for data processing to obtain the difference spectrum, gas volume in the flushing zone, and gas saturation.

[0114] Figure 6 This is the result of the interpretation of the nuclear magnetic resonance logging of Well H21. Track 3 is the resistivity, Track 4 is the T2 distribution under long waiting time and short echo interval, Track 5 is the T2 distribution under long waiting time and long echo interval, Track 6 is the inverted differential spectrum, Track 7 is the calculated gas volume of the flushing zone after inversion, Track 8 is the calculated gas saturation of the flushing zone, Track 9 is the fluid volume analysis, and Track 10 is the formation analysis. At 3350-3361m, the resistivity is about 18Ω.m, the inverted differential spectrum signal is obvious, and it is distributed in the direction of T2 decrease. The calculated differential spectrum porosity (△φ) is about 5%, and the gas saturation of the flushing zone is about 40%, which is interpreted as a gas layer. The oil test result is a daily gas production of 173163m 3 , the daily water production is 0, and the gas layer is displayed, verifying that the fluid identification is accurate.

[0115] In addition, the above method was used to process and interpret data from 9 layers in 4 other wells in the study area, and the interpretation results were consistent with the oil test results by 88.9%, which shows that the above method is accurate and feasible, and the effect is very significant.

[0116] Embodiment 4:

[0117] As shown in FIG. 7 , this embodiment further provides a nuclear magnetic resonance logging dual TE observation mode data processing and saturation model building system 20, comprising:

[0118] An acquisition module 21 is used to extract echo trains through a nuclear magnetic resonance dual TE logging mode;

[0119] The processing module 22 is used to derive the equation of echo string amplitude difference and fluid type, fluid content and fluid relaxation parameter based on the nuclear magnetic resonance dual TE logging relaxation mechanism; perform nonlinear inversion on the constructed objective function based on the particle swarm inversion method to obtain the diffusion coefficient and relaxation parameter of the fluid, and generate echo string difference reflecting the fluid information; perform linear inversion on the generated echo string difference by the least squares method to obtain the difference spectrum reflecting the hydrocarbon information; construct a saturation model, and calculate the hydrocarbon volume and saturation of the reservoir flushing zone based on the saturation model.

[0120] The embodiment of the present invention further provides a computing device, comprising: a processor, a memory storing a computer program, wherein when the computer program is executed by the processor, the method described above is executed. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0121] The embodiment of the present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above. All implementations in the above method embodiment are applicable to this embodiment and can achieve the same technical effect.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for processing data in dual TE observation mode of nuclear magnetic resonance logging and constructing a saturation model, characterized in that: include: Extract echo trains through NMR dual TE logging mode; According to the relaxation mechanism of NMR dual TE logging, the equations of echo train amplitude difference, fluid type, fluid content and fluid relaxation parameters are derived. Based on the particle swarm inversion method, the constructed objective function is nonlinearly inverted to obtain the diffusion coefficient and relaxation parameter of the fluid, and the echo string difference reflecting the fluid information is generated; The generated echo train difference is linearly inverted by the least square method to obtain the difference spectrum reflecting the hydrocarbon-containing information; A saturation model is constructed, and the hydrocarbon-containing volume and saturation of the reservoir flushing zone are calculated based on the saturation model.

2. The method for data processing and saturation model construction of NMR logging dual TE observation mode according to claim 1, characterized in that: In the nuclear magnetic resonance dual TE logging mode, under the waiting time, long echo interval and short echo interval are respectively used for measurement to obtain two groups of echo trains.

3. The method for data processing and saturation model construction of NMR logging dual TE observation mode according to claim 2, characterized in that: In the relaxation mechanism of NMR dual TE logging, assuming that the reservoir contains only gas and water phases, the echoes measured in the short echo interval mode include: surface relaxation and volume relaxation; the magnetization intensity at time t under long waiting time and short echo interval is: t=T ES *i,i=1,2,3,...,m L The echoes measured in the long echo interval mode include: surface relaxation, volume relaxation and diffusion relaxation; the magnetization intensity at time t under long waiting time and long echo interval is: t=T EL *i,i=1,2,3,...,m S ; Where, T WL is the long waiting time; T ES is the short echo interval; T EL is a long echo interval; is the magnetization intensity at time t under long waiting time and short echo interval, A / m; is the magnetization intensity at time t under long waiting time and long echo interval, A / m; M 0water,j is the magnetization intensity of the jth component water, A / m; M gas is the magnetic intensity of the gas, A / m; T 2water,j is the transverse relaxation time of water of the jth component, ms; T 2,gas is the transverse relaxation time of gas, ms; T 1,gas is the longitudinal relaxation time of gas, ms; T 1,water is the longitudinal relaxation time of water, ms. is the transverse relaxation time of water diffusion, ms; T 2D,gas is the transverse relaxation time of gas diffusion, ms.

4. The method for data processing and saturation model building in dual TE observation mode of nuclear magnetic resonance logging according to claim 3, characterized in that: Transverse relaxation time of water diffusion and the transverse relaxation time T of gas diffusion 2D,gas The calculation formulas are: Where D water is the diffusion coefficient of water, cm 2 / ms;D gas is the diffusion coefficient of gas, cm 2 / ms; G is the magnetic field gradient value, Gs / cm; γ is the gyromagnetic ratio of hydrogen protons.

5. The method for data processing and saturation model building in dual TE observation mode of nuclear magnetic resonance logging according to claim 4, characterized in that: The two sets of echo string data with different echo intervals are normalized, and the echo string with long echo interval is taken as the standard. At the same time t, the formula for thinning out the echo string with short echo interval is: To eliminate the influence of water, transform the above formula into:

6. The method for data processing and saturation model building in dual TE observation mode of nuclear magnetic resonance logging according to claim 5, characterized in that: The calculation formula after eliminating the influence of water signal is: Where, ΔM is the echo string difference; when T 1,gas , T 2,gas When it is known that the calculation formula after eliminating the influence of the water signal is linear, the calculation formula after eliminating the influence of the water signal is linearly inverted according to the least square method or the SVD method to obtain a double TE difference spectrum containing reservoir fluid information; when T 1,gas , T 2,gas When it is unknown, the calculation formula after eliminating the influence of water signal is nonlinear, and nonlinear inversion is performed on the calculation formula after eliminating the influence of water signal, and the particle swarm algorithm is used to search for the optimal gas and water diffusion coefficients and relaxation parameters.

7. The method for data processing and saturation model building in dual TE observation mode of nuclear magnetic resonance logging according to claim 6, characterized in that: When the particle swarm algorithm is used to search for the optimal gas and water diffusion coefficients and relaxation parameters, the fitness function of the particle swarm algorithm inversion is: The calculation formula for gas saturation in the reservoir flushing zone is: Where f is the residual sum of squares; Δφ(0) is the porosity difference calculated by the CPMG sequence acquisition of double TE echo trains; φ gas is the gas-bearing porosity of the formation flushing zone; φ is the formation porosity; HI g It is the hydrogen content index of the gas flushed into the formation.

8. A nuclear magnetic resonance logging dual TE observation mode data processing and saturation model building system, characterized by: include: An acquisition module, used for extracting echo trains through a nuclear magnetic resonance dual TE logging mode; The processing module is used to derive the equation of echo train amplitude difference and fluid type, fluid content and fluid relaxation parameter according to the relaxation mechanism of nuclear magnetic resonance dual TE logging; perform nonlinear inversion on the constructed objective function based on the particle swarm inversion method to obtain the diffusion coefficient and relaxation parameter of the fluid, and generate echo train difference reflecting the fluid information; perform linear inversion on the generated echo train difference by the least square method to obtain the difference spectrum reflecting the hydrocarbon information; construct a saturation model, and calculate the hydrocarbon volume and saturation of the reservoir flushing zone according to the saturation model.

9. A computing device, characterized in that include: one or more processors; A storage device, used for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program, which, when executed by a processor, implements the method according to any one of claims 1 to 7.