A method and device for joint inversion of multi-twist, multi-torque-efficiency nuclear magnetic resonance logging observation data

Abstract

The application discloses a nuclear magnetic resonance well logging multi-TW multi-TE observation data joint inversion method and device, and the method comprises the following steps: generating a plurality of groups of first echo train data according to nuclear magnetic resonance multi-TW multi-TE well logging data; constructing a joint equation based on the plurality of groups of first echo train data, and recombining the first echo train data into second echo train data; and constructing a signal matrix based on the second echo train data to perform joint inversion on the second echo train data. According to the method, different echo train data obtained in the multi-TW multi-TE mode is recombined, a joint inversion result of a plurality of echo data is obtained through a new joint inversion method, and finally, the multi-solution problem is eliminated.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration technology, and in particular to a method and apparatus for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging. Background Technology

[0002] Nuclear magnetic resonance (NMR) logging is an important geophysical logging technology that uses the interaction between hydrogen protons and an applied magnetic field to observe reservoir information. It can provide parameters such as total formation porosity, effective porosity, mobile fluids, and bound fluids, playing a crucial role in oilfield exploration and development. To facilitate fluid identification, NMR logging has developed two observation modes: long and short waiting times (dual TW) and long and short echo intervals (dual TE). In current oil and gas exploration, the simultaneous measurement of multiple observation modes is increasingly common. For example, the DTWE2 mode of the MRIL-P instrument and the D9TWE3 mode of the MRT instrument can obtain multi-TW and multi-TE echo data, including long waiting time and short echo interval groups, short waiting time and short echo interval groups, long waiting time and long echo interval groups, and short waiting time and long echo interval groups.

[0003] In nuclear magnetic resonance dual-TW (TW) logging and its analysis, based on the different polarizabilities of fluids at different waiting times—that is, oil, gas, and water are fully polarized during long waiting times, while only water is fully polarized during short waiting times—qualitative and quantitative evaluation of reservoir fluids can be achieved by comparing and analyzing the nuclear magnetic resonance logging responses at different waiting times. In nuclear magnetic resonance dual-TE (TTE) logging and its data analysis, due to the influence of diffusion relaxation, the T2 distribution corresponding to longer echo intervals shifts forward compared to shorter echo intervals, and the forward shift is more pronounced for fluids with larger diffusion coefficients. Current processing methods involve analyzing dual-TW and dual-TE logging data separately. Specifically:

[0004] (1) Based on dual TW logging, the difference spectrum analysis (DSM) and time domain analysis (TDA) methods are used. This method has a better identification effect when the difference between the longitudinal relaxation time of oil and water is large.

[0005] (2) Based on dual TE logging, the spectral shift analysis (SSM) and quantitative diffusion analysis (DIFAN) methods are used. This method has a better identification effect when the difference between oil and water diffusion relaxation is large.

[0006] However, existing nuclear magnetic resonance logging interpretations have the following main drawbacks:

[0007] (1) When nuclear magnetic resonance multi-TW and multi-TE logging modes are measured simultaneously, the data are analyzed independently for dual-TW and dual-TE modes. However, the two observation data and interpretation results are often inconsistent, leading to confusion in the qualitative identification and quantitative evaluation of reservoir fluids.

[0008] (2) When the fluid viscosity and oil-water relaxation parameters are uncertain and the fluid contribution to the nuclear magnetic resonance logging response is small, the T2 distribution difference of the two TWs is not obvious, and the quantitative analysis method - time domain analysis (TDA) is not accurate enough.

[0009] (3) When the difference in fluid diffusion relaxation is small, the difference in T2 distribution of the two TEs is not obvious, and the quantitative analysis method - diffusion analysis is not accurate enough.

[0010] Therefore, it is imperative to invent a joint inversion method based on logging data from different modes of nuclear magnetic resonance logging. Summary of the Invention

[0011] This invention provides a method and apparatus for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging, which solves the problem of inconsistencies and multiple solutions that often occur when different logging data are analyzed separately in the prior art.

[0012] A method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging, the method comprising:

[0013] Multiple sets of first echo train data were generated based on nuclear magnetic resonance multi-TW multi-TE logging data;

[0014] A joint equation is constructed based on multiple sets of the first echo train data, and then recombined into the second echo train data;

[0015] A signal matrix is ​​constructed based on the second echo train data to perform joint inversion on the second echo train data.

[0016] Optionally, the method further includes:

[0017] The difference spectrum and the oil saturation of the flushing zone were obtained based on the joint inversion results.

[0018] Optionally, multiple sets of first echo train data can be generated based on nuclear magnetic resonance logging data, including:

[0019] Relaxation mechanism analysis of nuclear magnetic resonance logging data;

[0020] Multiple sets of first echo train data were generated based on the relaxation mechanism analysis results.

[0021] Optionally, the method further includes: generating different sets of first echo train data based on multi-TW multi-TE observation modes.

[0022] Optionally, based on the multi-TW multi-TE observation mode, the first echo train data includes: a long waiting time short echo interval group, a short waiting time short echo interval group, a long waiting time long echo interval group, and a short waiting time long echo interval group.

[0023] Optionally, based on the dual-TW dual-TE observation mode, the first echo train data includes: a long waiting time short echo interval group, a short waiting time short echo interval group, and a long waiting time long echo interval group.

[0024] Optionally, the T2 distribution of the difference spectrum can be obtained by the least squares method.

[0025] Optionally, the method further includes: identifying fluids in the study area based on the difference spectrum and oil saturation of the flushing zone obtained after joint inversion.

[0026] This invention discloses a joint inversion device for multi-TW and multi-TE observation data from nuclear magnetic resonance logging, the device comprising:

[0027] The first generation unit is used to generate multiple sets of first echo train data based on nuclear magnetic resonance multi-TW multi-TE logging data;

[0028] The second generation unit is used to construct joint equations based on multiple sets of the first echo train data and recombine them into the second echo train data.

[0029] The joint inversion unit is used to construct a signal matrix based on the second echo train data to perform joint inversion on the second echo train data.

[0030] The present invention provides a computer-readable storage medium storing one or more programs, which can be executed by one or more processors to implement the steps of the joint inversion method for multi-TW and multi-TE observation mode data of nuclear magnetic resonance logging as described in any of the above claims.

[0031] The present invention relates to a method and apparatus for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging. First, different echo train data obtained from the multi-TW and multi-TE modes are recombined. Then, a new method for joint inversion is used to obtain joint inversion results of multiple echo data, thereby eliminating multiple solutions. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating the joint inversion method based on nuclear magnetic resonance logging data in an embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram of the joint inversion device based on nuclear magnetic resonance logging data in an embodiment of the present invention;

[0034] Figures 3A-3D This is a schematic diagram comparing the oil saturation results calculated based on the embodiments of the present invention with the model saturation;

[0035] Figure 4 This is the result of nuclear magnetic resonance logging interpretation of well D5 based on the method of the embodiments of the present invention. Detailed Implementation

[0036] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0037] It should be understood that in the various embodiments of this document, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0038] In current oil and gas exploration, simultaneous measurements using multiple TW and multiple TE modes in nuclear magnetic resonance (NMR) logging are becoming increasingly common. However, analyzing the dual-TW and dual-TE data separately often yields inconsistencies. Therefore, there is an urgent need to develop joint inversion techniques for multiple echo train data sets under NMR logging multiple-TW and multiple-TE modes.

[0039] Specifically, embodiments of the present invention provide a joint inversion method based on nuclear magnetic resonance logging data, such as... Figure 1 As shown, the method includes:

[0040] Step 100: Generate multiple sets of first echo train data based on NMR multi-TW multi-TE logging data. Specifically, generate multiple sets of first echo train data based on NMR logging data from the DTWE2 mode of the MRIL-P instrument or the D9TWE3 mode of the MRT instrument. The DTWE2 mode of the MRIL-P instrument and the D9TWE3 mode of the MRT instrument observe large amounts of data with rich information, reflecting fluid properties, and are therefore suitable for obtaining multiple sets of first echo train data in this embodiment of the invention. Similarly, this specific embodiment of the invention is also applicable to data obtained by other similar instruments. In this specific embodiment, the physical basis of NMR logging technology is the magnetism of atomic nuclei and their interaction with an applied magnetic field. The measured first echo train data consists of relaxation decay curves composed of hundreds of spin echoes. The spin echo train obtained by measuring the transverse relaxation time in the CPMG sequence of low-field NMR does not exhibit single exponential decay, but rather the sum of multiple exponential decays.

[0041] Step 200: Construct joint equations based on multiple sets of the first echo string data and recombine them into second echo string data. Specifically, construct joint equations to recombine the generated multiple sets of first echo string data into new second echo string data. Specifically, regarding step 100 above, if a multi-TW multi-TE observation mode is used, the obtained first echo string data can be divided into two categories: one category is echo strings with short echo intervals and long / short waiting times; the other category is echo strings with long echo intervals and long / short waiting times. Based on the short echo interval, construct joint equations for the echo string formulas corresponding to the first echo string data with long and short waiting times to obtain the first magnetization vector difference at different waiting times within the short echo interval; based on the long echo interval, construct joint equations for the echo string formulas corresponding to the first echo string data with long and short waiting times to obtain the second magnetization vector difference at different waiting times within the long echo interval. Recombine the first and second magnetization vector differences into new second echo string data.

[0042] Step 300: Construct a signal matrix based on the second echo train data to perform joint inversion on the second echo train data. Specifically, the most important problem in nuclear magnetic resonance logging data processing is obtaining the T2 distribution from the echo train inversion. Therefore, after generating the second echo train data in step 200, the signal matrix is ​​reconstructed according to the set observation parameters, and the second echo train data is jointly inverted. The key to multi-exponential fitting of nuclear magnetic resonance logging data is how to solve for the T2 relaxation time of various types of pores and the proportion of pores from the equations. In a preferred embodiment, the LSOR algorithm is used for calculation. The LSOR algorithm is a method proposed by Paige and Sanders that uses the Lanczos iterative method to solve least squares problems. This method has low computational cost and can easily simplify calculations by utilizing the sparsity of the matrix, thus making it suitable for solving large sparse problems.

[0043] The method described in the above embodiments of the present invention first reassembles different echo train data obtained from multiple TW and multiple TE modes, and obtains joint inversion results of multiple echo data through a new joint inversion method, thereby eliminating multiple solutions.

[0044] The joint inversion method based on nuclear magnetic resonance logging data described in the specific embodiments of the present invention preferably further includes: obtaining the difference spectrum and the oil saturation of the flushed zone based on the joint inversion results.

[0045] The joint inversion method based on nuclear magnetic resonance logging data described in the specific embodiments of the present invention preferably includes generating multiple sets of first echo train data based on the nuclear magnetic resonance logging data, including:

[0046] Relaxation mechanism analysis was performed on nuclear magnetic resonance logging data. In a specific embodiment, under short echo interval (TES), long and short waiting times (T...) were used respectively. WL T WS Measurements were performed, yielding two sets of first echo train data. According to the nuclear magnetic resonance relaxation mechanism, since the T1 of oil in the reservoir is much longer than that of water in the rock, the effect of diffusion relaxation is relatively small under TES conditions and can be ignored. Therefore, the measured echoes include surface relaxation and volume relaxation. Under a long echo interval (TEL), measurements were performed using both long and short waiting times, yielding two sets of first echo train data. According to the nuclear magnetic resonance relaxation mechanism, under TEL conditions, i.e., when TE is 3.6 ms, the effect of diffusion relaxation is significant and cannot be ignored. Therefore, the measured echoes include surface relaxation, volume relaxation, and diffusion relaxation.

[0047] Multiple sets of first echo train data were generated based on the relaxation mechanism analysis results.

[0048] In a specific embodiment, under short echo interval (TES), the reservoir is usually hydrophilic, and water is in contact with the surface of rock particles. The formulas for the echo train amplitude measured by long and short waiting times are as follows (1) and (2):

[0049]

[0050]

[0051] In the formula, M L,TES(t) M represents the magnetization at time t under short echo interval and long waiting time, in A / m; S,TES(t) M represents the magnetization at time t under short echo interval and short waiting time, in A / m; 0wj Let M be the magnetization intensity of the j-th component water, in A / m; oil T represents the magnetization of the oil, in A / m. 2,wj Let T be the transverse relaxation time of water for the j-th component, in milliseconds; 2,oil T represents the transverse relaxation time of the oil, in milliseconds (ms). 1,oil T represents the longitudinal relaxation time of the oil, in milliseconds (ms). 1,water Let be the longitudinal relaxation time of water, in milliseconds (ms).

[0052] Subtracting the magnetization intensities of the long and short waiting times yields the vector difference as shown in formula (3). This vector difference can reflect the fluid type.

[0053]

[0054] Under long echo interval (TEL), measurements were performed using two different polarization times, long and short, respectively, yielding two sets of first echo train data. According to the nuclear magnetic resonance relaxation mechanism, under TEL conditions, i.e., when TE is 3.6 ms, the effect of diffusion relaxation is significant and cannot be ignored. Therefore, the measured echoes include surface relaxation, bulk relaxation, and diffusion relaxation. The echo train amplitudes measured with long and short waiting times are respectively calculated using formulas (4) and (5):

[0055]

[0056]

[0057] In the formula, M L,TEL (t) represents the magnetization at time t under a long echo interval and long waiting time, in A / m; M S,TEL (t) represents the magnetization at time t under a long echo interval and short waiting time, in A / m; T 2DW Let be the water diffusion coefficient, in cm. 2 / s;T 2DO Let be the diffusion coefficient of the oil, in cm. 2 / s.

[0058] Subtracting the magnetization intensities for the long and short waiting times yields the vector difference as shown in formula (6):

[0059]

[0060] Let Δα oil and Δα water These are the polarization functions of oil and water, respectively. Then formulas (3) and (6) become formulas (7) and (8) respectively:

[0061]

[0062]

[0063] The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging, as described in a specific embodiment of the present invention, preferably further includes: generating different sets of first echo train data based on different observation modes. Specifically, the number of sets of first echo train data generated by the multi-TW and multi-TE observation mode or the dual-TW and dual-TE observation mode is different. The multi-TW and multi-TE observation mode can generate 4 sets of first echo train data, while the dual-TW and dual-TE observation mode can generate 3 sets of first echo train data. However, the number of sets of first echo train data does not affect the subsequent joint inversion.

[0064] The joint inversion method based on nuclear magnetic resonance logging data described in the specific embodiments of the present invention is preferably based on a multi-TW multi-TE observation mode, wherein the first echo train data includes: long waiting time short echo interval group (TWL, TES), short waiting time short echo interval group (TWS, TES), long waiting time long echo interval group (TWL, TEL), and short waiting time long echo interval group (TWS, TEL).

[0065] The joint inversion method for multi-TW and multi-TE observation data from nuclear magnetic resonance logging described in a specific embodiment of the present invention is preferably based on a dual-TW and dual-TE observation mode. The first echo train data includes: a long waiting time short echo interval group (TWL, TES), a short waiting time short echo interval group (TWS, TES), and a long waiting time long echo interval group (TWL, TEL).

[0066] In a specific embodiment of the present invention, the joint inversion of the second echo train data includes:

[0067] Assuming that there are m and m' observed echoes under short and long echo intervals, respectively, and n relaxation components, the echo vectors under different modes can be expressed as formulas (9) and (11):

[0068]

[0069] t(i)=i×T ES ,i=1,2,3,...,m.

[0070] Represented in matrix form as formula (10):

[0071] ΔM1=A1p+ε1 (10)

[0072]

[0073] t(i′)=i′×T EL ,i′=1,2,3,...,m′.

[0074] Represented in matrix form as formula (12):

[0075] ΔM2=A2p+ε2 (12)

[0076] in, Combine echo signals from different modes for processing:

[0077]

[0078] In the formula, ΔM1 represents the echo signal difference between the short echo interval and the long and short polarization time observation modes; ΔM2 represents the echo signal difference between the long echo interval and the long and short polarization time observation modes; pj Let T be the component porosity of the j-th type of pore; 2j Tj is the T2 relaxation time for the j-th type of pore; t is time; Tj ES For short echo intervals; T EL ε is the long echo interval; ε, ε1, ε2 are random noise; A1 is an m×n matrix; A2 is an m′×n matrix; m and m′ are the number of echoes; △M1=[△M1(1),△M1(2),…,△M1(m)] T ;△M2=[△M2(1),△M2(2),…,△M2(m′)] T p = [p1, p2, ..., p n ] T .

[0079] In the specific embodiments of the present invention, the joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging preferably obtains the T2 distribution of the difference spectrum through the least squares method.

[0080] The joint inversion method based on nuclear magnetic resonance logging data described in the specific embodiments of the present invention preferably further includes: fluid identification of the study area based on the difference spectrum and oil saturation of the flushed zone obtained after joint inversion.

[0081] A joint inversion device for multi-TW and multi-TE observation mode data from nuclear magnetic resonance logging, as described in a specific embodiment of the present invention, is as follows: Figure 2 As shown, the device includes:

[0082] The first generation unit 201 is used to generate multiple sets of first echo train data based on nuclear magnetic resonance logging data;

[0083] The second generation unit 202 is used to construct joint equations based on multiple sets of the first echo train data and recombine them into the second echo train data.

[0084] The joint inversion unit 203 is used to construct a signal matrix based on the second echo train data to perform joint inversion on the second echo train data.

[0085] To verify the accuracy of the above methods, rock models with a porosity of 13% and different oil-water contents (oil saturation) were set up, with oil saturation of 0%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, and 90%. Based on rock physics experiments, the transverse and longitudinal relaxation times of oil were set to 500 ms and 4000 ms, respectively, and the longitudinal relaxation time of water was set to 500 ms. The hydrogen index of both oil and water was 1.0. The long waiting time was set to 12.98 s, the short waiting time to 1.0 s, the short echo interval to 0.9 ms, and the long echo interval to 3.6 ms. Using the preset model, given parameters, and oil-water contents, and based on the nuclear magnetic resonance logging response mechanism, multi-TW and multi-TE observation echo trains were generated using equations (1), (2), (4), and (5). The T2 distribution of the difference spectrum was finally obtained using the least squares method.

[0086] Synthesized echoes and their inverted difference spectra for rock models with different saturation levels were analyzed. Under short echo interval observation mode, the amplitude of the difference spectrum T2 distribution increases with increasing oil saturation. Under long echo interval mode, the difference spectrum was obtained by inverting echo trains with different waiting times. Due to diffusion relaxation, the T2 distribution shifts towards decreasing T2 with increasing oil saturation, and the amplitude increases. Joint inversion of the two echo train differences showed that the amplitude of the difference spectrum T2 distribution increases with increasing oil saturation, indicating that this difference spectrum can be used for fluid identification.

[0087] Meanwhile, the oil-bearing volume under different models was obtained by inversion, and the corresponding oil saturation was calculated and compared with the model saturation. Figure 3A 3B, 3C, and 3D show a comparison between the calculated oil saturation and the model saturation. After joint inversion, the calculated oil saturation is consistent with the model. Figure 3A Comprehensive comparison, for example Figure 3B The oil saturation calculated after joint inversion showed better results. In contrast, the oil saturation calculated by the TDA method and diffusion analysis method differed significantly from the model saturation, such as... Figure 3C , Figure 3D Table 1 shows the comparison results of the absolute errors of different saturation models. The average absolute error of the oil saturation calculated after joint inversion is 3.9%. The average absolute errors of the oil saturation calculated by the TDA method and the diffusion analysis method DIFAN are 6.58% and 6.74%, respectively. In comparison, the method described in this embodiment of the invention has higher accuracy in calculating oil volume and oil saturation, which can meet the needs of production.

[0088] Table 1 Comparison of absolute errors of models with different saturation levels

[0089]

[0090] This invention also provides a computer-readable storage medium storing one or more programs that can be executed by one or more processors to implement the steps of the joint inversion method based on nuclear magnetic resonance logging data as described in any of the above specific embodiments.

[0091] The present invention was applied in a specific embodiment in the Yanchang Formation of a basin in central and western China. An application example is given below.

[0092] The complex oil-water layer in the western Yanchang Formation of a basin in central and western China is an important target layer for oil and gas. The formation water salinity varies widely, resulting in weak resistivity indicative of oil-bearing potential and low fluid discrimination accuracy in resistivity logging. Therefore, nuclear magnetic resonance logging (NMR) was performed using an MRT instrument in the D9TWE3 observation mode. Five sets of echo data were obtained. Sets A and B were acquired using a short echo interval (TES) (0.9 ms) long-short polarization time mode, with an echo number (NE) of 500, a long wait time (TWL) of 12988 ms, and a short wait time (TWS) of 1000 ms. Sets D and E were acquired using a long echo interval (TEL) (3.6 ms) long-short polarization time mode, with an echo number (NE) of 125, a long wait time (TWL) of 12998 ms, and a short wait time (TWS) of 1000 ms. Set C contained observation parameters for the clay formation, with an echo number (NE) of 10 and a wait time (TW) of 20 ms.

[0093] Using the above method, echo train data was split and combined, jointly inverted, and saturation was calculated. Figure 4 The following is an interpretation of the nuclear magnetic resonance logging results for well D5 using a specific embodiment of this invention. Track 3 shows resistivity; track 4 shows the T2 distribution under long waiting time and short echo interval; track 5 shows the T2 distribution under short waiting time and short echo interval; track 6 shows the T2 distribution under long waiting time and long echo interval; track 7 shows the T2 distribution under short waiting time and long echo interval; track 8 shows the difference spectrum after joint inversion; track 9 shows the calculated oil saturation of the flushed zone after joint inversion; track 10 shows fluid volume analysis; and track 11 shows formation analysis. At 2890-2900m, the resistivity is approximately 8 Ω·m, interpreted as a water layer. However, the difference spectrum signal from the joint inversion is significant, and the calculated oil saturation of the flushed zone is high, approximately 40%, interpreted as an oil layer. The oil test results show a daily oil production of 58.31t and a daily water production of 0, indicating an oil layer and verifying the accuracy of fluid identification.

[0094] Furthermore, the above method was used to process and interpret data from 11 layers in 9 other wells in the study area, and the interpretation results showed a 90% agreement rate with the oil testing results. Therefore, the method of this embodiment is accurate, feasible, and highly effective.

[0095] This invention is applicable not only to data processing and interpretation of oil-water layers in oil reservoirs, but also to data processing and interpretation of gas-water layers in gas reservoirs.

[0096] The method described in this invention, specifically in the multi-TW and multi-TE observation modes of nuclear magnetic resonance (NMR) logging, derives formulas for macroscopic magnetization vectors, fluid volume, and relaxation parameters. The NMR relaxation mechanism conforms to the characteristics of the observation modes, realizing a novel joint inversion method for NMR logging data from multiple TW and multiple TE modes. It calculates the oil-bearing volume and oil saturation of the reservoir flushed zone. Based on the difference spectrum obtained after joint inversion and the oil saturation of the reservoir flushed zone, fluid type identification can be performed. The joint inversion theory and novel fluid identification approach proposed in this invention are applicable not only to multi-TW and multi-TE strategy modes but also to dual-TW and dual-TE measurement mode data, without affecting subsequent data processing and results.

[0097] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0098] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0099] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0100] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0101] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging, characterized in that, The method includes: Multiple sets of first echo train data were generated based on nuclear magnetic resonance multi-TW multi-TE logging data; A joint equation is constructed based on multiple sets of the first echo train data, and then recombined into the second echo train data; A signal matrix is ​​constructed based on the second echo train data to perform joint inversion on the second echo train data.

2. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 1, characterized in that, The method further includes: The difference spectrum and the oil saturation of the flushing zone were obtained based on the joint inversion results.

3. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 1, characterized in that, Multiple sets of first echo train data were generated based on nuclear magnetic resonance multi-TW multi-TE logging data, including: Relaxation mechanism analysis of nuclear magnetic resonance logging data; Multiple sets of first echo train data were generated based on the relaxation mechanism analysis results.

4. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 3, characterized in that, The method also includes generating different sets of first echo train data based on multi-TW multi-TE observation modes.

5. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 4, characterized in that, Based on the multi-TW multi-TE observation mode, the first echo train data includes: long waiting time short echo interval group, short waiting time short echo interval group, long waiting time long echo interval group, and short waiting time long echo interval group.

6. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 4, characterized in that, Based on the dual-TW dual-TE observation mode, the first echo train data includes: a long waiting time short echo interval group, a short waiting time short echo interval group, and a long waiting time long echo interval group.

7. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 2, characterized in that, The T2 distribution of the difference spectrum was obtained using the least squares method.

8. The method for joint inversion of multi-TW and multi-TE observation data from nuclear magnetic resonance logging according to claim 2, characterized in that, The method further includes: identifying fluids in the study area based on the difference spectrum and oil saturation of the flushing zone obtained after joint inversion.

9. A joint inversion device for multi-TW and multi-TE observation data from nuclear magnetic resonance logging, characterized in that, The device includes: The first generation unit is used to generate multiple sets of first echo train data based on nuclear magnetic resonance multi-TW multi-TE logging data; The second generation unit is used to construct joint equations based on multiple sets of the first echo train data and recombine them into the second echo train data. The joint inversion unit is used to construct a signal matrix based on the second echo train data to perform joint inversion on the second echo train data.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores one or more programs, which can be executed by one or more processors to implement the steps of the joint inversion method for multi-TW and multi-TE observation data of nuclear magnetic resonance logging as described in any one of claims 1 to 8.

Images