An oil-water well fluid production profile identification, inversion method, system and program product

Abstract

The application discloses an oil-water well liquid production profile identification and inversion method, system and program product, relates to the oil and gas field development technical field, and solves the problems that in the existing liquid production profile monitoring method based on DAS and DTS technologies, DAS signals are easily affected by fluid flow interference, signal quality is poor, liquid point information cannot be accurately reflected, DTS signal inversion liquid production profile information lacks sufficient spatial constraint conditions, and inversion results are large in deviation; the application extracts local liquid production characteristics in DAS signals through blind source separation, effectively identifies liquid point positions in combination with a statistical screening method, introduces the identified liquid point positions as constraint conditions into a DTS inversion process, and effectively improves the precision and stability of the inversion results; the application not only overcomes the limitations of a traditional F-K spectrum analysis method in an oil-water well environment, but also can introduce effective spatial constraints in DTS inversion, significantly reduces inversion errors, and improves the interpretation precision of the liquid production profile.

Description

Technical Field

[0001] This invention relates to the field of oil and gas field development technology, and more specifically, to a method, system, and program product for identifying and retrieving production profiles of oil and water wells. Background Technology

[0002] In oil and gas field development, accurately identifying the production profile along the depth direction of the wellbore is of great significance for optimizing injection and production strategies, improving reservoir recovery, and extending the production life of the oilfield. Production profile monitoring helps engineers to grasp the fluid distribution within the wellbore in real time, thereby effectively adjusting production strategies, especially in the management of multi-well and multi-layer reservoirs, providing precise data support.

[0003] Currently, with the development of distributed fiber optic sensing technology, the combined application of distributed acoustic sensing (DAS) and distributed temperature sensing (DTS) has become an important trend in production profile monitoring. DAS technology acquires real-time acoustic signals along the depth direction of the wellbore by monitoring the propagation of sound waves within the wellbore; while DTS technology reflects fluid movement and heat conduction through temperature changes. The combination of these two technologies provides strong data support for monitoring production profiles. However, in practical applications, the integration of DAS and DTS still faces a series of technical challenges.

[0004] First, the application of DAS signals in oil and water wells presents significant challenges. Oil and water wells typically operate at low flow rates, and the fluid within the well is a complex two-phase mixture of oil and water. In this environment, DAS signals are often affected by complex acoustic interference caused by fluid disturbances, resulting in poor signal quality. Especially when oil and water are mixed, the properties and fluctuations of the fluid can cause strong interference, affecting the propagation characteristics of acoustic waves and making it impossible for DAS signals to accurately reflect liquid point information. Furthermore, due to the low flow rate and complex acoustic wave propagation characteristics, the applicability of traditional frequency-wavenumber (FK) analysis methods in oil and water well environments is limited. This method is typically used for frequency analysis of acoustic waves in high-speed gas wells, but it cannot effectively extract clear acoustic velocity characteristics in oil and water wells, leading to an inability to accurately identify production areas.

[0005] Secondly, DTS signals present problems in the inversion process of product profiles. The DTS inversion process converts temperature data measured by fiber optic sensors into product profile information. However, the inversion problem itself is severely ill-posed, lacking sufficient spatial constraints, which can easily lead to significant deviations in the inversion results. Specifically, without sufficient spatial constraints, the inversion results may have multiple solutions or be unstable, making it difficult to provide an accurate and reliable interpretation of the product profile. Therefore, the lack of explicit identification of the actual effluent point location during DTS inversion can lead to inconsistent or even erroneous inversion results due to misjudgment or lack of constraints.

[0006] Therefore, this application proposes a method, system, and program product for identifying and retrieving production profiles in oil and water wells, thereby solving the problems mentioned in the background art. Summary of the Invention

[0007] The purpose of this application is to provide a method, system, and program product for identifying and inverting production profiles in oil and water wells. This addresses the problems in existing production profile monitoring methods based on DAS and DTS technologies. DAS signals are susceptible to fluid flow interference, resulting in poor signal quality and inaccurate reflection of fluid point information. Furthermore, DTS signal-derived production profile information lacks sufficient spatial constraints, leading to large deviations in the inversion results. This application extracts local production features from DAS signals through blind source separation and effectively identifies fluid point locations using statistical screening methods. The identified fluid point locations are then introduced as constraints into the DTS inversion process, effectively improving the accuracy and stability of the inversion results. This application not only overcomes the limitations of traditional FK spectrum analysis methods in oil and water well environments but also introduces effective spatial constraints into DTS inversion, significantly reducing inversion errors and improving the interpretation accuracy of production profiles.

[0008] This paper first provides a method for identifying and inverting the production profile of oil and water wells, including: Step S1, acquiring DAS and DTS signals based on a distributed optical fiber system arranged in the wellbore, and preprocessing and extracting low frequencies from the DAS signals; Step S2, performing blind source separation on the extracted DAS signals to extract each independent source signal; Step S3, identifying potential production point locations by combining statistical screening based on each independent source signal; Step S4, using the potential production point locations as constraints to perform DTS inversion, and obtaining the final interpretation result according to the forward modeling solution strategy.

[0009] In one possible implementation, step S1 includes: discretely sampling the acoustic wave signal at spatial intervals and time intervals to obtain a DAS signal; decomposing the DAS signal into signals of different frequencies based on a two-dimensional spatiotemporal Fourier transform; and filtering the signals of different frequencies to extract the low-frequency band signal.

[0010] In one possible implementation, step S2 includes: performing mean-removal processing on the extracted DAS signal, and whitening processing on the mean-removed signal; determining the number of independent sources based on the whitened signal, which is taken as the number of mutually independent effluent points; and inputting the number of independent sources into an independent component analysis algorithm to perform blind source separation on the extracted signal to obtain each independent source signal.

[0011] In one possible implementation, step S3 includes: calculating the low-frequency energy index or amplitude index of each independent source signal, distributing the low-frequency energy index or amplitude index along the depth direction to obtain the sound wave intensity distribution along the depth; and selecting spatial locations where the sound wave intensity exceeds the sound wave intensity threshold as potential liquid outlet locations based on the sound wave intensity distribution along the depth and a set statistical threshold or hypothesis testing method.

[0012] In one possible implementation, step S4 includes: Step S41, constructing a mathematical model of wellbore-reservoir fluid and heat conduction coupling to describe the temperature and pressure changes of the fluid in the wellbore, as well as the flow and heat conduction process of multiphase fluid in the reservoir; Step S42, constructing and initializing a single-well spatial grid model, setting the grid well index of the potential fluid production point identified by blind source separation to 1, and setting the rest to 0, as the boundary conditions for subsequent simulation; Step S43, simulating the wellbore temperature and oil and water production profile at each time point based on the forward modeling solution strategy, extracting the actual observed wellbore temperature and oil and water production profile based on the DTS signal, and optimizing the model parameters by comparing the simulation results with the actual observation results; Step S44, calculating the final horizontal well oil and water production profile based on the model with optimized parameters.

[0013] In one possible implementation, step S41 includes: constructing a wellbore fluid thermodynamic behavior model to describe the changes in fluid temperature and pressure within the wellbore; and constructing a reservoir multiphase flow model and a reservoir heat conduction process model.

[0014] In one possible implementation, step S42 includes: collecting various data related to the reservoir and well; performing spatial coordinate transformation on the structural surface and well trajectory data, so that the transformed well trajectories are located in the same... In the plane; a regular hexahedral mesh is generated based on the structural surface after spatial coordinate transformation to simulate the three-dimensional space of the reservoir and wellbore; initial attribute values ​​are assigned to each mesh cell; the mesh well index of the potential fluid production point location identified by blind source separation is set to 1, and the rest are set to 0, as the boundary conditions for subsequent simulations.

[0015] In one possible implementation, step S43 includes: performing forward modeling based on a mathematical model of wellbore-reservoir fluid and heat conduction coupling and a single-well spatial grid model to simulate wellbore temperature and oil and water production profiles at various times; extracting wellbore temperature and oil and water production profiles at corresponding times based on DTS signals and DAS signal segments as actual observation results; and comparing the results obtained from forward modeling with the results obtained from actual observations based on an integrated progressive data assimilation method to optimize model parameters.

[0016] This application also provides a water well production profile identification and inversion system to implement the oil and water well production profile identification and inversion method described above. The system includes: a signal acquisition module for acquiring DAS and DTS signals based on a distributed optical fiber system deployed in the wellbore, and preprocessing and extracting low-frequency signals from the DAS signals; a signal separation module for performing blind source separation on the extracted DAS signals and extracting signals from each independent source; a production point identification module for identifying potential production point locations based on statistical screening of each independent source signal; and a DTS inversion module for performing DTS inversion with the potential production point locations as constraints, and obtaining the final interpretation result based on a forward modeling solution strategy.

[0017] This application also provides a computer program product that, when run on an electronic device, causes the electronic device to execute the oil and water well production profile identification and inversion method as described above.

[0018] Compared with the prior art, this application has the following beneficial effects: The oil and water well production profile identification, inversion method, system and program products provided in this application extract local (low frequency, ultra-low frequency) production signals through blind source separation of DAS signals, and identify the liquid point location by statistical screening, which is then used as a constraint condition to guide the inversion of distributed temperature sensing DTS, thereby realizing the joint and accurate interpretation of oil and water well production profiles; This invention is not only applicable to the accurate identification and interpretation of production profiles of various oil and water wells, but can also be applied to oil and water wells with different flow levels and different fluid types, and has great technical promotion value and market potential; It is especially suitable for oil-water two-phase wells with low flow rates and oil-water-gas composite wells, and has unique advantages in accurate monitoring of such wells. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0020] Figure 1 A flowchart of the oil and water well production profile identification and inversion method provided in the embodiments of this application;

[0021] Figure 2 The DAS response waterfall diagram of the entire water injection process provided in the embodiments of this application (intensity is taken as the logarithm with base 2);

[0022] Figure 3 The DAS spatiotemporal Fourier transform and frequency extraction results provided in the embodiments of this application;

[0023] Figure 4 The spatiotemporal distribution diagram of 100Hz after DAS decomposition provided in the embodiments of this application;

[0024] Figure 5This is a 10Hz spatiotemporal distribution diagram after DAS decomposition provided in an embodiment of this application;

[0025] Figure 6 This is a 3Hz spatiotemporal distribution diagram after DAS decomposition provided in an embodiment of this application;

[0026] Figure 7 This is a 1Hz spatiotemporal distribution diagram after DAS decomposition provided in an embodiment of this application;

[0027] Figure 8 Spatiotemporal distribution map of 0.5Hz after DAS decomposition provided in the embodiments of this application;

[0028] Figure 9 A schematic diagram of the local acoustic signal obtained after separation of a certain depth blind source, as provided in an embodiment of this application;

[0029] Figure 10 This is a local acoustic wave spatiotemporal distribution map obtained after DAS blind source separation provided in an embodiment of this application;

[0030] Figure 11 A schematic diagram of the three-dimensional mesh of the single-well coupled numerical model provided in the embodiments of this application;

[0031] Figure 12 A schematic diagram of the wellbore temperature profile distribution at a certain moment for each model after data assimilation provided in the embodiments of this application;

[0032] Figure 13 A schematic diagram of the horizontal section water absorption profile obtained for the purpose of explaining the embodiments of this application;

[0033] Figure 14 This is a structural diagram of the oil and water well production profile identification and inversion system provided in the embodiments of this application. Detailed Implementation

[0034] In the following, the terms “comprising” or “may include” as used in the various embodiments of this application indicate the presence of the claimed function, operation, or element, and do not limit the addition of one or more functions, operations, or elements. Furthermore, as used in the various embodiments of this application, the terms “comprising,” “having,” and their cognates are intended only to indicate a specific feature, number, step, operation, element, component, or combination of the foregoing, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations of the foregoing, or the possibility of adding one or more combinations of the foregoing.

[0035] The terminology used in the various embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the various embodiments of this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in a generally used dictionary) are to be interpreted as having the same meaning as in the context of the relevant technical field and are not to be interpreted as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.

[0037] Please see Figure 1 As shown, Figure 1 This is a flowchart of a method for identifying and inverting production profiles in oil and water wells. The method includes: Step S1, acquiring DAS and DTS signals based on a distributed optical fiber system deployed in the wellbore, and preprocessing and extracting low frequencies from the DAS signals; Step S2, performing blind source separation on the extracted DAS signals to extract individual source signals; Step S3, identifying potential production point locations by combining statistical screening based on the individual source signals; Step S4, using the potential production point locations as constraints for DTS inversion, and obtaining the final interpretation result based on a forward modeling solution strategy.

[0038] Specifically, this application identifies the location of the production point of oil and water wells based on blind source separation and statistical screening of DAS signals, and interprets the production profile information of oil and water wells by introducing production point location constraints based on DTS signals.

[0039] The improvement of this application lies in extracting local production signals through blind source separation (BSS) technology of DAS signals and identifying the location of liquid points using statistical screening methods. These locations are then used as constraints to guide distributed temperature sensing (DTS) inversion, thereby achieving a joint and accurate interpretation of production profiles in oil and water wells. This application not only overcomes the limitations of traditional FK spectral analysis methods in oil and water well environments but also introduces effective spatial constraints into DTS inversion, significantly reducing inversion errors and improving the interpretation accuracy of production profiles. It can accurately extract DAS signal features related to production activity in oil and water wells, solving the problem that traditional methods cannot adapt to the complex flow environment of oil and water wells, while improving the accuracy of DTS inversion and ensuring high precision and stability of production profile interpretation.

[0040] In one possible implementation, step S1 includes: discretely sampling the acoustic wave signal at spatial intervals and time intervals to obtain a DAS signal; decomposing the DAS signal into signals of different frequencies based on a two-dimensional spatiotemporal Fourier transform; and filtering the signals of different frequencies to extract the low-frequency band signal.

[0041] Specifically, firstly, acoustic signals are collected in a discrete manner in space and time using a distributed optical fiber system. The collected signals are represented as follows:

[0042] ;

[0043] in: It is the spatial sampling point number (total number) (spatial sampling points); It is the time sampling point number (total) (number of time sampling points); the spatial interval of sampling is... The sampling time interval is ; Let m be the position coordinates of the m-th spatial sampling point along the fiber optic axis; This represents the time corresponding to the nth time sampling point.

[0044] Then, a two-dimensional Fourier transform is performed on the acquired discrete signal to convert the signal from the space-time domain to the space-frequency domain. The formula for the two-dimensional Fourier transform is:

[0045] ;

[0046] in, It is a spatial frequency index (corresponding to) ), It is a time frequency index (corresponding to) Through the aforementioned two-dimensional spatiotemporal Fourier transform, the original spatiotemporal signal is decomposed into multiple spatiotemporal fields with specific frequency characteristics. These frequency fields may include, but are not limited to, 0.01Hz, 0.02Hz, 1Hz, 100Hz, etc., depending on the time-frequency index. .

[0047] Finally, the separated frequency band signals are processed to remove isolated spike noise, instrument drift, and other interference, improving signal quality and reliability, thus making subsequent analysis and applications more accurate. For distributed fiber optic systems, the main focus is on extracting low-frequency and even ultra-low-frequency signals to enhance local acoustic signals related to fluid production activities and improve the signal-to-noise ratio.

[0048] In one possible implementation, step S2 includes: performing mean-removal processing on the extracted DAS signal, and whitening processing on the mean-removed signal; determining the number of independent sources based on the whitened signal, which is taken as the number of mutually independent effluent points; and inputting the number of independent sources into an independent component analysis algorithm to perform blind source separation on the extracted signal to obtain each independent source signal.

[0049] Specifically, blind source separation (BSS) refers to the process of recovering the original source signal as much as possible from the observed mixed signal when the details of the mixing process and the characteristics of the source signal are unknown. The DAS signal in this application is applicable to this process, due to the unknown... The acoustic waves generated at each fluid outlet propagate along the wellbore, and the acquired DAS signal is a mixture of observation signals from each fluid outlet location. Therefore, blind source separation is required to extract the mixed observation signals and separate independent source signals from them. Each independent source signal generates an acoustic signal at a specific depth underground.

[0050] First, the extracted mixed observation signals are subjected to mean removal processing. This process is repeated for each mixed observation signal. The mean of each dimension is 0; then the mean-reduced signal is whitened. This allows for the mixing of observed signals. To achieve isotropy, the covariance matrix of the signal must become the identity matrix. Principal component analysis (PCA) can be used specifically for this purpose.

[0051] Let the covariance matrix be ;in It is a mixed observation signal. Indicates the expected value. yes transpose;

[0052] Perform eigenvalue decomposition or singular value decomposition (SVD): on the covariance matrix Perform eigenvalue decomposition to obtain ;in It is an eigenvector matrix, and its column vectors are eigenvectors; It is a diagonal matrix, and the elements on the diagonal are... The eigenvalues ​​are arranged in descending order; It is the transpose matrix;

[0053] Then the whitening matrix: ;

[0054] Signals after albinism: ;

[0055] After this processing, the whitening signal The covariance matrix is ​​the identity matrix. ,Right now .

[0056] Secondly, the number of independent sources is determined based on the whitening signal. The whitening signal obtained from PCA processing is decomposed into eigenvalues, and the resulting eigenvalues ​​correspond to the energies of each principal component. The number of sources is typically determined using the eigenvalue energy decrease inflection point. Methods include: inflection point detection, automatic determination using information criteria (such as AIC, BIC), and energy accumulation methods: for example, accumulating to over 95% of the total energy as the number of effective principal components (i.e., the number of sources).

[0057] Finally, blind source separation is performed on the extracted DAS signal based on the Independent Component Analysis (ICA) algorithm. The mathematical model for the solution is established as follows: yes DAS signals from independent outlet location sources yes A mixed observation signal. The mixing process is modeled as a linear instantaneous mixing: ,in: , It is an unknown mixture matrix; It is time. The goal is to estimate a dissociation matrix. , so that: That is, through the demixing matrix. Mixed observation signals Converted into a separated signal ,make As close as possible to the original signal That is, each This corresponds to an independent effluent point location. The unmixing matrix can be solved using Independent Component Analysis (ICA). :

[0058] Independent component analysis (ICA) is a signal processing technique that aims to recover the original independent signals from a mixture of signals. A common optimization criterion is to maximize non-Gaussianity (or minimize mutual information), because according to the central limit theorem, the mixed signal tends to approximate a Gaussian distribution.

[0059] A common loss function is negative entropy, which measures the non-Gaussianity of a signal. Negative entropy is defined as:

[0060] ;

[0061] in: It is entropy, which measures the uncertainty of a signal; They are Gaussian variables with the same covariance; It is a signal processed by the ICA algorithm; The larger, Beyond Gaussian;

[0062] Solving this problem can be done using algorithms such as FastICA, Infomax, and JADE. Here, we take the FastICA algorithm as an example, using an iterative formula based on maximizing non-Gaussianity (e.g., using the logarithmic cosh function or the kurtosis function) (using the logarithmic cosh function as an example):

[0063] ;

[0064] Among these: normalization needs to be performed after each iteration. ). This is the updated unmixing matrix; It is the mixed solution matrix before the update; It is the input signal vector; It is a nonlinear function (such as) ), used to approximate non-Gaussianity; yes The derivative; This represents the expected value, that is, for all input signals. The average.

[0065] Understandably, step S2 utilizes the DAS low-frequency signal for blind source separation, successfully overcoming the poor applicability of the traditional FK spectrum method in oil and water wells. DAS signals in oil and water wells are often affected by acoustic interference, making it difficult for traditional FK analysis methods to effectively extract useful information. Through blind source separation technology, clear local production signals can be extracted from complex interference signals, thus providing reliable input for further analysis.

[0066] In one possible implementation, step S3 includes: calculating the low-frequency energy index or amplitude index of each independent source signal, distributing the low-frequency energy index or amplitude index along the depth direction to obtain the sound wave intensity distribution along the depth; and selecting spatial locations where the sound wave intensity exceeds the sound wave intensity threshold as potential liquid outlet locations based on the sound wave intensity distribution along the depth and a set statistical threshold or hypothesis testing method.

[0067] Specifically, potential effluent points are identified based on the blind source separation results. Independent source signals are extracted from the mixed signal using blind source separation technology. These independent source signals correspond to local signals at different spatial locations and depths, and their low-frequency energy or amplitude indices are calculated to obtain the sound wave intensity distribution along the depth. By setting statistical thresholds (e.g., mean + 3 standard deviations) or using hypothesis testing methods (e.g., significance tests based on p-values), spatial locations with statistically significant abnormal increases in sound wave intensity are screened out and identified as potential effluent point locations.

[0068] Understandably, step S3 uses statistical testing to screen spatial locations with significant changes in sound wave intensity as the outlet locations, thereby effectively suppressing misjudgments caused by noise or accidental disturbances. Statistical methods include setting thresholds (e.g., mean + 3 standard deviations) or using hypothesis testing (e.g., significance test based on p-value) to ensure high accuracy in outlet identification.

[0069] In one possible implementation, step S4 includes: Step S41, constructing a mathematical model of wellbore-reservoir fluid and heat conduction coupling to describe the temperature and pressure changes of the fluid in the wellbore, as well as the flow and heat conduction process of multiphase fluid in the reservoir; Step S42, constructing and initializing a single-well spatial grid model, setting the grid well index of the potential fluid production point identified by blind source separation to 1, and setting the rest to 0, as the boundary conditions for subsequent simulation; Step S43, simulating the wellbore temperature and oil and water production profile at each time point based on the forward modeling solution strategy, extracting the actual observed wellbore temperature and oil and water production profile based on the DTS signal, and optimizing the model parameters by comparing the simulation results with the actual observation results; Step S44, calculating the final horizontal well oil and water production profile based on the model with optimized parameters.

[0070] Specifically, based on the blind source separation results, the location of the effluent point is constrained for DTS inversion. The specific steps are as follows:

[0071] Step S41 involves constructing a mathematical model of the coupling between the wellbore, reservoir fluid, and heat conduction. Further, step S41 includes: constructing a wellbore fluid thermodynamic behavior model to describe the changes in fluid temperature and pressure within the wellbore; and constructing a reservoir multiphase flow model and a reservoir heat conduction process model.

[0072] Specifically, a thermodynamic behavior model of wellbore fluids is constructed:

[0073] ;

[0074] in, The temperature of the fluid inside the pipe. The initial temperature of the formation. A For relaxation distance, For quality flow, The specific heat capacity of the mixed fluid. It is the Joule-Thomson coefficient. The density of the oil-water mixture. For flow rate, For pressure, The well trajectory dip angle, The coefficient of friction, This represents the depth of the wellbore. It is the acceleration due to gravity; This represents a differential operator.

[0075] Relaxation distance A We adopt the empirical formula derived by Shiu and Begg based on statistical regression:

[0076] ;

[0077] in, arrive It is an empirical coefficient. For time, Where is the wellbore radius. The density of the formation oil, The specific gravity of the formation gas. The density of the liquid inside the wellbore. This refers to the wellhead pressure.

[0078] arrive The specific values ​​are shown in the table below:

[0079] .

[0080] Constructing a multiphase flow model for the reservoir:

[0081] The flow of oil, water, and gas phases within the reservoir satisfies the mass conservation equation:

[0082] ;

[0083] in, The porosity of the reservoir. For the sake of the prime minister The density of (such as oil phase, water phase, gas phase), For the sake of the prime minister saturation Components c In phase mass fraction in For absolute penetration rate, For the sake of the prime minister Severe (equal to) relative penetration Divide by fluid viscosity ), For the sake of the prime minister Pressure For the sake of the prime minister The flow rate, Components c In phase Source and sink items in the middle, This represents the gradient operator.

[0084] Constructing a reservoir heat conduction process model:

[0085] ;

[0086] in, For the sake of the prime minister Specific heat capacity, For the sake of the prime minister The isobaric coefficient of thermal expansion, For the sake of the prime minister The flow rate, For rock density, For the specific heat capacity of the rock, For temperature, For depth, is the thermal conductivity of the rock.

[0087] Step S42 Initializes the single-well spatial grid model. Further, step S42 includes: collecting various data related to the reservoir and well; performing spatial coordinate transformation on the structural surface and well trajectory data, so that the transformed well trajectories are located in the same... In the plane; a regular hexahedral mesh is generated based on the structural surface after spatial coordinate transformation to simulate the three-dimensional space of the reservoir and wellbore; initial attribute values ​​are assigned to each mesh cell; the mesh well index of the potential fluid production point location identified by blind source separation is set to 1, and the rest are set to 0, as the boundary conditions for subsequent simulations.

[0088] Specifically, the basic data preparation includes: collecting reservoir top and bottom structural surfaces, well trajectories, well logging interpretations (porosity, permeability, water saturation), fluid PVT parameters, wellbore structure data, fiber optic temperature profiles, and wellhead metering production data. Spatial coordinate transformation involves performing spatial coordinate transformations on the structural surface and well trajectory data to ensure the well trajectories are located as close as possible to the same coordinate plane. In the plane, simplified modeling is implemented. A hexahedral mesh is generated based on the transformed construction surface: 1. 1. The directional grid boundary conforms to the top and bottom structural surfaces; 2. Directional grid size ≤ 1 meter Direction 10–20 meters, 3. Considering the specific conditions of the reservoir, the direction should be ≤5 meters. Extending the direction a certain distance (e.g., 100 meters) to cover the effective reservoir area. Initialization attribute assignment: 1. Same Layer assignment corresponds to porosity and permeability; 2. Same 3. Assign cross-sectional values ​​to correspond to oil / water saturation; 4. Calculate initial pressure and temperature values ​​based on wellhead pressure and geothermal gradient. 5. Set well index for perforated areas: Set the grid well index of potential fluid production points identified by blind source separation to 1, and set the index of other locations to 0, as boundary conditions for subsequent simulations.

[0089] Step S43: Forward numerical simulation and DTS profile fitting. Further, step S43 includes: performing forward modeling based on a mathematical model of wellbore-reservoir fluid and heat conduction coupling and a single-well spatial grid model to simulate wellbore temperature and oil / water production profile yield at each time point; extracting wellbore temperature and oil / water production profile yield at corresponding time points based on DTS signal extraction and DAS signal truncation, using these as actual observation results; and comparing the results obtained from the forward modeling with the results obtained from actual observation based on an integrated progressive data assimilation method to optimize model parameters.

[0090] Specifically, the forward modeling strategy is as follows: 1. Spatial discretization: The continuous physical space is discretized into a finite number of grid cells for numerical computation. Central difference is used for fluid dynamics equations, and second-order central difference is used for heat conduction equations. 2. Temporal discretization: The continuous time process is discretized into a series of time steps for numerical simulation of time evolution. The Crank-Nicolson method can be used for time discretization. 3. Coupled numerical solution: In the coupled problem of fluid flow and heat conduction, fluid flow affects temperature distribution, and temperature changes affect fluid density and viscosity, thus affecting flow. Therefore, an alternating iterative coupling strategy of flow and heat transfer is used. This alternating iterative coupling strategy gradually approximates the solution of the coupled system by alternately solving the flow equation and heat conduction equation in each time step. 4. Output results, i.e., the goal of numerical simulation: During the numerical simulation, the wellbore temperature and oil and water production profiles of the well are calculated and recorded at each time step.

[0091] Temperature and yield inversion based on the Integrated Progressive Data Assimilation (ES-MDA) method: Model parameters are updated through multiple rounds of assimilation to fit DTS observation data. The specific steps are as follows:

[0092] 1. Select actual observation data: Select the temperature profile (wellbore temperature) and wellhead metering data (oil and water production profile) from the DTS signal corresponding to the DAS signal segment (multiple time points are possible), and set them as the actual observation data vector. ;

[0093] 2. Initialize model parameters: In the initialized single-well spatial grid model, set... Permeability and water saturation in different directions are used as adjustable variables, and an initial model parameter vector is constructed. ;

[0094] 3. Assimilation steps and expansion coefficient settings: Set the total number of data assimilation steps. and in the The expansion coefficient during secondary assimilation It satisfies the following relationship:

[0095] ;

[0096] 4. Number and definition of model set: Define the number of models in the set. And after each assimilation, the model set is recorded as follows:

[0097] ;

[0098] in, ;when hour, Indicates that based on the prior distribution from Initial model parameters obtained from sampling; when hour, Indicates the first After the second assimilation The parameters of each model;

[0099] 5. Perform forward modeling and disturbance handling: For arrive Perform the following steps:

[0100] A. Forward Model Calculation: Based on the established multi-field coupled calculation model, forward model calculations are performed on each model in the model set to obtain a set of prediction results:

[0101] ;

[0102] in, According to the model A vector composed of wellbore temperature and horizontal well oil and water production profile data obtained after forward modeling;

[0103] B. Perturb the actual observation data: Perturb the actual observation data and calculate the corrected observation data vector:

[0104] ;

[0105] in, A vector of actual observation data; To introduce Gaussian distributed random errors, it is usually derived from the standard normal distribution. Extracted from, among which It is an identity matrix, indicating that the errors are independent and identically distributed; The autocovariance matrix of the observation error;

[0106] C. Model Update: Based on the forward modeling results, each model in the model set is updated according to the following formula:

[0107] ;

[0108] in, For the first After the first assimilation The parameter vector of a model; For the first After the first assimilation The parameter vector of a model; Let be the covariance matrix of the model parameters; The covariance matrix of the data; The autocovariance matrix of the observation error;

[0109] 6. Calculate the mean of the model parameters after assimilation: This yields the mean value of the parameters after assimilation. The model set after sub-assimilation The mean of the model parameters is calculated based on this set:

[0110] ;

[0111] in, This represents the total number of models in the model set. In the first After the first assimilation, the second The parameter vector of a model, From 1 to , representing each model in the model set. The purpose of this formula is to calculate the result after completion. After one iteration of data assimilation, the average value of all model parameters in the model set is obtained by summing the parameter vectors of all models and then dividing by the total number of models. This is achieved by obtaining the average parameter vector. It can be used as the final model parameter estimate for model optimization.

[0112] Finally, the final explanation is obtained using a forward model: based on the mean of the model parameters. Optimize the model parameters, and use a forward modeling method based on the optimized model to calculate the final production profile of the horizontal well. .

[0113] Understandably, step S4 introduces spatial location constraints based on blind source analysis results in the DTS inversion, significantly reducing inversion uncertainty and improving the consistency and accuracy of the analytical results. By using the outlet point location extracted from the DAS signal as a constraint in the inversion process, it ensures that the temperature profile inversion results conform to the physical characteristics of the actual producing area.

[0114] Please see Figure 2-13 As shown, this application also provides specific application examples to help those skilled in the art fully understand the implementation process of this application. Taking the water intake profile of a water injection open-hole horizontal well in the Middle East as an example, the following steps are performed: Figure 1The method for identifying and inverting the production profile of oil and water wells, as shown, includes: Step S1: Processing DAS data and identifying the location of fluid points based on blind source analysis results. A distributed fiber optic system is deployed in the horizontal wellbore to acquire DAS and DTS signals. The well depth ranges from 2400 meters to 3100 meters, and the well trajectory is basically horizontal. The reservoir porosity averages 25%, with large permeability variations, indicating heterogeneity. The method is derived from the DAS signal response waterfall diagram (e.g., ...) throughout the water injection process. Figure 2 As shown in the figure, the horizontal axis represents time, the vertical axis represents depth, and the color represents signal amplitude. The figure shows the change of the root mean square (RMS) of the signal with frequency range from 1.0 Hz to 10.0 Hz over time and depth. After selecting a stable water injection section from the continuously acquired DAS data, a spatiotemporal Fourier transform (such as...) was used. Figure 3 As shown in the figure, the signals at various frequencies (100Hz, 10Hz, 3Hz, and 1Hz) are extracted, and their spatiotemporal distribution is as follows: Figure 4 , 5 As shown in 6 and 7 ( Figure 4-7 The horizontal axis represents the equivalent period T(s) in the frequency domain after the Fourier transform, i.e., the reciprocal of the frequency; the vertical axis represents the depth; and the color represents the signal amplitude. It can be seen that the low frequency (1Hz) exhibits more pronounced spatial distribution non-uniformity. After comprehensive analysis, the spatiotemporal distribution map of 0.5Hz is extracted, as shown below. Figure 8 As shown in the figure (the horizontal axis represents the equivalent period T(s) in the frequency domain after Fourier transform, which is the reciprocal of the frequency, the vertical axis represents the depth, and the color represents the signal amplitude).

[0115] Step S2: The extracted spatiotemporal signals are analyzed using blind source separation to obtain local acoustic signals at various depths, such as... Figure 9 As shown, the horizontal axis represents the frequency domain after Fourier transform, and the vertical axis represents the signal amplitude. Red represents the interference (measured) results, and blue represents the extracted local acoustic signal.

[0116] Step S3: Further form as Figure 10 The local acoustic spatiotemporal distribution at each depth obtained after DAS blind source separation is shown in the figure (the horizontal axis represents the equivalent period T(s) in the frequency domain after Fourier transform, i.e., the reciprocal of the frequency; the vertical axis represents the depth; and the color represents the signal amplitude). Based on its 0.5Hz acoustic amplitude intensity and statistical test method (assumption: whether the acoustic amplitude intensity at this depth is abnormal), the location of the liquid outlet is determined.

[0117] Step S4: Construct the numerical simulation model. Construct a single-well geological model and a numerical simulation model. The geological model includes formation pressure, porosity, permeability, water / oil saturation, and temperature. Import parameters such as PVT and relative permeability curves to establish a single-well pipe flow-seepage numerical simulation model, such as... Figure 11As shown. Based on the results of blind source analysis and statistical analysis, the locations of the liquid outlets are identified, and perforation sections are set up.

[0118] The ES-MDA data assimilation method is used to fit temperature profiles and wellhead water cut. Profile data corresponding to the observation time are selected on the DTS temperature waterfall plot, and the ES-MDA method is used to assimilate the observed data at the selected time, such as... Figure 12 As shown (horizontal axis represents temperature, vertical axis represents depth), black dots represent observed true values, red x-axis represents the average or best estimate calculated by the ES-MDA method, gray dots represent the predicted values ​​of each model in the ensemble, and the red part is the DTS temperature curve calculated under the average parameters. The final water absorption profile calculated by the average model after data assimilation is shown below. Figure 13 As shown, the data includes the cumulative water intake profile and the water intake along the wellbore profile. The vertical axis represents the depth along the horizontal well. The horizontal axis of the right figure represents the total injection volume at that depth during the observation period. The horizontal axis of the left figure represents the sum of the total injection volume at each point calculated along the depth direction (from bottom to top). It is worth noting that the total injection volume of the entire well during the observation period is the total injection volume at the outlet (the maximum value).

[0119] It is understood that the oil-water well production profile identification and inversion method provided in this application extracts local (low-frequency, ultra-low-frequency) production signals through blind source separation of DAS signals, identifies the liquid point location through statistical screening, and then uses it as a constraint condition to guide the inversion of distributed temperature sensing (DTS), thereby achieving joint and accurate interpretation of oil-water well production profiles. This invention is not only applicable to the accurate identification and interpretation of production profiles of various oil-water wells, but can also be applied to oil-water wells with different flow rates and different fluid types, and has great technical promotion value and market potential. It is especially suitable for oil-water two-phase wells and oil-water-gas composite wells with low flow rates, and has unique advantages in accurate downhole monitoring of such wells.

[0120] Please see Figure 14 As shown, Figure 14 This is a structural diagram of the oil and water well production profile identification and inversion system provided in an embodiment of this application. The water well production profile identification and inversion system is used to achieve, for example... Figure 1 The method for identifying and inverting the production profile of oil and water wells is shown. The system includes: a signal acquisition module for acquiring DAS and DTS signals based on a distributed optical fiber system deployed in the wellbore, and preprocessing and extracting low-frequency signals from the DAS signals; a signal separation module for performing blind source separation on the extracted DAS signals and extracting signals from each independent source; a production point identification module for identifying potential production point locations based on statistical screening of each independent source signal; and a DTS inversion module for performing DTS inversion with the potential production point locations as constraints, and obtaining the final interpretation result based on a forward modeling solution strategy.

[0121] This application also provides a computer program product that, when run on an electronic device, causes the electronic device to perform actions such as... Figure 1 The method for identifying and inverting the production profile of oil and water wells is shown.

[0122] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for identifying and inverting the production profile of oil and water wells, characterized in that, include: Step S1: Acquire DAS and DTS signals using a distributed optical fiber system based on wellbore layout, and preprocess and extract low frequencies from the DAS signals. Step S2: The extracted DAS signal is subjected to mean removal processing, and the signal after mean removal processing is subjected to whitening processing; the number of independent sources is determined based on the signal after whitening processing, which is taken as the number of mutually independent effluent points; the number of independent sources is substituted into the independent component analysis algorithm to perform blind source separation on the extracted signal to obtain the signals of each independent source. Step S3: Calculate the low-frequency energy index or amplitude index of each independent source signal, distribute the low-frequency energy index or amplitude index along the depth direction to obtain the sound wave intensity distribution along the depth; based on the sound wave intensity distribution along the depth and combined with the set statistical threshold or hypothesis testing method, screen out the spatial locations where the sound wave intensity exceeds the sound wave intensity threshold as potential liquid outlet locations. Step S4: Construct a mathematical model of wellbore-reservoir fluid and heat conduction coupling to describe the temperature and pressure changes of the fluid in the wellbore, as well as the flow and heat conduction processes of multiphase fluids within the reservoir; construct and initialize a single-well spatial grid model, setting the grid well index of potential fluid production points identified by blind source separation to 1, and setting the rest to 0, as boundary conditions for subsequent simulations; perform forward modeling based on the mathematical model of wellbore-reservoir fluid and heat conduction coupling and the single-well spatial grid model to simulate wellbore temperature and oil and water production profiles at various times; extract wellbore temperature and oil and water production profiles at corresponding times based on DTS signal extraction and DAS signal truncation segments, as actual observation results; Based on the integrated progressive data assimilation method, the results obtained from forward modeling are compared with the results obtained from actual observations, and the model parameters are optimized. The final oil and water production profile of the horizontal well is calculated based on the model with optimized parameters.

2. The method for identifying and inverting production profiles of oil and water wells according to claim 1, characterized in that, Step S1 includes: DAS signals are obtained by discretely sampling acoustic signals with spatial and time intervals. Based on the two-dimensional spatiotemporal Fourier transform, the DAS signal is decomposed into signals of different frequencies; Filter signals of different frequencies to extract the low-frequency band signals.

3. The method for identifying and inverting production profiles of oil and water wells according to claim 1, characterized in that, A mathematical model coupling wellbore-reservoir fluid and heat conduction is constructed to describe the temperature and pressure changes of the fluid within the wellbore, as well as the flow and heat conduction processes of the multiphase fluid within the reservoir. A thermodynamic behavior model of wellbore fluids is constructed to describe the changes in fluid temperature and pressure within the wellbore. Construct a reservoir multiphase flow model and a reservoir heat conduction process model.

4. The method for identifying and inverting production profiles of oil and water wells according to claim 1, characterized in that, A single-well spatial grid model was constructed and initialized. The grid well index for potential fluid outlet locations identified by blind source separation was set to 1, and the index for other locations was set to 0. These were used as boundary conditions for subsequent simulations. Collect various data related to reservoirs and wells; Spatial coordinate transformation was performed on the structural surface and well trajectory data, and the transformed well trajectories are located in the same... In the plane; Based on the structural surfaces after spatial coordinate transformation, a regular hexahedral mesh is generated to simulate the three-dimensional space of the reservoir and wellbore; Assign initial attribute values ​​to each grid cell; The grid well index of the potential effluent point location identified by blind source separation is set to 1, and the index of the remaining locations is set to 0, as the boundary condition for subsequent simulations.

5. A system for identifying and inverting the production profile of oil and water wells, characterized in that, A method for identifying and inverting the production profile of an oil and water well as described in any one of claims 1-4, comprising: The signal acquisition module is used to acquire DAS signals and DTS signals based on a distributed optical fiber system arranged in the wellbore, and to preprocess and extract low-frequency signals from the DAS signals. The signal separation module is used to perform blind source separation on the extracted DAS signal and extract each independent source signal; The liquid outlet identification module is used to identify potential liquid outlet locations by combining statistical filtering based on signals from various independent sources. The DTS inversion module is used to perform DTS inversion with the potential effluent point location as a constraint, and obtain the final interpretation result according to the forward modeling solution strategy.

6. A computer program product, characterized in that, When the computer program product is run on an electronic device, the electronic device performs a method for identifying and inverting the production profile of an oil and water well as described in any one of claims 1-4.

Images