A method for inverting deep water drive front in oil reservoir based on dynamic response of natural potential

Abstract

The application discloses a kind of oil reservoir deep water drive front inversion methods based on natural potential dynamic response, belong to petroleum development engineering and geophysical monitoring technical field.The method first constructs the long-distance water drive oil physical simulation experiment of target reservoir, establishes the natural potential distribution of unimpinged area in front of water drive front in oil-water two-phase displacement process, and further reveals that the field strength gradient coefficient and intercept coefficient in the law follow the certain quadratic polynomial evolution characteristics with front position, and accordingly constructs generalized space-time evolution analytical model.In field application, real-time potential response is obtained using electrode array arranged on production well, and the monitoring data is substituted into the above-mentioned model to construct a quadratic characteristic equation about water drive front position, and the effective root is screened by solving the equation and combining reservoir physical boundary, to realize accurate positioning of deep water drive front.The application breaks through the limitation of existing technology, which depends on complex numerical simulation or only studies single-phase flow, and realizes quantitative inversion of "measuring single point to know far end" using clear physical field analytical solution, with the advantages of high calculation efficiency, high positioning accuracy and the like.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas development engineering and geophysical monitoring technology, specifically relating to a method for monitoring and inverting the location of the deep water drive front of water-injected oil reservoirs using natural potential signals. Background Technology

[0002] In oilfield water injection development, accurate monitoring of the migration location of the water drive front is crucial for controlling water cut rise and optimizing injection-production strategies. Currently, mainstream monitoring technologies include 4D seismic imaging, inter-well electromagnetic imaging, and distributed fiber optic sensing. However, these technologies have certain limitations: while 4D seismic and inter-well electromagnetic imaging offer high spatial resolution, their acquisition costs are high and discontinuous, resulting in monitoring blind spots; while distributed fiber optic sensing can provide real-time monitoring, its detection range is mainly limited to the vicinity of the wellbore, making it difficult to detect fluid changes at deeper levels within the wellbore.

[0003] The self-potential (SP) method, as a low-cost, passive monitoring technique, utilizes the potential signals generated by underground fluid flow for monitoring and has broad application prospects. However, existing SP-based methods are mostly focused on single-phase flow research and numerical simulation, lacking analytical solutions for physical models of oil-water two-phase displacement processes. Furthermore, complex electromagnetic interference in the field environment often results in low signal-to-noise ratios for single-point measurements, making it difficult to guarantee the accuracy of inversion. Therefore, there is an urgent need for a method based on well-defined physical laws that can characterize the transport features of two-phase flows. Summary of the Invention

[0004] The purpose of this invention is to provide a method for inverting the location of the water-drive front based on a generalized evolution model of the natural potential field. By constructing a clear logarithmic-quadratic spatiotemporal evolution mathematical model and combining it with production well electrode array acquisition technology, a low-cost, quantitative location of the deep water-drive front of the reservoir can be achieved.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for inverting the position of the water-drive front based on a generalized evolution model of the natural potential field, comprising the following steps: Step S1: Construct a baseline evolution model based on physical simulation principles: Construct a long-distance waterflooding physical model simulating the target reservoir medium, and collect spontaneous potential data along the unaffected area ahead of the waterflooding front at different times during the waterflooding process; based on the experimental data, determine the spontaneous potential at any point within the unaffected area. Depending on the distance between that point and the water injection end Follows the logarithmic decay law Further extract data at different propulsion times to establish the field strength gradient coefficient in the logarithmic decay law. With intercept coefficient Water-driven leading edge position The nonlinear evolution function relationship is used to construct a coefficient that varies with the leading edge position. and A generalized spatiotemporal evolution model.

[0006] Step S2: Acquire real-time monitoring data of the production well to be tested: In the reservoir area to be tested, determine the relative spatial coordinates of the injection well and the production well. To improve the signal-to-noise ratio and representativeness of the monitoring data, an electrode array is deployed on the production well. The real-time spontaneous potential response of the production well relative to the surface reference electrode or wellhead reference point is obtained using the electrode array, and the real-time spontaneous potential value used for inversion is determined based on the array data. and the corresponding spatial distance .

[0007] Step S3: Perform inverse inversion based on single-point signal: convert the real-time natural potential value... and spatial distance Substituting the generalized spatiotemporal evolution model constructed in step S1, we construct a model for the unknown water-driven front position. The quadratic characteristic equation is obtained; by solving the quadratic characteristic equation and selecting the effective roots according to the physical boundary conditions, the position of the water-drive front in the deep reservoir at the current moment is obtained. .

[0008] Further, in step S1, the coefficient , Water-driven leading edge position The nonlinear evolution function relationship is specifically limited to a quadratic polynomial relationship: in, and These are constants that characterize the electrical properties of a specific reservoir medium, determined through regression analysis of physical simulation experimental data.

[0009] Furthermore, the general state equation of the generalized spatiotemporal evolution model is: ;in, When the water drive leading edge is located Predicting spatial location at time The natural potential value generated at that location.

[0010] Furthermore, the quadratic characteristic equation in step S3 is specifically constructed as follows: By rearranging the equation, we obtain information about... The standard form of a quadratic equation in one variable: .

[0011] Furthermore, the solution to the quadratic characteristic equation includes a valid solution screening step: calculating the two real roots of the equation and eliminating invalid solutions based on reservoir physical boundary conditions; the physical boundary conditions include: the calculated leading edge position. It must be greater than 0 and less than the injection-production well spacing. (Right now If both roots are within the range, then the root that conforms to the continuity of fluid transport is selected as the effective solution by combining the leading edge position data at historical moments.

[0012] Further, step S2, which involves deploying an electrode array on the production well, specifically includes: deploying multiple measuring electrodes longitudinally along the wellbore. The determination of the real-time natural potential value for inversion based on the array data... Specifically, this includes: performing signal-to-noise ratio (SNR) analysis on the multi-channel signals acquired by the electrode array, and selecting the electrode data with the highest SNR as the... Alternatively, a weighted average can be calculated on the measurements of the electrode array within the target reservoir depth range, and the average value can be used as the... .

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. Breaking through the limitations of single-phase flow and theoretical innovation: This invention, targeting the natural potential response characteristics in the oil-water two-phase displacement process, reveals for the first time the physical law of the evolution of the potential field coefficient as a quadratic function with the leading edge position, and establishes a clear analytical mathematical model, making up for the shortcomings of existing research which mostly focuses on single-phase flow or relies on complex numerical simulations.

[0014] 2. Strong anti-interference capability: By introducing production well electrode array technology and optimizing the processing of multi-channel data (such as signal-to-noise ratio filtering or weighted averaging), the interference of on-site environmental noise is effectively overcome, and the accuracy of inversion input is improved.

[0015] 3. High computational efficiency and accurate positioning: It enables the rapid quantitative back-calculation of the invisible groundwater drive front position by using the potential signal of a single monitoring point (or array comprehensive value) at a remote end (production well) and solving a quadratic equation. Furthermore, it effectively eliminates mathematical pseudo-solutions through physical boundary constraints, resulting in high positioning accuracy and facilitating real-time application in the field. Attached Figure Description

[0016] Figure 1 This is a schematic flowchart of the quantitative inversion method for the water-drive front position described in this invention; Figure 2 This is a schematic diagram of the long core multi-electrode experimental device used to establish a benchmark model in Embodiment 1 of the present invention; Figure 3 This is a graph showing the potential distribution and logarithmic fitting of the unaffected area in front of the leading edge at different propulsion times in Example 1. Figure 4 This is a schematic diagram of the fitting relationship between the field gradient coefficient A and the position of the water-driven leading edge in Embodiment 1 of the present invention; Figure 5 A schematic diagram of the fitting relationship between the intercept coefficient B and the position of the water-drive leading edge in Embodiment 1 of the present invention; Figure 6 This is a schematic diagram of the electrode array layout and inversion principle of the production well in Embodiment 3 of the present invention. Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0018] Example 1: This example aims to establish the evolution law of the natural potential field of the target reservoir through physical simulation experiments and obtain the characteristic constants required by the generalized model.

[0019] 1. Experimental setup and procedure: such as Figure 2 As shown, a multi-electrode monitoring model for long core samples was constructed. The main body of the model consisted of a 50cm long, 2cm inner diameter acrylic sand-filled tube, simulating the target reservoir medium. Measurement electrodes (E1-E10) were placed every 5cm along the axial direction, with the outlet electrode E10 serving as the reference electrode. A waterflooding experiment was conducted, and the electrodes were sequentially moved to a certain position at the waterflooding front (denoted as...). Simultaneously record the potential data of each electrode in the unaffected area in front relative to the reference point.

[0020] 2. Establish a spatial logarithmic decay model: Extract potential distribution data of the unaffected area ahead when the water-drive leading edge reaches different positions. For example... Figure 3 As shown, the experimental data indicate that the potential With distance Strictly follows the logarithmic decay law: Fit the data at different times and obtain the correlation coefficient. All values ​​are greater than 0.89, verifying the universality of this pattern.

[0021] 3. Construct a generalized spatiotemporal evolution model: Further analyze the coefficients in the above logarithmic model. and Changes that occur as the leading edge advances. For example... Figure 4 As shown, the field strength gradient coefficient Following the position of the leading edge It exhibits quadratic growth: In this embodiment, .like Figure 5 As shown, the intercept coefficient Following the position of the leading edge It exhibits a quadratic function descent: In this embodiment, Substituting the above relationships into the fundamental logarithmic equation, we obtain a generalized spatiotemporal evolution model applicable to this reservoir condition: .

[0022] Example 2: This example uses the experimental data obtained in Example 1 to simulate the working condition of "known monitoring point potential and unknown leading edge position" to verify the inversion accuracy of the generalized evolution model proposed in this invention.

[0023] 1. Verification object: Select the E5 electrode located in the middle of the long core model (actual location). E5 was chosen as the verification point because it is located at the geometric center of the model, is least affected by boundary effects, and can represent the typical characteristics of deep reservoirs. At the same time, the potential signal generated at the leading edge at this point is significantly attenuated when it is transmitted to the reference point, and the signal is weak, which can effectively test the model's ability to resolve weak signals.

[0024] 2. Inversion process: Assuming only the location is known The real-time potential signal measured at this location And the leading edge position Unknown. Substituting the known conditions into the generalized spatiotemporal evolution model determined in Example 1, a model for... The characteristic equation of a quadratic variable: .

[0025] 3. Result Analysis: Solving the above equations yields the calculated solution. Compare the calculated solution with the actual physical location ( The absolute deviation was 1.4 cm, and the relative error was 6.2%.

[0026] This example demonstrates that the generalized model constructed based on full-process data has high reliability and can achieve high-precision positioning of the far-end water-drive leading edge using a single-point weak electrical signal, with the relative error controlled within the allowable range for engineering applications, thus verifying the effectiveness of the method.

[0027] Example 3: This example aims to illustrate how to combine the experimentally verified theoretical model and method with electrode array technology to apply to the field monitoring of actual oil reservoirs.

[0028] 1. The deployment of the on-site monitoring system, such as Figure 6 As shown, taking a specific injection-production well group as an example: Coordinate definition: The wellhead of the injection well is set as the origin of the coordinate system. The straight-line distance between the production well and the injection well is... (For example ).

[0029] Electrode array deployment: To overcome the complex electromagnetic noise environment at the site, an electrode array is deployed on one side of the production well. Deployment methods may include: arranging multiple induction electrodes longitudinally outside the production well casing or in the target layer. ).

[0030] Reference point: Set the point at infinity on the ground surface or the stable potential point at the wellhead as the common reference electrode.

[0031] 2. Data Acquisition and Preprocessing: During monitoring, the natural potential response data of each electrode in the array are acquired synchronously. This is to obtain high-quality input values ​​for inversion. The following strategy is used to process array data: Signal-to-noise ratio (SNR) optimization: The SNR of each channel signal is calculated in real time, and channels affected by sudden interference (such as lightning or industrial current) are automatically eliminated, and the channel data with the highest SNR is selected.

[0032] Weighted average: A weighted average of the measurements from all valid channels in the array. This is to eliminate random electrodynamic noise generated by fluid flow in the wellbore.

[0033] 3. Inversion Calculation and Effective Solution Screening: Assuming that the characteristic constants of this block have been determined through formation parameter calibration (similar to the process in Example 1), Spatial distance of production wells and the processed potential value Substituting into the generalized model equation: ; Organize and solve the problem about A quadratic equation in one variable. Physical boundary screening: The equation may have two mathematical solutions, and invalid solutions need to be eliminated by considering the physical boundary conditions.

[0034] Spatial constraints: Leading edge position Must meet .

[0035] Timing constraints: the current time step It should be greater than the previous moment. (This conforms to the unidirectional propulsion law of fluid). Through the above steps, the quantitative inversion of the position of the deep underground water drive front can be achieved at the production wellhead.

[0036] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for quantitative inversion of the position of the water-drive front based on a generalized evolution model of the natural potential field, characterized in that, Includes the following steps: Step S1: Construct a baseline evolution model based on physical simulation principles: Construct a long-distance waterflooding physical model simulating the target reservoir medium, and collect the spontaneous potential data along the unaffected area ahead of the waterflooding front at different times during the waterflooding process; based on the data, determine the spontaneous potential at any point within the unaffected area. Depending on the distance between that point and the water injection end Follows the logarithmic decay law: Extract data at different propulsion times and establish the field strength gradient coefficient in the logarithmic decay law. With intercept coefficient Water-driven leading edge position The nonlinear evolution function relationship is used to construct a coefficient that varies with the leading edge position. and A generalized spatiotemporal evolution model; Step S2: Obtain real-time potential monitoring data of the production well to be tested: In the reservoir area to be tested, determine the relative spatial coordinates of the injection well and the production well. Multiple measuring electrodes are arranged longitudinally along the wellbore of the production well to form an electrode array. The real-time spontaneous potential response of the production well relative to a surface reference electrode or wellhead reference point is obtained using the electrode array. Based on the array data, the real-time spontaneous potential value used for inversion is determined. and the corresponding spatial distance Among them, the real-time natural potential value is determined. The method includes: performing signal-to-noise ratio (SNR) analysis on the multi-channel signals acquired by the electrode array, and selecting the electrode data with the highest SNR as... Alternatively, a weighted average can be calculated on the measurements of the electrode array within the target reservoir depth range, and the weighted average can be used as the... ; Step S3: Perform inverse inversion based on single-point signal: convert the real-time natural potential value... and spatial distance Substituting the generalized spatiotemporal evolution model constructed in step S1, we construct a model for the unknown water-driven front position. The quadratic characteristic equation is used to determine the current position of the water-drive front in the deep reservoir. .

2. The method according to claim 1, characterized in that, In step S1, the coefficient , Water-driven leading edge position The nonlinear evolution function relationship is specifically limited to a quadratic polynomial relationship: Where a1, b1, c1, a2, b2, c2 are constants that characterize the electrical properties of a specific reservoir medium, determined by regression analysis of physical simulation experimental data.

3. The method according to claim 2, characterized in that, The general state equation of the generalized spatiotemporal evolution model is: ;in, When the water drive leading edge is located Predicting spatial location at time The natural potential value generated at that location.

4. The method according to claim 3, characterized in that, The quadratic characteristic equation in step S3 is specifically constructed as follows: By rearranging the equation, we obtain information about The standard form of a quadratic equation in one variable: .

5. The method according to claim 4, characterized in that, Solving the quadratic characteristic equation further includes an effective solution screening step: calculating the two real roots of the equation and eliminating invalid solutions based on reservoir physical boundary conditions; the physical boundary conditions include: the calculated leading edge position. It must be greater than 0 and less than the injection-production well spacing. ,Right now If both roots are within the range, then the root that conforms to the continuity of fluid transport is selected as the effective solution by combining the leading edge position data at historical moments.

6. The method according to claim 1, characterized in that, In step S1, the physical model is a long core sand-filled tube model or a long core holder model, and multiple non-polarized electrodes are arranged along the axial direction. The non-polarized electrodes are used to synchronously collect the natural potential response at different positions during the water-drive front advancement process.

Images