Water plugging well selection and layer selection evaluation method considering seepage time-varying characteristics

Abstract

The invention discloses a water shutoff well selection and layer selection evaluation method considering seepage time-varying characteristics, and solves the problem that a traditional static evaluation method cannot track the dynamic evolution of oil reservoir seepage. The model collects and processes multi-dimensional data of an oil reservoir, constructs a fuzzy evaluation matrix, generates a dynamic weight by adopting a time-varying game theory-dynamic CRITIC combined weighting method, updates a weight variation coefficient in real time by combining a seepage time-varying state space model and Kalman filtering, and constructs a multi-period association evaluation matrix by introducing a time delay factor. And finally, establishing a dynamic evaluation model to realize time-varying prediction and evaluation of the water plugging effect. According to the model, double-time-dimension modeling and a dynamic coupling mechanism are fused, the accuracy of water shutoff, well selection and layer selection of the low-permeability reservoir is improved, and the model has engineering application value and technical innovation.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploitation, and in particular to a water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage. Specifically, it is an evaluation model technology for accurately screening oil wells and layers suitable for implementing water plugging measures during the development process of an oil reservoir, taking into account the dynamic changes of reservoir seepage characteristics over time. This technology runs through the middle and late stages of oil reservoir exploitation. By deeply analyzing and processing various oil reservoir data, it provides a scientific and efficient decision-making basis for water plugging operations, so as to improve the oil recovery rate of the oil reservoir and optimize the exploitation benefits. Background Art

[0002] In the field of oil exploitation, water plugging operations are of crucial significance for improving the oil recovery rate of oil wells and optimizing the development effect of oil reservoirs. The accuracy of well and layer selection for water plugging plays a decisive role in the measure effect, and the scientificity and accuracy of its decision-making are directly related to the success or failure of water plugging measures and economic benefits.

[0003] Traditional water plugging well and layer selection evaluation methods mostly focus on the consideration of static geological parameters and production data, such as indicators like the permeability, porosity, current water cut of the oil reservoir, and previous oil production. With the progress of oil reservoir exploitation, a large number of practices and studies have shown that the seepage characteristics inside the oil reservoir are not constant. Taking the mined rock mass as an example, affected by mining activities, the pore and fracture structures of coal mine surrounding rocks will change significantly, thereby causing its seepage characteristics and seepage field to show dynamic change characteristics. This change does not occur instantaneously but has obvious time-varying characteristics. The change of the seepage field will in turn affect the displacement and stress fields of the surrounding rock, prompting the creep characteristics of the rock mass to change. There is a complex and close strong coupling effect between the two. In the oil reservoir environment, this seepage time-varying characteristic also widely exists. As the exploitation time goes by, the oil reservoir pressure continuously drops, the rock pore structure gradually deforms due to stress changes, and coupled with the long-term interaction between the fluid and rock minerals, the size, shape, and connectivity of the seepage channels continue to change, resulting in obvious changes in seepage laws over time.

[0004] However, existing water plugging well and layer selection evaluation models have exposed many problems in practical applications due to the failure to fully consider the seepage time-varying characteristics. On the one hand, there are deviations in the evaluation of the water plugging potential of wells and layers. Some well layers judged to be suitable for water plugging based on static parameters, after implementing water plugging measures, due to the dynamic changes of seepage conditions, fail to achieve the expected oil production increase and water cut reduction effect, and may even cause new water channeling due to the change of the seepage field, resulting in a further decline in the productivity of oil wells. On the other hand, there is a lack of forward-looking prediction of seepage changes during the long-term development process of the oil reservoir, making it difficult for the water plugging plan to adapt to the dynamic evolution of the oil reservoir and unable to achieve continuous and efficient oil reservoir development management.

[0005] It is urgent to develop a water plugging well and layer selection evaluation model that can fully consider the time-varying characteristics of seepage. For the first time, the time-varying seepage theory is combined with fuzzy evaluation to construct a two-time-dimensional evaluation model of "dynamic contingency coefficient + time series prediction". This model will fill the current technical gap, provide a more accurate and scientific decision-making basis for water plugging operations, and effectively improve the effectiveness of water plugging measures and the overall benefit of reservoir development. Summary of the Invention

[0006] 1. A method for evaluating water plugging well and layer selection considering the time-varying characteristics of seepage, characterized by comprising the following steps:

[0007] S1. Construct an influencing factor index system: Collect the basic geological data, production dynamic data of each well in the target reservoir area, and water injection-related data such as injection water pressure and injection volume. Clean the collected data, remove outliers and incorrect data, and perform standardization processing to make different types of data comparable.

[0008] S2. Fuzzification processing: Define corresponding fuzzy language variable sets for each index according to its characteristics, and use the trapezoidal membership function to determine the membership degrees of each index data on different fuzzy language variables.

[0009] S3. Determine the initial weight: Use the time-varying game theory - dynamic CRITIC combined weighting method to generate the basic dynamic weight w base (t), and its variation law includes the seepage time decay factor and the sliding window analysis of dynamic CRITIC.

[0010] S4. Establish a seepage time-varying state space model: Based on the seepage mechanics theory, combined with the reservoir geological characteristics and production dynamic data, establish a seepage model considering time-varying characteristics.

[0011] S5. Use the Kalman filter to update the contingency coefficient in real time: Use the Kalman filter algorithm to update the contingency coefficient in real time, and track the time-varying law of the reservoir seepage characteristics by inputting the reservoir dynamic monitoring data.

[0012] S6. Construct a multi-period correlation evaluation matrix: Introduce a time delay factor into the fuzzy evaluation model to construct a "multi-period correlation evaluation matrix" to quantify the dynamic correlation relationship between the water plugging effect and the well and layer selection parameters at different time nodes;

[0013] S7. Application and result output of the dynamic evaluation model: Based on the time-varying contingency coefficient and the multi-period correlation evaluation matrix, establish a dynamic evaluation model of the water plugging effect to realize the time-varying prediction and evaluation of the water plugging effect in low-permeability reservoirs.

[0014] 2. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the step of constructing the influencing factor index system, a multi-dimensional and multi-layer index system of reservoir geomechanics and two-phase seepage with sixteen indexes is comprehensively considered. Specifically, it includes: formation pressure, formation permeability, Young's modulus, Poisson's ratio, brittleness index, in-situ stress difference coefficient, rock compressibility, crude oil viscosity, water saturation, production time, injection pressure, injection volume, fracture development degree, reservoir heterogeneity, breakthrough index, injection-production correlation index, etc.

[0015] In step S1, the water flooding degree index is determined by comparing the change range and growth rate of water saturation in the oil layer within different time periods; the calculation of the permeability variation coefficient considers the distribution characteristics of permeability data at different time points to reflect the time-varying situation of permeability heterogeneity.

[0016] In the formula, S w,j (t) is the water saturation of the j-th layer at time t, %.

[0017] In the formula, C v,j (t) is the permeability variation coefficient of the j-th layer at time t, dimensionless; σ k,j (t) is the permeability standard deviation of the j-th layer at time t, mD; μ k,j (t) is the permeability mean value of the j-th layer at time t, mD.

[0018] 3. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the fuzzy processing step, the trapezoidal membership function is used to calculate the membership degree, and an initial fuzzy evaluation matrix is constructed:

[0019] When the index is unimodal distribution, let b = c, at this time the function degenerates into a triangular form, and the peak point is b.

[0020] 4. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that the time-varying game theory - dynamic CRITIC combined weighting method is used to generate the basic dynamic weight w base (t), and its variation law includes the seepage time decay factor and the sliding window analysis of dynamic CRITIC.

[0021] Based on the sliding time window analysis of the objective weight of the index:

[0022] Calculation of index conflict:

[0023] In the formula, C j is the conflict of the j-th index, dimensionless; σ j is the standard deviation of the j-th index within the sliding window, with the unit same as the original dimension of the index. For example, the unit of permeability is mD; r jk is the Pearson correlation coefficient between index j and index k within the sliding window, dimensionless; m is the total number of indices.

[0024] Dynamic update: A sliding time window T is adopted, and the data within the window is updated at a time step of Δt (30 d). The window width is determined according to the seepage response time (3 months is taken here).

[0025] Seepage response factor: α j (t) = e -β·ΔP(t) / μ

[0026] In the formula, α j (t) is the seepage response factor of the j-th index at time t, dimensionless; β is the permeability attenuation coefficient, MPa-1; ΔP(t) is the change in the average pressure gradient within the time window, MPa; μ is the fluid viscosity, mPa·s.

[0027] Dynamic CRITIC objective weight:

[0028] In the formula, is the dynamic CRITIC weight of the j-th index, dimensionless.

[0029] Introduce a seepage time decay factor to correct the subjective weight:

[0030] Seepage time decay factor:

[0031] In the formula, γ(t) is the time decay factor, dimensionless; γ0 is the initial decay coefficient, dimensionless; k d is the decay rate, d-1; t is the development time, d.

[0032] Corrected judgment matrix: A(t) = S(t) ⊙ γ(t) + (1 - γ(t)) · A0

[0033] In the formula, S(t) is the time-sensitive matrix; A(t) is the time-varying judgment matrix; A0 is the initial judgment matrix (static AHP result)

[0034] Solve the eigenvector corresponding to the largest eigenvalue of A(t) by the eigenvalue method, and normalize it to obtain the subjective weight

[0035] Solve the optimal combination of subjective and objective weights and embed a time-varying adjustment mechanism. The game theory combination optimization objective function is as follows:

[0036] In the formula, the first term is the difference between subjective and objective weights; the second term is the weight change rate penalty term; λ(t) is the time-varying adjustment coefficient, d - 1; ΔQ(t) is the cumulative liquid production rate change, m 3 / d; Q ref is the reference liquid production volume, m 3 .

[0037] Solve using the Lagrange multiplier method to obtain the Nash equilibrium weights

[0038] In the formula, δj(t) is the weight change trend term, calculated by the moving average of historical data.

[0039] Through the time-varying game theory - dynamic CRITIC model, couple the seepage time decay factor and the index conflict analysis to generate the initial weights that conform to the dynamic laws of the reservoir.

[0040] 5. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the step of analyzing the time-varying characteristics of seepage, when establishing the seepage model considering time-varying characteristics, consider the compressibility of the fluid, the deformation characteristics of the rock, and the interaction of oil-water two-phase seepage, and more accurately describe the time-varying process of fluid seepage in the reservoir by introducing appropriate mathematical equations and parameters.

[0041] Based on Darcy's law and the material balance equation, establish a time-varying seepage model considering fluid compressibility and rock deformation, and the control equation is:

[0042] In the formula, k(t) is the permeability varying with time, mD; μ is the fluid viscosity, mPa·s; p(t) is the reservoir pressure at time t, MPa; φ is the porosity, %; c t is the comprehensive compressibility coefficient, MPa -1 .

[0043] Adopt the state space model to construct the time-varying equation of seepage parameters, take seepage parameters such as permeability and pressure as state variables, and establish the state equation: x k = A k x k-1 + B k u k + w k

[0044] In the formula, x k is the state variable vector at the k-th moment; A k is the state transition matrix, reflecting the time autocorrelation of the state variables; B k is the input matrix; u k is the input vector; w k is the process noise vector. x k = [k k , p k , G k T

[0045] In the formula, k k is the permeability at the k-th moment, mD; p k is the formation pressure at the k-th moment, MPa; G k is the starting pressure gradient at the k-th moment, MPa·m -1 ; u k = [Q ik , p k T

[0046] In the formula, Q ik is the water injection intensity, m 3 / d; p k is the production pressure difference, MPa.

[0047] 6. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the step of real-time updating of the Kalman filter weight coefficient, the initial weight determined in step S3 is used as the initial value of the weight coefficient, and the Kalman filter algorithm is used to update the weight coefficient in real time. By inputting the reservoir dynamic monitoring data, the prior estimate value of the weight coefficient is predicted through the time update step, and then the posterior estimate value is corrected by using the monitoring data and the Kalman gain through the measurement update step, so as to realize the dynamic tracking of the reservoir stress sensitivity effect and the change of the starting pressure gradient.

[0048] Predict the prior estimate value of the weight coefficient through the state equation

[0049] Introduce the seepage parameter data z k monitored in real time, and construct the measurement equation: z k = H k x k + v k

[0050] In the formula, H k ​​is the measurement matrix, v k is the measurement noise vector.

[0051] The Kalman gain K k :

[0052] where is the prior estimate covariance matrix, R k is the measurement noise covariance matrix.

[0053] Using the Kalman gain to correct the prior estimate value to obtain the posterior estimate value of the contingency coefficient

[0054] By continuously iterating the above process, the contingency coefficient is updated in real time so that it can track the dynamic evolution of the reservoir seepage characteristics. This process ensures that the contingency coefficient can be adjusted in a timely manner according to the actual situation of the reservoir, laying a solid foundation for accurately constructing the multi-period correlation evaluation matrix and applying the dynamic evaluation model subsequently.

[0055] 7. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the step of constructing the multi-period correlation evaluation matrix, a time delay factor is introduced into the fuzzy evaluation model to construct a "multi-period correlation evaluation matrix".

[0056] For the evaluation indexes of each period, calculate the membership degree to form a multi-period correlation evaluation matrix:

[0057] where R is the multi-period correlation evaluation matrix; n is the number of delay periods; m is the number of evaluation indexes, m = 16; r ij is the membership degree of index j at time t in the i-th fuzzy level.

[0058] 8. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, based on the time-varying contingency coefficient and the multi-period correlation evaluation matrix, a dynamic evaluation model of the water plugging effect is established to realize the time-varying prediction and evaluation of the water plugging effect of low-permeability reservoirs.

[0059] Comprehensive evaluation vector calculation:

[0060] where B is the comprehensive evaluation vector; W is the weight vector of the fuzzy evaluation model.

[0061] Dynamic evaluation index:

[0062] Time-varying sensitive index identification: If |w k,j - w k-1，j | > 0.2, then index j is a sensitive index. Calculate the dynamic evaluation index for water plugging well and layer selection according to the comprehensive evaluation vector B. Weight the elements in the comprehensive evaluation vector and sum them up to get a specific value. The larger the value, the more suitable the well and layer are for water plugging. Brief Description of the Drawings

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0064] Figure 1 It is a schematic flowchart of the evaluation method for water plugging well and layer selection based on considering the time-varying characteristics of seepage provided by the embodiment of the present invention. Detailed Embodiments

[0065] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following will describe the preferred embodiments of the present invention with reference to the accompanying drawings. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0066] S1. Construct an influencing factor index system: Collect the basic geological data, production dynamic data of each well in the target reservoir area, and water injection-related data such as injection water pressure and injection volume. Clean the collected data, remove outliers and incorrect data, and perform standardization processing to make different types of data comparable.

[0067] S2. Fuzzification processing: Define a corresponding set of fuzzy language variables for each index according to its characteristics, and use the trapezoidal membership function to determine the membership degree of each index data on different fuzzy language variables.

[0068] S3. Determine the initial weight: Use the time-varying game theory - dynamic CRITIC combined weighting method to generate the basic dynamic weight w base (t), whose variation law includes the seepage time decay factor and the sliding window analysis of dynamic CRITIC.

[0069] S4. Establish a seepage time-varying state space model: Based on the seepage mechanics theory, combined with the reservoir geological characteristics and production dynamic data, establish a seepage model considering time-varying characteristics.

[0070] S5. Kalman filter for real-time update of contingency coefficients: Use the Kalman filter algorithm to update the contingency coefficients in real time. By inputting reservoir dynamic monitoring data, track the time-varying law of reservoir seepage characteristics.

[0071] S6. Construction of multi-period correlation evaluation matrix: Introduce a time delay factor into the fuzzy evaluation model to construct a "multi-period correlation evaluation matrix" to quantify the dynamic correlation between water plugging effect and well and layer selection parameters at different time nodes.

[0072] S7. Application and result output of dynamic evaluation model: Based on the time-varying contingency coefficients and multi-period correlation evaluation matrix, establish a dynamic evaluation model for water plugging effect to realize the time-varying prediction and evaluation of water plugging effect in low-permeability reservoirs.

[0073] 2. A method for evaluating well and layer selection for water plugging considering time-varying seepage characteristics according to claim 1, characterized in that in the step of constructing the influencing factor index system, a multi-dimensional and multi-layer index system considering sixteen indexes of reservoir geomechanics and two-phase seepage is comprehensively considered. Specifically, it includes: formation pressure, formation permeability, Young's modulus, Poisson's ratio, brittleness index, in-situ stress difference coefficient, rock compressibility, crude oil viscosity, water saturation, production time, injection pressure, injection volume, fracture development degree, reservoir heterogeneity, breakthrough index, injection-production correlation index, etc.

[0074] In step S1, the water flooding degree index is determined by comparing the change range and growth rate of water saturation in the oil layer within different time periods; the calculation of the permeability variation coefficient considers the distribution characteristics of permeability data at different time points to reflect the time-varying situation of permeability heterogeneity.

[0075] In the formula, S w,j (t): Water saturation of the jth layer at time t, %.

[0076] In the formula, C v,j (t) is the permeability variation coefficient of the jth layer at time t, dimensionless; σ k,j (t) is the permeability standard deviation of the jth layer at time t, mD; μ k,j (t) is the permeability mean value of the jth layer at time t, mD.

[0077] The missing data is supplemented by the cubic spline interpolation method. The formula is:

[0078] In the formula, x i is the time of the known data point, a i , bi , c i , d i are interpolation coefficients.

[0079] 3. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the fuzzification step, a trapezoidal membership function is used to calculate the membership degree, and an initial fuzzy evaluation matrix is constructed:

[0080] When the index has a unimodal distribution, let b = c, and at this time the function degenerates into a triangular form, and the peak point is b.

[0081] 4. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that a time-varying game theory-dynamic CRITIC combined weighting method is used to generate the basic dynamic weight w base (t), and its variation law includes a seepage time decay factor and a sliding window analysis of dynamic CRITIC.

[0082] Based on the sliding time window to analyze the objective weight of the index:

[0083] Calculation of index conflict:

[0084] In the formula, C j is the conflict of the j-th index, dimensionless; σ j is the standard deviation of the j-th index within the sliding window, and the unit is the same as the original dimension of the index. For example, the unit of permeability is mD; r jk is the Pearson correlation coefficient between index j and index k within the sliding window, dimensionless; m is the total number of indexes.

[0085] Dynamic update: A sliding time window T is used, and the data within the window is updated at a time step of Δt (30 d), and the window width is determined according to the seepage response time (here it is taken as 3 months).

[0086] Seepage response factor: α j (t) = e -β·ΔP(t) / μ

[0087] In the formula, α j (t) is the seepage response factor of the j-th index at time t, dimensionless; β is the permeability attenuation coefficient, MPa-1; ΔP(t) is the change in the average pressure gradient within the time window, MPa; μ is the fluid viscosity, mPa·s.

[0088] Objective weight of dynamic CRITIC:

[0089] In the formula, is the dynamic CRITIC weight of the j-th index, dimensionless.

[0090] Introduce the seepage time decay factor to correct the subjective weight:

[0091] Seepage time decay factor:

[0092] In the formula, γ(t) is the time decay factor, dimensionless; γ0 is the initial decay coefficient, dimensionless; k d is the decay rate, d-1; t is the development time, d.

[0093] Correct the judgment matrix: A(t) = S(t) ⊙ γ(t) + (1 - γ(t)) · A0

[0094] In the formula, S(t) is the time-sensitive matrix; A(t) is the time-varying judgment matrix; A0 is the initial judgment matrix (static AHP result)

[0095] Solve the eigenvector corresponding to the largest eigenvalue of A(t) by the eigenvalue method, and normalize it to obtain the subjective weight

[0096] Solve the optimal combination of the subjective and objective weights and embed the time-varying adjustment mechanism. The game theory combination optimization objective function is:

[0097] In the formula, the first term is the difference between the subjective and objective weights; the second term is the weight change rate penalty term; λ(t) is the time-varying adjustment coefficient, d-1; ΔQ(t) is the cumulative liquid production rate change, m 3 / d; Q ref is the reference liquid production, m 3 .

[0098] Solve using the Lagrange multiplier method to obtain the Nash equilibrium weight

[0099] In the formula, δj(t) is the weight change trend term, calculated by the moving average of historical data.

[0100] Through the time-varying game theory - dynamic CRITIC model, couple the seepage time decay factor with the index conflict analysis to generate the initial weight that conforms to the dynamic law of the reservoir.

[0101] 5. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the step of analyzing the time-varying characteristics of seepage, when establishing the seepage model considering the time-varying characteristics, the compressibility of the fluid, the deformation characteristics of the rock, and the interaction of oil-water two-phase seepage are considered, and by introducing appropriate mathematical equations and parameters, the time-varying process of fluid seepage in the reservoir is more accurately described.

[0102] Based on Darcy's law and the material balance equation, a time-varying seepage model considering fluid compressibility and rock deformation is established, and the control equation is:

[0103] In the formula, k(t) is the permeability varying with time, mD; μ is the fluid viscosity, mPa·s; p(t) is the reservoir pressure at time t, MPa; φ is the porosity, %; c t is the comprehensive compressibility coefficient, MPa -1 .

[0104] The time-varying equation of seepage parameters is constructed by using the state space model. The seepage parameters such as permeability and pressure are used as state variables, and the state equation is established: x k = A k x k-1 + B k u k + w k

[0105] In the formula, x k is the state variable vector at the k-th moment; A k is the state transition matrix, reflecting the time autocorrelation of the state variables; B k is the input matrix; u k is the input vector; w k is the process noise vector. x k = [k k , p k , G k T

[0106] In the formula, k k is the permeability at the k-th moment, mD; p k is the formation pressure at the k-th moment, MPa; G k is the starting pressure gradient at the k-th moment, MPa·m -1 ; u k = [Q ik , p k T

[0107] In the formula, Qik is the water injection intensity, m 3 / d; p k is the production pressure difference, MPa.

[0108] Considering stress sensitivity, the time-varying equation of permeability is corrected, and the stress sensitivity coefficient α is introduced. The relationship between the corrected permeability and the effective stress is:

[0109] In the formula, k0 is the initial permeability, mD; α is the stress sensitivity coefficient, MPa -1 ; σ is the effective stress, MPa; σ0 is the initial effective stress, MPa.

[0110] 6. A water plugging well and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that, in the step of real-time updating the contingency coefficient by the Kalman filter, the initial weight determined in step S3 is used as the initial value of the contingency coefficient, and the Kalman filter algorithm is used to update the contingency coefficient in real time. By inputting the reservoir dynamic monitoring data, the prior estimate value of the contingency coefficient is predicted through the time update step, and then the posterior estimate value is corrected by using the monitoring data and the Kalman gain through the measurement update step, so as to realize the dynamic tracking of the stress sensitivity effect and the change of the starting pressure gradient of the reservoir.

[0111] Predict the prior estimate value of the contingency coefficient through the state equation

[0112] Introduce the seepage parameter data z of real-time monitoring k , and construct the measurement equation: z k =H k x k +v k

[0113] In the formula, H k is the measurement matrix, v k is the measurement noise vector.

[0114] The Kalman gain K k :

[0115] In the formula, is the prior estimate covariance matrix, R k is the measurement noise covariance matrix.

[0116] Use the Kalman gain to correct the prior estimate value to obtain the posterior estimate value of the contingency coefficient

[0117] By continuously iterating the above process, the contingency coefficient is updated in real time so that it can track the dynamic evolution of the reservoir seepage characteristics. This process ensures that the contingency coefficient can be adjusted in a timely manner according to the actual situation of the reservoir, laying a solid foundation for accurately constructing the multi-period correlation evaluation matrix and applying the dynamic evaluation model subsequently.

[0118] 7. A method for evaluating water plugging well and layer selection considering time-varying seepage characteristics according to claim 1, characterized in that, in the step of constructing the multi-period correlation evaluation matrix, a time delay factor is introduced into the fuzzy evaluation model to construct a "multi-period correlation evaluation matrix".

[0119] For the evaluation indexes of each period, calculate the membership degree to form a multi-period correlation evaluation matrix:

[0120] In the formula, R is the multi-period correlation evaluation matrix; n is the number of delay periods; m is the number of evaluation indexes, m = 16; r ij is the membership degree of index j at time t in the i-th fuzzy level.

[0121] 8. A method for evaluating water plugging well and layer selection considering time-varying seepage characteristics according to claim 1, characterized in that, based on the time-varying contingency coefficient and the multi-period correlation evaluation matrix, a dynamic evaluation model of water plugging effect is established to realize the time-varying prediction and evaluation of the water plugging effect of low-permeability reservoirs.

[0122] Calculation of comprehensive evaluation vector:

[0123] In the formula, B is the comprehensive evaluation vector; W is the weight vector of the fuzzy evaluation model.

[0124] Dynamic evaluation index:

[0125] Time-varying sensitive index identification: If |w k,j -w k-1,j |>0.2, then index j is a sensitive index. Calculate the dynamic evaluation index of water plugging well and layer selection according to the comprehensive evaluation vector B, and sum the elements in the comprehensive evaluation vector with weights to obtain a specific value. The larger the value, the more suitable the well and layer are for water plugging.

[0126] The above are only some preferred embodiments of the present invention. Any person skilled in the art may modify the above-described technical solutions or modify them into equivalent technical solutions. Therefore, the corresponding simple modifications or equivalent transformations made according to the technical solutions of the present invention all fall within the scope of protection required by the present invention.

Claims

1. A water plugging well selection and layer selection evaluation method considering the time-varying characteristics of seepage, characterized in that: The following steps are involved: S1. Construct an index system of influencing factors: Collect basic geological data, production dynamic data, and water injection pressure, injection volume and other water injection related data of each well in the target reservoir area. Clean the collected data, remove outliers and erroneous data, and perform standardization to make different types of data comparable. S2. Fuzzy processing: define the corresponding fuzzy linguistic variable set for each indicator according to its characteristics, and use the trapezoidal membership function to determine the membership degree of each indicator data on different fuzzy linguistic variables. S3. Initial weight determination: Use time-varying game theory-dynamic CRITIC combined weighting method to generate basic dynamic weight w base (t), and its variation pattern includes the seepage time attenuation factor and the sliding window analysis of dynamic CRITIC. S4. Establish a time-varying state space model of seepage: Based on the seepage mechanics theory, combined with reservoir geological characteristics and production dynamic data, establish a seepage model that takes time-varying characteristics into account. S5. Real-time update of contingency coefficients by Kalman filtering: The contingency coefficients are updated in real time by using the Kalman filtering algorithm, and the time-varying laws of reservoir seepage characteristics are tracked by inputting reservoir dynamic monitoring data. S6. Construction of multi-period correlation evaluation matrix: Introduce time delay factor into the fuzzy evaluation model to construct a "multi-period correlation evaluation matrix" to quantify the dynamic correlation between water plugging effect and well selection parameters at different time nodes; S7. Application of dynamic evaluation model and result output: Based on the time-varying weight coefficient and multi-period correlation evaluation matrix, a dynamic evaluation model of water plugging effect is established to achieve time-varying prediction and evaluation of water plugging effect in low permeability reservoirs.

2. A water shutoff well selection and layer selection evaluation method considering the time-varying seepage characteristics according to claim 1, characterized in that: In the step of constructing the index system of influencing factors, a multi-dimensional and multi-layer index system of sixteen indicators of reservoir geomechanics and two-phase seepage is comprehensively considered. In step S1, the flooding degree index is determined by comparing the change range and growth rate of water saturation in the oil layer in different time periods; the calculation of the permeability variation coefficient takes into account the distribution characteristics of permeability data at different time points to reflect the time-varying situation of permeability heterogeneity.

3. A water shutoff well selection and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that: In the fuzzification processing step, a trapezoidal membership function is used to calculate the membership degree and construct an initial fuzzy evaluation matrix.

4. A water shutoff well selection and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that: The basic dynamic weight w is generated by using the time-varying game theory-dynamic CRITIC combination weighting method base (t), and its variation pattern includes the seepage time attenuation factor and the sliding window analysis of dynamic CRITIC. The objective function of game theory combinatorial optimization is: In the formula, is the subjective weight of the jth indicator, dimensionless; is the dynamic weight of the jth indicator, dimensionless; λ(t) is the time-varying adjustment coefficient, d-1; ΔQ(t) is the cumulative liquid production change rate, m 3 / d;Q ref is the reference liquid production, m 3 . Dynamic CRITIC objective weights: a j (t)=e -β·ΔP(t) / μ In the formula, C j is the conflict of the jth indicator, dimensionless; σ j is the standard deviation of the jth indicator in the sliding window, and its unit is the same as the original dimension of the indicator; r jk is the Pearson correlation coefficient between indicator j and indicator k in the sliding window, dimensionless; m is the total number of indicators; α j (t) is the seepage response factor of the jth index at time t, dimensionless; β is the permeability attenuation coefficient, MPa -1 ; ΔP(t) is the average pressure gradient change in the time window, MPa; μ is the fluid viscosity, mPa·s. Subjective weight: A(t)=S(t)☉γ(t)+(1-γ(t))·A0 Where S(t) is the time-sensitive matrix; A(t) is the time-varying judgment matrix; A0 is the initial judgment matrix (static AHP result); γ(t) is the time attenuation factor, dimensionless; γ0 is the initial attenuation coefficient, dimensionless; k d is the decay rate, d-1; t is the development time, d. Solve the eigenvector corresponding to the maximum eigenvalue of A(t) and normalize it to get the subjective weight Through the time-varying game theory-dynamic CRITIC model, the seepage time attenuation factor and the index conflict analysis are coupled to generate the initial weights that conform to the dynamic laws of the reservoir.

5. A water shutoff well selection and layer selection evaluation method considering the time-varying seepage characteristics according to claim 1, characterized in that: In the step of analyzing the time-varying characteristics of seepage, when establishing the seepage model taking into account the time-varying characteristics, the compressibility of the fluid, the deformation characteristics of the rock and the interaction between the oil-water two-phase seepage are considered, and by introducing appropriate mathematical equations and parameters, the time-varying process of fluid seepage in the reservoir is described more accurately. Based on Darcy's law and material balance equation, a time-varying seepage model considering fluid compressibility and rock deformation is established, and the control equation is: Where k(t) is the permeability that changes with time, mD; μ is the fluid viscosity, mPa·s; p(t) is the reservoir pressure at time t, MPa; φ is the porosity, %; c t is the comprehensive compression coefficient, MPa -1 . The state space model is used to construct the time-varying equation of seepage parameters. The seepage parameters such as permeability and pressure are used as state variables to establish the state equation: x k =A k x k-1 +B k u k +w k x k =[k k ,p k ,G k ] T u k =[Q ik ,p k ] T In the formula, x k is the state variable vector at the kth moment; A k is the state transfer matrix, reflecting the time autocorrelation of the state variables; B k is the input matrix; u k is the input vector; w k is the process noise vector; k k is the permeability at the kth moment, mD; p k is the formation pressure at the kth moment, MPa; G k is the starting pressure gradient at the kth moment, MPa·m -1 ;Q ik is the water injection intensity, m 3 / d;p k is the production pressure difference, MPa.

6. A water shutoff well selection and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that: In the step of real-time updating of the contingency coefficient by Kalman filtering, the initial weight determined in step S3 is used as the initial value of the contingency coefficient, and the contingency coefficient is updated in real time using the Kalman filtering algorithm. By inputting dynamic monitoring data such as reservoir pressure, permeability, and water content, the prior estimate of the contingency coefficient is predicted through the time updating step, and then the a posteriori estimate is obtained by using the monitoring data and the Kalman gain correction through the measurement updating step, thereby realizing dynamic tracking of the reservoir stress sensitivity effect and the start-up pressure gradient change. Predicting the prior estimates of the contingency coefficients through the equation of state Introducing real-time monitoring of seepage parameter data k , construct the measurement equation: z k =H k x k +v k In the formula, H k is the measurement matrix, v k is the measurement noise vector. Kalman gain K k : In the formula, is the prior estimated covariance matrix, R k is the measurement noise covariance matrix. Use the Kalman gain to correct the prior estimate and obtain the posterior estimate of the weight coefficient By continuously iterating the above process, the contingency coefficient is updated in real time, making it possible to track the dynamic evolution of reservoir seepage characteristics.

7. A water shutoff well selection and layer selection evaluation method considering the time-varying seepage characteristics according to claim 1, characterized in that: In the step of constructing the multi-period correlation evaluation matrix, a time delay factor is introduced into the fuzzy evaluation model to construct a "multi-period correlation evaluation matrix". Where R is the multi-period correlation evaluation matrix; n is the number of time periods; m is the number of evaluation indicators, m = 16; r ij is the membership degree of index j at the i-th fuzzy level at time t.

8. A water plugging well selection and layer selection evaluation method considering the time-varying characteristics of seepage according to claim 1, characterized in that: Based on the time-varying weight coefficient and the multi-period correlation evaluation matrix, a dynamic evaluation model for water plugging effect is established to achieve time-varying prediction and evaluation of water plugging effect in low-permeability oil reservoirs. Comprehensive evaluation vector calculation: Where B is the comprehensive evaluation vector; W is the weight vector of the fuzzy evaluation model. Dynamic evaluation index: In the formula, w k,j is the time-varying weight, r ij It is the element of multi-period correlation evaluation matrix. Identification of time-varying sensitive indicators: If |w k，j -w k-1，j |>0.2, then index j is a sensitive index. The dynamic evaluation index for water plugging well and layer selection is calculated based on the comprehensive evaluation vector B, and the weighted sum of the elements in the comprehensive evaluation vector is used to obtain a specific value. The larger the value, the more suitable the well and layer are for water plugging.

Images