Streamline prediction method based on flow field parameters between injection and production wells

Abstract

The invention relates to the technical field of injection and production well flow fields, in particular to a streamline prediction method based on flow field parameters between injection and production wells, which comprises the following steps: constructing an injection and production well coordinate system, and acquiring a plurality of flow field feature points between two end points by taking the coordinates of a water injection well and a production well as the two end points; performing polynomial fitting by using the flow field feature points to obtain a flow field curve; selecting a plurality of flow field parameter sampling points on the plurality of flow field curves, and calculating a flow field parameter mean value; constructing a particle swarm algorithm, and optimizing a correction coefficient of the flow field parameter mean value by taking the flow field parameter mean value as an input parameter and the water production rate as a target function; and calculating the corrected flow field parameter according to the correction coefficient of the flow field parameter. The method solves the problems that a large amount of oil reservoir data is needed for obtaining streamline field data of a target oil reservoir through an existing numerical simulation method, and the influence of permeability on a streamline is not considered.

Description

Technical Field

[0001] The invention relates to the technical field of injection-production well flow fields, and in particular to a streamline prediction method based on flow field parameters between injection-production wells. Background Art

[0002] In actual oil reservoirs, due to the horizontal and vertical reservoir heterogeneity, the relationship between injection and production wells is complex, and the differences in intra-layer, inter-layer, plane water content level, and production degree are getting bigger and bigger. Therefore, it is particularly important to determine the parameter characterization between injection and production wells and clarify the injection and production relationship from a quantitative perspective. The patent name is a streamline field characterization method and device for oil reservoirs.

[0003] The patent with publication number CN112302638A discloses a streamline field characterization method and device for an oil reservoir. The streamline field data of a target oil reservoir is obtained through a numerical simulation model. This method requires a lot of reservoir data and does not consider the effect of permeability on streamlines. Summary of the invention

[0004] In view of the shortcomings of the existing methods, the present invention solves the problem that the existing numerical simulation methods need a large amount of reservoir data to obtain the streamline field data of the target reservoir and do not consider the influence of permeability on streamlines.

[0005] The technical solution adopted by the present invention is: a streamline prediction method based on flow field parameters between injection and production wells comprises the following steps:

[0006] Step 1: Construct the injection and production well coordinate system, take the coordinates of the injection well and the production well as the two end points, obtain several flow field characteristic points between the two end points; use the flow field characteristic points to perform polynomial fitting to obtain the flow field curve;

[0007] As a preferred embodiment of the present invention, the acquisition of several flow field characteristic points includes:

[0008] The injection wells and production wells are equally spaced into several odd-numbered points, and a straight line perpendicular to the x-axis is set with the odd-numbered point as the center; a ray is set with the injection well as the starting point and the angle θ with the x-axis, and a ray is set with the injection well as the starting point and the angle 180-θ with the x-axis. The intersection of the two rays and the straight line at the odd-numbered point is the flow field characteristic point.

[0009] As a preferred embodiment of the present invention, the polynomial fitting formula is:

[0010] y=a0x n-1 +a1x n-2 +...a n

[0011] Among them, a n is the fitting coefficient, x is the independent variable, and n is the number of odd points.

[0012] As a preferred implementation manner of the present invention, the number of odd points=3.

[0013] Step 2: Select a number of flow field parameter sampling points on a number of flow field curves and calculate the mean value of the flow field parameters;

[0014] As a preferred embodiment of the present invention, the flow field parameters include: permeability, porosity and water saturation.

[0015] Step 3: Construct a particle swarm algorithm, take the mean value of flow field parameters as input parameters, and the water production rate as the objective function, and optimize the correction coefficient of the mean value of flow field parameters.

[0016] As a preferred embodiment of the present invention, the particle swarm algorithm includes:

[0017] Focusing on production wells and water production rate The water production rate in the water injection stage and the oil production stage is solved by using the average flow field parameters and Darcy's law. Assume that the correction coefficient C of the mean value of flow field parameters is par ij , par is the flow field parameter; find the objective function The target function is selected as the loop termination indicator during iteration, and the Is it satisfied? If yes, terminate, otherwise continue to iterate; where β is the target threshold, It is the water production rate solved by the particle swarm algorithm after correction.

[0018] Step 4, calculating the corrected flow field parameters according to the correction coefficients of the flow field parameters;

[0019] As a preferred embodiment of the present invention, the formula of the corrected flow field parameter is:

[0020]

[0021] in, is the mean value of flow field parameters.

[0022] As a preferred embodiment of the present invention, a streamline prediction system based on flow field parameters between injection and production wells includes: a memory for storing instructions executable by a processor; and a processor for executing instructions to implement a streamline prediction method based on flow field parameters between injection and production wells.

[0023] As a preferred embodiment of the present invention, a computer readable medium stores computer program codes, and when the computer program codes are executed by a processor, a streamline prediction method based on flow field parameters between injection and production wells is implemented.

[0024] Beneficial effects of the present invention:

[0025] 1. The present invention accurately finds the flow field curve between the production well and the water injection well by nonlinear fitting; it is simpler than the existing numerical simulation model;

[0026] 2. Taking the flow field curve as the benchmark, the permeability, porosity and water saturation of the flow field characteristic points as the training parameters, and using the particle swarm algorithm to correct the flow field parameters between injection and production wells, it can conveniently and quickly meet the needs of oilfield on-site engineering and technical personnel to understand the relationship between reservoir injection and production parameters in a short time, effectively improve the implementation efficiency of on-site measures, improve economic benefits and implementation effects, and provide a basis for subsequent analysis of well differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of the streamline prediction method based on flow field parameters between injection and production wells of the present invention;

[0028] Figure 2 It is a schematic diagram of flow field characteristic points of a generating well and a water injection well of the present invention;

[0029] Figure 3 It is a schematic diagram of a flow field curve of the present invention. DETAILED DESCRIPTION

[0030] The present invention is further described below in conjunction with the accompanying drawings and embodiments. This figure is a simplified schematic diagram, which only illustrates the basic structure of the present invention in a schematic manner, and therefore it only shows the components related to the present invention.

[0031] like Figure 1 As shown, a streamline prediction method based on flow field parameters between injection and production wells includes the following steps:

[0032] Step 1: Construct the injection and production well coordinate system, take the coordinates of the injection well and the production well as the two end points, obtain several flow field characteristic points between the two end points; use the flow field characteristic points to perform polynomial fitting to obtain the flow field curve;

[0033] Obtaining several flow field feature points between the two end points includes:

[0034] The injection wells and production wells are equally spaced into a number of odd points n, and a straight line perpendicular to the x-axis is drawn with the odd point as the center; a ray is drawn with the injection well as the starting point and the angle θ with the x-axis, and a ray is drawn with the injection well as the starting point and the angle 180-θ with the x-axis. The intersection of the ray and the straight line at the odd point is the flow field characteristic point.

[0035] The injection and production well coordinate system starts from the injection well and ends at the production well in the x-axis direction, with the y-axis perpendicular to the x-axis.

[0036] like Figure 2 As shown in the figure, the blue injection well O in the well network is used as the origin of the injection and production well coordinate system. The coordinates of point O (x i ,yi ) as the origin O(0,0), the red production well P j The coordinates of (L O pj,0),L O pj is the injection well O and the production well P j The distance Figure 2 For example, when the number of odd points is n=3, when θ is a certain angle, the coordinates of the three flow field feature points are (L Opj / 4,y1)、(L Opj / 2,y2)、(3L Opj / 4,y3);y1=L Opj / 4*tanθ,y2=L Opj / 2*tanθ,y2=3L Opj / 4*tanθ; -90°<θ<90°;

[0037] The formula for polynomial fitting is: y = a0x n-1 +a1x n-2 +...a n ; Since the more odd points there are, the more complex the polynomial fitting becomes, through actual testing, when n=3, both the computational complexity and accuracy requirements can be met.

[0038] like Figure 3 As shown, there are six simulated flow field curves when θ is ±30°, ±45°, and ±60° respectively;

[0039] For example, when θ = 60°, the curve fitting is

[0040] Step 2: Select a number of flow field parameter sampling points on a number of flow field curves and calculate the mean value of the flow field parameters;

[0041] Flow field parameters include: permeability K, porosity φ and water saturation S w ;

[0042] Taking the sampling point as the characteristic point of the flow field as an example, the mean permeability is k0, k1 is the permeability of the injection well, 3 flow field characteristic points, and production well; similarly, the mean porosity can be obtained Mean water saturation

[0043] Step 3: Construct a particle swarm algorithm, take the mean of flow field parameters as input parameters, and the water production rate as the objective function, and optimize the correction coefficient of flow field parameters;

[0044] In the particle swarm algorithm, the production well is the center and the water production rate is the is the fitting index; the water production rate in the water injection stage and the oil production stage is solved by Darcy's law using the average flow field parameters. Assume that the correction coefficient C of the mean value of flow field parameters is par ij , par is the flow field parameter; find the objective function The target function is selected as the loop termination indicator during iteration, and the Is it satisfied? If yes, terminate, otherwise continue iteration; β is the target threshold, f calculate is the water production rate solved by the particle swarm algorithm after correction, and in this embodiment, β=1%.

[0045] Specifically include:

[0046] Suppose in a two-dimensional target search space, there are n' particles forming a population X = (X1, X2, ..., X n’ ), where the i-th particle is a two-dimensional vector, representing the position of the i-th particle in the two-dimensional search space and also represents a potential solution to the problem;

[0047] For particle i, the vector X i Represents the correction coefficient of flow field parameters; vector V i represents the acceleration factor of the flow field parameters;

[0048] Particle Speed:

[0049] Particle Position:

[0050] Among them, ω is the inertia weight. A large value can make the algorithm have a stronger global search ability, and a small value makes the algorithm tend to local search; ω is initially set to 0.9 and linearly decreases to 0.4 with the increase of the number of iterations. In this way, we can focus on the global search first, so that the search space converges quickly to a certain area, and then use local fine search to obtain a high-precision solution;

[0051] c1 and c2 are two learning factors, which are non-negative constants, usually 2, but there are other values. Usually c1 = c2 and the range is between 0 and 4.

[0052] r1 and r2 are random numbers distributed in the interval [0,1];

[0053] In order to prevent blind searching of particles, the particle position and velocity are generally limited to a certain range. If the current acceleration of the particle causes the particle velocity V in a certain dimension to i Exceed the maximum speed V of this dimension dmax , then the speed of this dimension is limited to the maximum speed upper limit V of this dimension dmax , V dmax The choice should not exceed the particle width range, if Vdmax If it is too large, the particle may fly over the location of the optimal solution; if it is too small, it may reduce the global search ability of the particle.

[0054] The number of particles in the particle swarm is usually 20-40. For most problems, 10 particles are enough to achieve good results. However, for more difficult problems or problems of a specific category, the number of particles can be 100 or 200. Generally speaking, the larger the swarm size, the greater the possibility of searching for the global optimal solution, and the better the optimization performance. However, the algorithm consumes more computing power, and the computing performance decreases relatively. However, the swarm size cannot be too small. If the swarm size is too small, the intelligence of the swarm intelligence optimization algorithm cannot be reflected, resulting in serious damage to the algorithm performance.

[0055] Termination condition: maximum number of cycles and minimum error requirement, this termination condition is determined by the specific problem. For the maximum allowed number of iterations, a larger number of iterations enables the algorithm to better search the solution space, so the possibility of finding the global optimal solution is also greater; correspondingly, a smaller maximum allowed number of iterations will reduce the possibility of the algorithm finding the global optimal solution.

[0056] Step 4, calculating the corrected flow field parameters according to the correction coefficients of the flow field parameters;

[0057] The formula of the corrected flow field parameters is:

[0058]

[0059] in, is the mean value of flow field parameters.

[0060] Taking the corrected permeability as an example, Get the corrected permeability of the five points

[0061] Similarly, the corrected porosity and water saturation can be calculated. Specific embodiment:

[0063] The injection-production relationship between the injection wells and the corresponding production wells in the four sub-layers of the XX reservoir was quantitatively analyzed. The results are shown in the following table. The corrected pore permeability is shown in Table 1.

[0064] Table 1 Comparison of corrected permeability and actual permeability of injection wells and production wells

[0065]

[0066] It can be seen from Table 1 that the difference between the corrected permeability and the actual permeability is very small, indicating that the method of the present invention is effective.

[0067] Based on the above ideal embodiments of the present invention, the relevant staff can make various changes and modifications without departing from the technical concept of the present invention through the above description. The technical scope of the present invention is not limited to the contents of the specification, and its technical scope must be determined according to the scope of the claims.

Claims

1. A streamline prediction method based on flow field parameters between injection and production wells, characterized in that: The following steps are involved: Step 1: Construct the injection and production well coordinate system, take the coordinates of the injection well and the production well as the two end points, and obtain several flow field characteristic points between the two end points; The flow field curve is obtained by polynomial fitting using the flow field characteristic points; Step 2: Select a number of flow field parameter sampling points on a number of flow field curves and calculate the mean value of the flow field parameters; Step 3: Construct a particle swarm algorithm, take the mean value of flow field parameters as input parameter, water production rate as target function, and optimize the correction coefficient of the mean value of flow field parameters; Step 4: Calculate the corrected flow field parameters according to the correction coefficients of the flow field parameters.

2. The streamline prediction method based on flow field parameters between injection and production wells according to claim 1 is characterized in that: The acquisition of several flow field feature points includes: The injection wells and production wells are equally spaced into several odd-numbered points, and a straight line perpendicular to the x-axis is set with the odd-numbered point as the center; a ray is set with the injection well as the starting point and the angle θ with the x-axis, and a ray is set with the injection well as the starting point and the angle 180-θ with the x-axis. The intersection of the two rays and the straight line at the odd-numbered point is the flow field characteristic point.

3. The streamline prediction method based on flow field parameters between injection and production wells according to claim 1 is characterized in that: The formula for polynomial fitting is: y=a0x n-1 +a1x n-2 +...a n Among them, a n is the fitting coefficient, x is the independent variable, and n is the number of odd points.

4. The streamline prediction method based on flow field parameters between injection and production wells according to claim 1, characterized in that: Flow field parameters include: permeability, porosity and water saturation.

5. The streamline prediction method based on flow field parameters between injection and production wells according to claim 1, characterized in that: Particle swarm optimization includes: Focusing on production wells and water production rate is the fitting index; let the correction coefficient C of the mean value of flow field parameters be par ij , par is the flow field parameter; find the objective function is the water production rate; determine the objective function Is it less than the target threshold? If yes, terminate, otherwise continue to iterate; where f calculate It is the water production rate solved by the particle swarm algorithm after correction.

6. The streamline prediction method based on flow field parameters between injection and production wells according to claim 1, characterized in that: The formula of the corrected flow field parameters is: in, is the mean value of flow field parameters, C par ij is the correction coefficient of the mean value of flow field parameters.

7. The streamline prediction method based on flow field parameters between injection and production wells according to claim 1, characterized in that: Number of odd points = 3.

8. A streamline prediction system based on flow field parameters between injection and production wells, characterized in that: include: a memory for storing instructions executable by a processor; A processor, used to execute instructions to implement the streamline prediction method based on flow field parameters between injection and production wells as described in any one of claims 1-7.

9. A computer readable medium storing a computer program code, characterized in that: When the computer program code is executed by a processor, the streamline prediction method based on flow field parameters between injection and production wells as described in any one of claims 1 to 7 is implemented.

Images