A method for collaboratively constructing random fields of permeability coefficient and porosity in heterogeneous aquifers

By introducing correlated random variables into heterogeneous aquifers, a stochastic field synergistic model of permeability coefficient and porosity is constructed, which solves the problem that the influence of parameter uncertainty is difficult to reflect in traditional methods, realizes the accurate spatial distribution of permeability coefficient and porosity, and improves the reliability and applicability of the model.

CN116029136BActive Publication Date: 2024-11-29CHINA UNIV OF MINING & TECH
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202310037190.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-01-10
Publication Date
2024-11-29
Estimated Expiration
2043-01-10

AI Technical Summary

Technical Problem

Existing technologies struggle to effectively reflect the impact of parameter uncertainties on numerical simulation results when constructing spatial distribution fields of permeability and porosity in heterogeneous aquifers. Furthermore, traditional methods are highly susceptible to human factors, which may lead to simulation failure.

Method used

By introducing associated random variables, a random field collaborative model of permeability coefficient and porosity is constructed to achieve spatial relationship control of positive correlation, negative correlation and no correlation. Exploration borehole layout methods such as diagonal layout method and quincunx layout method are adopted, and the relevant distance is calculated by combining recursive spatial method and variogram method to generate a normal distribution random field of permeability coefficient and porosity.

Benefits of technology

It achieves an effective spatial correlation between permeability coefficient and porosity, and can construct different types of correlations according to research needs, thereby improving the accuracy and reliability of the model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116029136B_ABST
    Figure CN116029136B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of environmental protection and water resources technology, specifically a method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer, the steps of which are as follows: Step 1: construct a two-dimensional rectangular coordinate system of a study area, collect permeability coefficient and porosity data of exploration holes in the aquifer in the study area, calculate the mean and variance of the permeability coefficient and porosity respectively, and determine the correlation distance of the permeability coefficient and porosity of the site; Step 2: grid the study area, calculate and generate a random field of logarithmic normal distribution of permeability coefficient; Step 3: generate a random field of porosity normal distribution by associating random variables. The present invention constructs a random field of porosity normal distribution based on the exploration data, and realizes the spatial variability and intrinsic correlation of permeability coefficient and porosity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of environmental protection and water resources, and in particular to a method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer. Background Art

[0002] With the development and widespread application of modern applied mathematics and computer technology, numerical simulation has gradually become the main means to study the laws of groundwater movement, quantitatively evaluate groundwater resources, and simulate the occurrence and development of some hydrogeological processes. In a large study area, hydrogeological parameters often have spatial variability and show heterogeneity, so it is often necessary to construct a spatial distribution field of hydrogeological parameters. On the other hand, constructing a single spatial distribution field of a certain parameter often has certain risks and cannot fully reflect the impact of parameter uncertainty on numerical simulation results. For this reason, it is necessary to construct a spatial distribution random field of hydrogeological parameters.

[0003] For heterogeneous strata, the traditional method of constructing a hydrogeological conceptual model is to artificially divide the study area into several regions based on the similarity of lithology, and then assign specific hydrogeological parameters to each region. However, the accuracy of the conceptual model constructed by this method is closely related to factors such as the researcher's level of understanding of the study area and the hydrogeological exploration situation. It is greatly affected by human factors and may suffer from "simulation failure".

[0004] At present, a lot of research has been done on the spatial variability of permeability parameters and porosity. It is generally believed that the permeability coefficient and porosity of aquifers are the main factors affecting groundwater flow and solute transport, and the permeability coefficient follows a log-normal distribution, while the porosity follows a normal distribution. For soils of a specific lithology, there may be three relationships between permeability coefficient and porosity: positive correlation, negative correlation, and no correlation. Therefore, how to construct a coordinated random field of permeability coefficient and porosity under the premise of ensuring the spatial distribution of the two parameters is of great significance for studying the impact of lithology heterogeneity on groundwater flow and solute transport. Summary of the invention

[0005] The purpose of the present invention is to provide a method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer. By introducing associated random variables, the random field of permeability coefficient and the random field of porosity are spatially associated, and positive association, negative association and unrelated artificial regulation are achieved.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer, comprising:

[0007] Step 1: Construct a two-dimensional rectangular coordinate system for the study area, collect the permeability and porosity data of the exploration holes in the aquifer in the study area, calculate the mean and variance of the permeability and porosity, and determine the correlation distance of the site permeability and porosity;

[0008] Step 2: Grid the study area and calculate the log-normal distribution random field of permeability coefficient;

[0009] Step 3: Generate a porosity normal distribution random field by correlating random variables.

[0010] Preferably, the exploration holes in step 1 should be arranged according to the characteristics of the area, shape, purpose, and pollution status of the study area, using the diagonal point distribution method, plum blossom point distribution method, chessboard point distribution method, or "S" shaped point distribution method. The relevant distance can be calculated using the recursive space method, the correlation function method, and the variation function method.

[0011] Preferably, in step 2, grid division should be performed according to the variability of the exploration data, the length of the cell should be determined, and the difference of the parameter values ​​in each cell should be ensured to be small. The random field formula of the logarithmic normal distribution of the permeability coefficient is:

[0012]

[0013] In the formula, K represents the site permeability coefficient, which is a random variable; E(K) represents the mean of the site permeability coefficient; cv represents the coefficient of variation, D(K) represents the variance of the site permeability coefficient; Lε1 represents a random variable that obeys the m-dimensional normal distribution N(0,C), where m is the number of grids in the site, C=(C ij ) m×m represents its covariance matrix, whose elements represents the spatial correlation between grid i and grid j, Δx ij with Δy ij represents the difference between the coordinates of grid i and grid j in the X and Y directions, L x and L y Represent the relevant distances of the site permeability in the X direction and the Y direction respectively.

[0014] Preferably, the derivation process of the associated random variable in step 3 is as follows:

[0015] Taking the logarithm of both sides of formula (1) we get:

[0016]

[0017] According to the properties of lognormal distribution, lnK obeys normal distribution, and the mean of lnK is Variance D(ln K)=ln(cv 2+1), so the above formula can be transformed as follows:

[0018]

[0019] Since the porosity obeys the normal distribution, referring to the normal distribution expression of formula (3), this paper proposes a porosity random field calculation formula as follows:

[0020]

[0021] Among them, n represents the site porosity, which is a random variable; E(n) represents the mean of the site porosity; D(n) represents the variance of the site porosity; Lε2 still represents a random variable that obeys the m-dimensional normal distribution N(0,C);

[0022] Since the permeability coefficient and porosity of the same research site are closely related, it can be seen that the random field of permeability coefficient and the random field of porosity are not independent, that is, Lε1 and Lε2 are not independent of each other. From formula (2), we can get Obey the normal distribution N(0,C). At this time, introduce the random variable η and the constant ρ(ρ∈[-1,1]), assuming that η also obeys the normal distribution N(0,C), and Lε1 and η are independent of each other, then It also obeys the normal distribution N(0,C), so we can make (4) Thus, the calculation formula of the porosity random field associated with the permeability random field is obtained:

[0023]

[0024] The porosity random field given by formula (5) is related to the permeability random field and also includes the randomness of the porosity itself;

[0025] Use n i represents the porosity of grid i, K i represents the permeability coefficient of grid i (i = 1, 2, L, m), and n is calculated according to formulas (3) and (5): i With lnK i The covariance of get It can be seen that ρ represents the correlation coefficient between porosity n and the logarithm of permeability lnK, and ρ∈[-1,1]; 0<ρ≤1 indicates positive correlation, -1≤ρ<0 indicates negative correlation, and ρ=0 indicates no correlation. Since permeability and porosity are associated through the spatial random variable Lε1, Lε1 is called the associated random variable;

[0026] The range of n is specified as [E(n)-δ, E(n)+δ], where 0<δ<E(n). In order to ensure that most sample points fall within the range of values ​​when randomly sampling porosity, the standard deviation can be limited according to the properties of normal distribution. Although restrictions However, there will still be a small number of points falling outside [E(n)-δ, E(n)+δ]. Therefore, it is stipulated that when n<E(n)-δ, the porosity is assigned to E(n)-δ; when n>E(n)+δ, the porosity is assigned to E(n)+δ.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention provides a method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer. By introducing associated random variables, the proposed associated random variables can effectively correlate permeability coefficient and porosity in space, and construct positively correlated, negatively correlated and uncorrelated spatial relationships according to research needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 Schematic diagram of the calculation model of the present invention.

[0030] Figure 2 It is a schematic diagram of the positively correlated permeability random field and porosity random field of the present invention.

[0031] Figure 3 It is a schematic diagram of the negatively correlated permeability random field and porosity random field of the present invention.

[0032] Figure 4 It is a schematic diagram of the random field of permeability coefficient and the random field of porosity which are unrelated in the present invention.

[0033] Figure 5 It is a schematic diagram of the positive correlation between the permeability coefficient and the porosity scatter point of the present invention.

[0034] Figure 6 It is a schematic diagram of the negative correlation between the permeability coefficient and the porosity scatter point of the present invention.

[0035] Figure 7 It is a schematic diagram showing that the permeability coefficient and porosity scatter points are uncorrelated in the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments.

[0037] The present invention proposes a method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer, comprising:

[0038] Step 1: Construct a two-dimensional rectangular coordinate system for the study area, collect the permeability and porosity data of the exploration holes in the aquifer in the study area, calculate the mean and variance of the permeability and porosity, and determine the correlation distance of the site permeability and porosity;

[0039] Step 2: Grid the study area and calculate the log-normal distribution random field of permeability coefficient;

[0040] Step 3: Generate a porosity normal distribution random field by correlating random variables.

[0041] The exploration holes in step 1 of the present invention should be arranged according to the characteristics of the area, shape, purpose, and pollution status of the study area, using the diagonal point arrangement method, the plum blossom point arrangement method, the chessboard point arrangement method, or the "S" shape point arrangement method. The relevant distance can be calculated using the recursive space method, the correlation function method, and the variation function method.

[0042] In step 2, the grid should be divided according to the variability of the exploration data, the length of the cell should be determined, and the difference of the parameter values ​​in each cell should be ensured to be small. The random field formula of the logarithmic normal distribution of the permeability coefficient is:

[0043]

[0044] In the formula, K represents the site permeability coefficient, which is a random variable; E(K) represents the mean of the site permeability coefficient; cv represents the coefficient of variation, D(K) represents the variance of the site permeability coefficient; Lε1 represents a random variable that obeys the m-dimensional normal distribution N(0,C), where m is the number of grids in the site, C=(C ij ) m×m represents its covariance matrix, whose elements represents the spatial correlation between grid i and grid j, Δx ij with Δy ij represents the difference between the coordinates of grid i and grid j in the X and Y directions, L x and L y Represent the relevant distances of the site permeability in the X direction and the Y direction respectively.

[0045] The derivation process of the associated random variable in step 3 is as follows:

[0046] Taking the logarithm of both sides of formula (1) we get:

[0047]

[0048] According to the properties of lognormal distribution, lnK obeys normal distribution, and the mean of lnK is Variance D(lnK)=ln(cv 2 +1), so the above formula can be transformed as follows:

[0049]

[0050] Since the porosity obeys the normal distribution, referring to the normal distribution expression of formula (3), a porosity random field calculation formula is as follows:

[0051]

[0052] Among them, n represents the site porosity, which is a random variable; E(n) represents the mean of the site porosity; D(n) represents the variance of the site porosity; Lε2 still represents a random variable that obeys the m-dimensional normal distribution N(0,C).

[0053] Since the permeability coefficient and porosity of the same research site are closely related, it can be seen that the random field of permeability coefficient and the random field of porosity are not independent, that is, Lε1 and Lε2 are not independent of each other. From formula (2), we can get It obeys the normal distribution N(0,C). Now we introduce the random variable η and the constant ρ(ρ∈[-1,1]). Assume that η also obeys the normal distribution N(0,C), and Lε1 and η are independent of each other. Then It also obeys the normal distribution N(0,C). We can make (4) Thus, the calculation formula of the porosity random field associated with the permeability random field is obtained:

[0054]

[0055] The porosity random field given by formula (5) is related to the permeability random field and also includes the randomness of the porosity itself.

[0056] Use n i represents the porosity of grid i, K i represents the permeability coefficient of grid i (i = 1, 2, L, m), and n is calculated according to formulas (3) and (5): i and ln K i The covariance of get It can be seen that ρ represents the correlation coefficient between porosity n and the logarithm of permeability lnK, and ρ∈[-1,1]; 0<ρ≤1 indicates positive correlation, -1≤ρ<0 indicates negative correlation, and ρ=0 indicates no correlation. Since permeability and porosity are associated through the spatial random variable Lε1, Lε1 is called the associated random variable.

[0057] Take porosity as an example to construct a normal distribution random field. Assuming that the site porosity n is a random variable, obeys the normal distribution, with a mean of E(n) and a variance of D(n), then theoretically, the value range of porosity n is (-∞,+∞), which is obviously inconsistent with the actual situation.

[0058] If n is not given a range, in order to avoid the random value of porosity n resulting in n≤0 or n≥1, it is stipulated that when n≤0, a very small positive number is used instead (such as 0.001), and when n≥1, a positive number close to and less than 1 is used instead (such as 0.999).

[0059] If the range of n is specified as [E(n)-δ, E(n)+δ], where 0<δ<E(n). In order to ensure that most sample points fall within the range of values ​​when randomly sampling porosity, the standard deviation can be limited according to the properties of normal distribution. Although restrictions However, there will still be a small number of points falling outside [E(n)-δ, E(n)+δ]. Therefore, it is stipulated that when n<E(n)-δ, the porosity is assigned to E(n)-δ; when n>E(n)+δ, the porosity is assigned to E(n)+δ.

[0060] Specific

[0061] Assume that the study area is a rectangular area with a length of 510m and a width of 300m. The target aquifer is a sandy confined aquifer with a thickness of 20m. The western boundary and the eastern boundary are given head boundaries, of which the eastern head is 25m and the western head is 30m. The northern boundary and the southern boundary are impermeable boundaries. Under natural conditions, the groundwater flow is a two-dimensional homogeneous, anisotropic, confined and stable flow flowing from west to east.

[0062] A coordinate system was established with the southwest corner of the study area as the origin.

[0063] Table 1 Probability distribution and value characteristics of geological parameters in the study area

[0064]

[0065] Note: NA indicates non-required parameters

[0066] Since the research area is divided into four areas, A, B, C, and D, the mean permeability coefficients are E(K A )=1m / d,E(K B )=23m / d,E(K C )=15m / d,E(K D )=10m / d; the mean porosity values ​​are E(n A )=0.27,E(n B )=0.32,E(n C )=0.22,E(n D )=0.18.Set L xA =L yA =L xB =L yB =L xC =LyC =L xD =L yD = 20 m, the porosity variance of the four zones is set to The number of grids for the site is n = 17000, and the cell length is 3 m. ρ = 0.8, ρ = -0.8, and ρ = 0 represent positive correlation, negative correlation, and no correlation between permeability and porosity, respectively. According to formulas (1) and (5), the Monte Carlo method is used to generate 400 sets of permeability random fields and porosity random fields that meet the given distribution characteristics, and one set of permeability random fields and porosity random fields are listed respectively.

[0067] The constructed permeability random field and porosity random field were statistically analyzed respectively. The permeability and porosity of the four areas A, B, C and D in the study area showed positive correlation, negative correlation and uncorrelated relationship, indicating that the proposed method of collaboratively constructing the permeability log-normal distribution random field and the porosity normal distribution random field based on associated random variables is feasible and reliable.

[0068] The above-mentioned specific embodiments are only several preferred embodiments of the present invention. Based on the technical solutions of the present invention and the relevant inspirations of the above-mentioned embodiments, those skilled in the art can make various alternative improvements and combinations to the above-mentioned specific embodiments.

Claims

1. A method for collaboratively constructing random fields of permeability coefficient and porosity in heterogeneous aquifers. It is characterized in that include: Step 1: Construct a two-dimensional rectangular coordinate system for the study area, collect the permeability and porosity data of the exploration holes in the aquifer in the study area, calculate the mean and variance of the permeability and porosity, and determine the correlation distance of the site permeability and porosity; Step 2: Grid the study area and calculate the log-normal distribution random field of permeability coefficient; Step 3: Generate a porosity normal distribution random field by correlating random variables; In step 2, the grid should be divided according to the variability of the exploration data, the length of the cell should be determined, and the difference of the parameter values ​​in each cell should be small. The random field formula of the logarithmic normal distribution of the permeability coefficient is: In the formula, K represents the site permeability coefficient, which is a random variable; E(K) represents the mean value of the site permeability coefficient; cv represents the coefficient of variation, D(K) represents the variance of the site permeability coefficient; Lε 1 represents a random variable that obeys the m-dimensional normal distribution N(0,C), where m is the number of grids in the site, and C=(C ij ) m×m represents its covariance matrix, whose elements represents the spatial correlation between grid i and grid j, Δx ij with Δy ij represents the difference between the coordinates of grid i and grid j in the X and Y directions, L x and L y They represent the relevant distances of the site permeability coefficient in the X and Y directions respectively; The derivation process of the associated random variable in step 3 is as follows: Taking the logarithm of both sides of formula (1) we get: According to the properties of lognormal distribution, lnK obeys normal distribution, and the mean of lnK is Variance D(ln K)=ln(cv 2 +1), so the above formula can be transformed as follows: Since the porosity obeys the normal distribution, referring to the normal distribution expression of formula (3), a porosity random field calculation formula is as follows: Where n represents the porosity of the site, which is a random variable; E(n) represents the mean of the porosity of the site; D(n) represents the variance of the porosity of the site; Lε 2 Still represents a random variable that obeys the m-dimensional normal distribution N(0,C); Since the permeability coefficient and porosity of the same research site are closely related, it can be seen that the random field of permeability coefficient and the random field of porosity are not independent, that is, Lε 1 With Lε 2 are not independent of each other, from formula (2) we can get It obeys the normal distribution N(0,C). Now we introduce the random variable η and the constant ρ(ρ∈[-1,1]). We assume that η also obeys the normal distribution N(0,C) and Lε 1 is independent of η, then It also obeys the normal distribution N(0,C), so we can make (4) Thus, the calculation formula of the porosity random field associated with the permeability random field is obtained: The porosity random field given by formula (5) is related to the permeability random field and also includes the randomness of the porosity itself; Use n i represents the porosity of grid i, K i represents the permeability coefficient of grid i (i = 1, 2, L, m), and n is calculated according to formulas (3) and (5): i and ln K i The covariance of get It can be seen that ρ represents the correlation coefficient between porosity n and the logarithm of permeability lnK, and ρ∈[-1,1]; 0<ρ≤1 indicates positive correlation, -1≤ρ<0 indicates negative correlation, and ρ=0 indicates no correlation. Since the permeability coefficient and porosity are related by the spatial random variable Lε 1 So Lε 1 It is called a correlated random variable; The range of n is specified as [E(n)-δ, E(n)+δ], where 0<δ<E(n). When randomly sampling porosity, the standard deviation is limited according to the properties of the normal distribution. It is stipulated that when n<E(n)-δ, the porosity is assigned as E(n)-δ; when n>E(n)+δ, the porosity is assigned as E(n)+δ.

2. According to the method for collaboratively constructing a random field of permeability coefficient and a random field of porosity in a heterogeneous aquifer according to claim 1, Features: The exploration holes in step 1 should be arranged according to the characteristics of the area, shape, purpose, and pollution status of the study area, using the diagonal point distribution method, plum blossom point distribution method, chessboard point distribution method, or "S" shape point distribution method. The relevant distance can be calculated using the recursive space method, correlation function method, and variation function method.

Citation Information

Patent Citations

  • Simulation method for directly generating logarithmic normal random field

    CN112071376A

  • In-situ mineral leaching analysis method based on ionic rare earth pore priority flow

    CN112464582A

  • Aquifer parameter field inversion method based on convolutional generative adversarial network

    CN114818548A

  • Method of adaptation of hydrodynamic model of productive formation of oil and gas condensate deposit taking into account uncertainty of geological structure

    RU2709047C1

  • Method and system for inversion of high-resolution aquifer storage coefficient based on gravity satellite data

    US20210375034A1