A Decision-Making Method for Exploration Well Deployment in Non-Ferrous Metal Mines Using Geometric Inversion Technology

Abstract

The present invention discloses a method for decision-making on deployment of exploration wells for nonferrous metal mineral resources using geometric inversion technology, comprising the following steps: S1, randomly generating simple geometric shapes to simulate the spatial distribution law of underground ore bodies, using a self-developed program to randomly generate geometric shapes; S2, reducing subjective tendencies and perceptions through millions of random samplings; S3, retaining models that match actual geophysical observation data in millions of random samplings; S4, obtaining a geometric probability distribution model by calculating the geometric distribution of the model at different depths; S5, selecting the location with the highest probability for deployment of exploration wells through statistical probability. The present invention adopts the above-mentioned method for decision-making on deployment of exploration wells for nonferrous metal mineral resources using geometric inversion technology, converts traditional subjective judgments into objective judgments based on Bayesian probability statistics, guides exploration activities through the idea of ​​probability theory, improves the success rate of exploration, reduces economic losses, and accurately and quantitatively characterizes the spatial distribution law of underground nonferrous metal ore bodies.

Description

Technical Field

[0001] The invention relates to the technical field of exploration and development of nonferrous metal mineral resources, and in particular to a nonferrous metal mineral exploration well deployment decision-making method using geometric inversion technology. Background Art

[0002] The traditional non-ferrous mineral resource exploration well design and deployment scheme relies on geophysical inversion models and subjective professional judgment. However, this method faces two major problems:

[0003] First, traditional geophysical inversion models mainly analyze physical properties such as density, magnetic susceptibility, and resistivity (hence the name physical property inversion). This physical property inversion has difficulties in describing the geometric characteristics of underground anomalies. During the deployment of exploration wells, we pay more attention to geometric characteristics rather than pure physical characteristics, which makes traditional physical property inversion models unsuitable for direct application in exploration well deployment.

[0004] Secondly, reliance on subjective professional judgment introduces uncertainty. Subjective judgment is often biased and may lead to increased risks and economic losses. In addition, subjective judgment is also subject to geographical restrictions and may not be widely applicable to exploration well designs in different regions. This limitation needs to be overcome through a more objective and systematic approach.

[0005] The present invention can improve the success rate of strategic nonferrous metal mineral exploration, help to discover more metal mineral resources with a higher success rate, thereby reducing exploration costs and improving economic value. Summary of the invention

[0006] The purpose of the present invention is to provide a non-ferrous metal mineral exploration well deployment decision-making method using geometric inversion technology, which transforms traditional subjective judgment into objective judgment based on Bayesian probability statistics, guides actual high-risk and high-cost exploration activities through the idea of ​​probability theory, reduces economic losses, and can accurately and quantitatively characterize the spatial distribution law of underground non-ferrous metal ore bodies.

[0007] To achieve the above object, the present invention provides a non-ferrous metal mineral exploration well deployment decision method using geometric inversion technology, comprising the following steps:

[0008] S1. Simulate the complex spatial distribution of underground ore bodies by randomly generating simple geometric shapes, using MATLAB or Python programs to randomly generate geometric shapes;

[0009] S2. Reduce subjective tendencies and perceptions through millions of random samplings;

[0010] S3, retain the models that match the actual geophysical observation data among millions of random samplings;

[0011] S4, obtaining a geometric probability distribution model by calculating the geometric distribution of the model at different depths;

[0012] S5. Based on statistical probability, select the location with the highest probability to deploy an exploration well.

[0013] Preferably, in step S1, the step of generating a geometric shape is:

[0014] S1-1. Set the spatial position range of the geometric shape according to the size of the work area;

[0015] S1-2. Randomly generate parameters to determine the geometric shape within the work area. The geometric parameters include: center coordinate position, major and minor axis lengths, and rotation angles;

[0016] S1-3. Through the computational geometry alpha-shape algorithm, discrete geometric shapes are connected into a unified whole to simulate the spatial distribution of underground ore bodies.

[0017] Preferably, in step S2, during the random sampling process, there are the following four sampling methods: a. changing the number of geometric shapes, b. changing the positions of geometric shapes, c. changing the sizes of geometric shapes, and d. changing the angles of geometric shapes. Each time, one of the four methods is randomly selected for iteration, and each random sampling method has a 25% probability of being selected.

[0018] Preferably, in step S3, the specific steps of retaining the model that matches the actual geophysical observation data among millions of random samplings are as follows:

[0019] S3-1. Perform geophysical forward modeling on the random ore body model to generate forward prediction data;

[0020] S3-2, compare the predicted data with the actual observed data to obtain the residual;

[0021] S3-3. Keep the model with residuals within a certain threshold, where the threshold is less than half of the number of actual geophysical observation data samples.

[0022] Preferably, in step S4, the specific implementation method of the geometric probability distribution model is as follows:

[0023] S4-1, convert the geometric model into binary, 1 represents ore body and 0 represents non-ore body;

[0024] S4-2, a large number of random sampling results are obtained at the same spatial location;

[0025] S4-3. Calculate the probability of ore body distribution by calculating the frequency of ore body occurrence.

[0026] The advantages and positive effects of the nonferrous metal mineral exploration well deployment decision-making method using geometric inversion technology described in the present invention are:

[0027] The present invention adopts the above-mentioned non-ferrous metal mineral exploration well deployment decision-making method using geometric inversion technology, transforms traditional subjective judgment into objective judgment based on Bayesian probability statistics, guides actual high-risk and high-cost exploration activities through the idea of ​​probability theory, reduces economic losses, and can accurately and quantitatively characterize the spatial distribution law of underground non-ferrous metal ore bodies.

[0028] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a schematic diagram of the probability model obtained by the binary model calculation of the present invention;

[0030] Figure 2 A diagram showing the relationship between the probability distribution and the deployment of exploration wells in the present invention;

[0031] Figure 3 It is a schematic diagram of the application of the geometric random inversion method of the present invention in actual mineral exploration;

[0032] Figure 4 Graph showing the relationship between the center coordinates of the ellipsoid and the semi-major axis length of the ellipsoid at different stages in the embodiment of the present invention, wherein A is the first stage, B is the second stage, and C is the third stage. DETAILED DESCRIPTION

[0033] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0034] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0035] A non-ferrous metal mineral exploration well deployment decision method using geometric inversion technology includes the following steps:

[0036] S1. Simulate the complex spatial distribution of underground ore bodies by randomly generating simple geometric shapes. Use MATLAB or Python programs to randomly generate geometric shapes. Set the spatial position range of the combined shapes according to the size of the work area. Randomly generate parameters to determine the geometric shapes within the work area. The geometric parameters include: center coordinate position, major and minor axis lengths, and rotation angles. Use the computational geometry algorithm "alpha-shape" to connect discrete geometric shapes into a unified whole to simulate the spatial distribution of underground ore bodies.

[0037] S2. Reduce subjective tendencies and perceptions through millions of random samplings. In the random sampling process, there are four sampling methods: a. changing the number of geometric shapes, b. changing the positions of geometric shapes, c. changing the sizes of geometric shapes, and d. changing the angles of geometric shapes. Each time, one of the four methods is randomly selected for iteration. Each random sampling method has a 25% probability of being selected.

[0038] S3. Keep the models that match the actual geophysical observation data among millions of random samplings; perform geophysical forward modeling on the random model ore body model to generate forward modeling prediction data; compare the prediction data with the actual observation data to obtain the residual; keep the models whose residuals are within a certain threshold. The threshold range is less than half of the number of samples of actual geophysical observation data.

[0039] S4. By calculating the geometric distribution of the model at different depths, a geometric probability distribution model is obtained; the geometric model is converted into binary, 1 represents an ore body and 0 represents a non-ore body; a large number of random sampling results are obtained at the same spatial position, that is, a series of arrays of 0 or 1; the distribution probability is calculated by calculating the frequency of 1. At a certain point, the number of random samplings is N, the number of 1 appearances is M, and the probability of the existence of an ore body at this point is M / N*100M / N*100.

[0040] S5. Obtain the statistical probability of the spatial distribution of the ore body through calculation, and select the location with the highest probability to deploy the exploration well.

[0041] Example

[0042] A non-ferrous metal mineral exploration well deployment decision method using geometric inversion technology (such as Figure 3 As shown), comprising the following steps:

[0043] 1. Simulate the complex spatial distribution of underground ore bodies by randomly generating simple geometric ellipses. Use self-developed MATLAB or Python programs to randomly generate ellipses. Specific steps: (1) Set the spatial position range of the geometric shape according to the size of the work area, for example, 0-1000 meters; (2) Randomly generate parameters to determine the ellipse within the work area. The geometric parameters include: center coordinate position, major and minor axis lengths, and rotation angles; (3) Use the computational geometry alpha-shape algorithm to connect discrete ellipses into a unified whole to simulate the spatial distribution of underground ore bodies.

[0044] 2. Reduce subjective tendencies and perceptions through millions of random samplings. In the random sampling process, there are four ways: (1) changing the number of ellipses; (2) changing the position of the ellipses; (3) changing the size of the ellipses; (4) changing the angle of the ellipses. Each time, one of the four ways is randomly selected for iteration, and each random sampling method has a 25% probability of being selected, such as Figure 1 shown.

[0045] 3. Retain the models that match the actual geophysical observation data among millions of random samplings. The specific steps are as follows: (1) Perform geophysical forward modeling on the random model ore body model to generate forward prediction data; (2) Compare the prediction data with the actual observation data to obtain the residual (RMS); (3) Retain the models whose residuals are within a certain threshold, and the threshold range is less than half of the number of samples of actual geophysical observation data.

[0046] 4. The geometric distribution of the model at different depths is calculated to obtain the geometric probability distribution model. The specific method of the geometric probability distribution model is as follows: (1) Convert the geometric model to binary, where 1 represents an ore body and 0 represents a non-ore body; (2) At the same spatial position, a large number of random sampling results are obtained, that is, a series of arrays of 0 or 1; (3) By calculating the frequency of 1, the probability of ore body distribution is calculated. At a certain point, the number of random sampling is N, the number of 1 appearances is M, and the probability of the existence of an ore body at this point is M / N*100M / N*100.

[0047] 5. Through statistical probability, select the location with the highest probability, that is, the area covered by the ellipse, and deploy the exploration well, such as Figure 2 shown.

[0048] Figure 1 How to calculate the spatial distribution probability model of the ore body through the binary model? Figure 1 Four binary models are randomly generated in the , where 1 represents the ore body distribution location and 0 represents the non-ore body distribution. All four models reflect that the middle grid is the ore body, but the ore body distribution in the corners is different. In the probability model, the probability that the middle position is the ore body is 100%, and the probability that the four corners are the ore body is 25%, that is, one quarter.

[0049] Figure 2 The relationship between probability distribution and exploration well deployment is shown. The dotted line represents a randomly generated ellipse, the solid trapezoid of the inclined stratum represents the stratum where the ore body is located, and the black solid line represents the exploration well. During the exploration process, in order to improve the success rate of exploration, maximize economic value and reduce costs, it is expected that an exploration well can accurately drill the target ore body. The location with the highest probability of the target ore body is where multiple ellipses overlap.

[0050] Figure 3This is a schematic diagram of the application of geometric random inversion method in actual mineral exploration. Figure 3 This is a well deployment decision diagram generated by geometric random inversion, where the background color is the total magnetic field intensity, the dot color represents the depth range, and the dot size represents the probability of drilling to find the target ore body. At any position in this diagram, the drilling depth and the probability of encountering an ore body can be known. Figure 3 Taking the position close to the north as an example, within the depth range of 50-200m at this position, there is a greater than 90% probability of hitting the target ore body; taking the position far from the north as an example, within the depth range of 200-280m at this position, there is a greater than 50-90% probability of hitting the target ore body.

[0051] Figure 4 In the figure, x, y, and z represent the center coordinates of the randomly generated ellipsoid, a represents the length of the semi-major axis of the randomly generated ellipsoid, and each scattered point represents the parameters of the randomly generated ellipsoid. Phase I: Generate a series of ellipsoids by random sampling, whose center coordinates and semi-major axis lengths follow a uniform distribution. At this stage, it is difficult to accurately determine the distribution of the ore body. Phase II: Use geophysical gravity data for matching. The results show that after verification with geophysical data, the parameter distribution range of the ellipsoid is significantly compressed, indicating that the uncertainty of the parameters is reduced, and the spatial distribution of the ore body can be roughly determined. Phase III: Further matching using geophysical borehole gravity data further compresses the parameter distribution range. The results show that at this stage, the position distribution of the ellipsoid is more specific. For example, taking the depth position as an example, the numerical range of z is significantly reduced, and the distribution depth of the ore body is basically determined.

[0052] Therefore, the present invention adopts the above-mentioned non-ferrous metal mineral exploration well deployment decision-making method using geometric inversion technology, transforms the traditional subjective judgment into an objective judgment based on Bayesian probability statistics, and guides actual high-risk and high-cost exploration activities through the idea of ​​probability theory, reduces economic losses, and can accurately and quantitatively characterize the spatial distribution law of underground non-ferrous metal ore bodies.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A non-ferrous metal mineral exploration well deployment decision-making method using geometric inversion technology, characterized in that: The steps include: S1. Simulate the complex spatial distribution of underground ore bodies by randomly generating simple geometric shapes, using MATLAB or Python programs to randomly generate geometric shapes; S2. Reduce subjective tendencies and perceptions through millions of random samplings; S3, retain the models that match the actual geophysical observation data among millions of random samplings; S4, obtaining a geometric probability distribution model by calculating the geometric distribution of the model at different depths; S5. Select the location with the highest probability to deploy the exploration well through statistical probability; In step S1, the steps of generating geometric shapes are: S1-1. Set the spatial position range of the geometric shape according to the size of the work area; S1-2. Randomly generate parameters to determine the geometric shape within the work area. The geometric parameters include: center coordinate position, major and minor axis lengths, and rotation angles; S1-3, through the computational geometry alpha-shape algorithm, discrete geometric shapes are connected into a unified whole to simulate the spatial distribution of underground ore bodies; In step S2, during the random sampling process, there are four sampling methods: a. changing the number of geometric shapes, b. changing the position of geometric shapes, c. changing the size of geometric shapes, and d. changing the angle of geometric shapes. One of the four methods is randomly selected for iteration each time, and each random sampling method has a 25% probability of being selected. In step S4, the specific implementation method of the geometric probability distribution model is as follows: S4-1, convert the geometric model into binary, 1 represents ore body and 0 represents non-ore body; S4-2, a large number of random sampling results are obtained at the same spatial location; S4-3. Calculate the probability of ore body distribution by calculating the frequency of ore body occurrence.

2. The method for decision-making on well deployment for nonferrous metal mineral exploration using geometric inversion technology according to claim 1 is characterized in that: In step S3, the specific steps of retaining the model that matches the actual geophysical observation data among millions of random samplings are as follows: S3-1. Perform geophysical forward modeling on random ore body models to generate forward prediction data; S3-2, compare the predicted data with the actual observed data to obtain the residual; S3-3. Keep the model with residuals within a certain threshold, where the threshold is less than half of the number of actual geophysical observation data samples.

Images