Evaluation method and system for fine division of water richness of unconsolidated aquifer in mine shaft

Abstract

An evaluation method and system for fine division of the water richness of an unconsolidated aquifer in a mine shaft. The method comprises: using an aquifer thickness, a sand and gravel bed thickness, a lithology-thickness index, a sandstone-to-formation thickness ratio and the number of sand and gravel beds to constitute evaluation indexes, and thus establishing an evaluation index matrix; unifying variation ranges of different evaluation index data, performing normalization processing to obtain a processed standard evaluation matrix, and thus constructing a water richness evaluation index system of a "bottom aquifer"; in light of the constructed water richness evaluation index system of the "bottom aquifer", according to the principle of a projection pursuit model, establishing an aquifer water richness evaluation model by means of DPS software; in the aquifer water richness evaluation model, performing linear projection and constructing a projection objective function; optimizing the projection objective function to determine an optimal projection direction; and substituting the optimal projection direction into an expression of linear projection to obtain projection eigenvalues capable of reflecting each evaluation index information, so as to perform comprehensive evaluation on an evaluation object.

Description

An evaluation method and system for fine zoning of water-richness of loose aquifers in mines Technical Field

[0001] The present invention belongs to the technical field of resource evaluation systems, and in particular relates to an evaluation method and system for fine zoning of water-richness of loose aquifers in mines. Background Art

[0002] Currently, objectively and rationally predicting and evaluating the water content of aquifers that influence and threaten safe coal seam recovery is a crucial prerequisite for mine water hazard prevention and control. This not only enables targeted planning and deployment of relevant water control projects, but also effectively prevents or reduces the likelihood of water inrush (or inrush) accidents in localized areas of high water content. Previous aquifer water content assessments have primarily focused on fractured aquifers in bedrock and karst aquifers in limestone, with relatively little evaluation of the water content of unconsolidated Cenozoic aquifers. The bottom aquifer (referred to as the "bottom aquifer") in the unconsolidated Cenozoic strata in the Huaibei mining area often directly overlies coal-bearing strata. When mining shallow coal seams, mining-induced fractures can lead to the upper "bottom aquifer," causing roof water inrush accidents. This aquifer is the material basis for roof water inrush, and its water content directly determines the volume and duration of the inrush. Therefore, an objective, reasonable and accurate prediction and evaluation of the water-richness of the "bottom content" is of great significance for ensuring the safe and efficient mining of shallow coal seams in mines and protecting groundwater resources.

[0003] Sedimentary characteristics are the most fundamental controlling factors for aquifer formation and groundwater storage. The thickness, lithologic structure, and spatial distribution of "bottom-containing" sediments formed in different sedimentary environments vary, leading to significant differences in their water-yield properties, exhibiting complex controlling mechanisms and nonlinear characteristics. In existing technologies, most aquifer water-yield evaluation methods employ methods or models to calculate the weights and combined values ​​of various evaluation indicators, establishing a linear relationship between aquifer water-yield and these indicators. However, these methods ignore the useful information regarding the nonlinear relationship between aquifer water-yield and these indicators, compromising the accuracy of water-yield evaluation results. Furthermore, traditional multi-factor comprehensive analysis methods assume that the evaluation indicator data generally follows a normal distribution. However, in reality, many high-dimensional data do not meet this assumption. The high-dimensional, nonlinear, and non-normal distribution of the evaluation indicator data in aquifer water-yield evaluation renders many excellent models and methods unsuitable for water-yield evaluation calculations, necessitating robust or nonparametric approaches.

[0004] Through the above analysis, the problems and defects of the existing technology are as follows: the water-richness evaluation of aquifers in the existing technology is mostly carried out around the fissure aquifers in bedrock, the karst aquifers in limestone, etc., and there are relatively few studies on the water-richness of loose aquifers in the Cenozoic. There is no mature water-richness evaluation method to objectively, reasonably and accurately predict and evaluate the water-richness of the "bottom aquifer". Summary of the Invention

[0005] In response to the problems existing in the prior art, the present invention provides a method and system for evaluating the fine zoning of water-richness of loose aquifers in mines.

[0006] The present invention is achieved by providing a method for evaluating the water-richness of loose aquifers in mines by fine zoning, the method comprising the following steps:

[0007] S101, the evaluation index matrix is ​​established by combining the thickness of the aquifer, the thickness of the gravel layer, the lithology-thickness index, the sand-to-ground ratio, and the number of gravel layers;

[0008] S102, unifying the variation ranges of different evaluation index data, normalizing the evaluation indexes with larger values ​​and those with smaller values, obtaining a normalized evaluation matrix, and constructing a "bottom-content" water-richness evaluation index system;

[0009] S103, combining the constructed “bottom-containing” water-richness evaluation index system and based on the principle of projection pursuit model, establish the aquifer water-richness evaluation model through DPS software;

[0010] S104, performing linear projection and constructing a projection objective function in the aquifer water-richness evaluation model;

[0011] S105, optimizing the projection objective function to determine the optimal projection direction;

[0012] S106, bringing the optimal projection direction back into the linear projection expression to obtain projection eigenvalues ​​that can reflect the information of each evaluation index;

[0013] S107: Comprehensively evaluate the evaluation object based on the distribution characteristics and size of the projection eigenvalues.

[0014] Furthermore, the aquifer thickness, gravel layer thickness, lithologic structure index, and sand-to-ground ratio in S101 are all positive indicators affecting the water-richness of the aquifer, while the number of gravel layers is a negative indicator affecting the water-richness.

[0015] Furthermore, in S105, the projection objective function is optimized by using a composite method to optimize and solve the projection objective function.

[0016] Furthermore, in said S106, the magnitude of the absolute value of each component of the optimal projection direction essentially reflects the contribution of each evaluation index to the bottom water-richness evaluation result. The larger the absolute value of each component, the greater the influence of the corresponding evaluation index on the bottom water-richness, and vice versa.

[0017] Furthermore, after visualization processing of the projection characteristic values ​​in S107, an isoline map of the bottom water-rich projection characteristic values ​​can be obtained.

[0018] Furthermore, the size of the projection characteristic value in S107 reflects the water-richness of the bottom, that is, the larger the projection characteristic value, the stronger the water-richness of the bottom.

[0019] Another object of the present invention is to provide an evaluation system for fine zoning of water-richness of loose aquifers in mines, which implements the evaluation method for fine zoning of water-richness of loose aquifers in mines. The evaluation system for fine zoning of water-richness of loose aquifers in mines comprises:

[0020] Evaluation index module, used to determine evaluation index;

[0021] An evaluation index system establishment module, connected with the evaluation index module, is used to establish an evaluation index system;

[0022] Projection pursuit module, used to provide a projection pursuit model;

[0023] The aquifer water-richness evaluation module is connected to the evaluation index system establishment module and the projection pursuit module, and is used to establish an aquifer water-richness evaluation model by combining the data of the evaluation index system establishment module and the model of the projection pursuit module through DPS software.

[0024] Another object of the present invention is to provide a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the evaluation method for the refined zoning of the water-richness of the loose aquifer in the mine.

[0025] Another object of the present invention is to provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to execute the steps of the evaluation method for refined zoning of water-richness of loose aquifers in mines.

[0026] Another object of the present invention is to provide an information data processing terminal, which is used to implement an evaluation system for fine-grained zoning of the water-richness of the loose aquifer in the mine.

[0027] First, objectively and rationally predicting and evaluating the water-richness of aquifers that affect and threaten safe coal seam recovery is a crucial prerequisite for mine water prevention and control efforts. This not only allows for targeted planning and deployment of relevant water prevention and control projects, but also effectively avoids or reduces the likelihood of water inrush (or sudden inrush) accidents in localized areas of high water content. Currently, domestic and international experts and scholars have conducted extensive and fruitful research on the evaluation of the water-richness of aquifers that affect coal seam mining. Evaluation methods primarily fall into three categories: geophysical methods (such as transient electromagnetic methods, direct current methods, and mine audio fluoroscopy), water pumping and drainage tests, and multi-factor comprehensive analysis. Geophysical methods are labor-intensive and costly, and their multi-solution nature means the interpretation of these results depends to a certain extent on the experience and technical skill of the interpreter. The unit water yield (q-value) obtained from pumping / draining tests can intuitively reflect the water-rich characteristics of a mine aquifer. However, due to the low level of hydrogeological exploration in most mining areas in my country and the small number of pumping / draining tests conducted, the number of q-values ​​obtained is very limited, failing to meet the requirements for detailed control of the distribution of aquifer water-richness, making it difficult to accurately evaluate and rationally zonate aquifer water-richness. Furthermore, due to the significant differences in coal seam occurrence, mine geology, and hydrogeological conditions across different mining areas, relying solely on q-values ​​to evaluate aquifer water-richness is incomplete. In contrast, the multi-factor comprehensive analysis method is a simple and effective approach that leverages a large amount of existing geological exploration borehole data and fully explores and utilizes multiple sources of information that influence aquifer water-richness, enabling a more comprehensive and accurate assessment of aquifer water-richness. This present invention provides a mature water-richness evaluation method that objectively, rationally, and accurately predicts and evaluates the water-richness of "bottom-containing" aquifers.

[0028] Second, further optimization is needed in the refined evaluation of aquifer water-yield and the selection of appropriate evaluation models. Previous studies on aquifer water-yield evaluation models based on multi-factor fusion analysis have mostly focused on fractured aquifers in bedrock, with relatively little research on the refined evaluation of water-yield in loose porous aquifers. Furthermore, previous studies on the quantitative evaluation of aquifer water-yield have mostly used methods to calculate the weights and comprehensive values ​​of the evaluation indicators, establishing a linear relationship between water-yield and the evaluation value. However, this approach ignores the useful information about the nonlinear relationship between water-yield and the evaluation indicators, affecting the accuracy of the evaluation results. Furthermore, traditional multi-factor comprehensive analysis methods for water-yield are based on the assumption that the indicator data generally follow a normal distribution. However, the evaluation indicator data affecting water-yield are high-dimensional, nonlinear, and not necessarily normally distributed, rendering many excellent models and methods unsuitable for quantitative water-yield evaluation.

[0029] This method is unaffected by high-dimensional, nonlinear, and non-normally distributed evaluation index data, breaking through the limitation of traditional comprehensive water-richness evaluation methods, which can only process normally distributed data. Although the model is based on linear projections of the data, it seeks nonlinear structures within these linear projections, making it suitable for solving nonlinear problems.

[0030] The present invention searches for the optimal projection direction entirely based on the structural characteristics of the data itself, avoids human subjectivity in weight assignment, reduces interference from human factors, enables the evaluation results to reflect the characteristics of the original data to the greatest extent, and improves the scientificity and accuracy of the evaluation results. The present invention can eliminate the interference of variables that are irrelevant to or have little relationship with the structure and characteristics of the data. Using the projection eigenvalue as the basis for evaluating the degree of water-richness, the water-richness of each region can be intuitively compared. The present invention can achieve an accurate evaluation of the water-richness of mine aquifers, and the model is relatively simple, intuitive, and easy to understand. The present invention has strong operability and effectiveness, and has good application value in the field of aquifer water-richness evaluation. The present invention provides a new evaluation method and approach for the evaluation of aquifer water-richness involving multiple factors.

[0031] Third, as auxiliary evidence for the inventiveness of the claims of the present invention, it is also reflected in the following important aspects:

[0032] The expected benefits and commercial value of the technical solution of the present invention after conversion are:

[0033] Accurately predicting the aquifers in mines is of great significance for ensuring the safe and efficient recovery of coal seams. Before arranging the working face and recovering the coal seams, each mine needs to conduct a water-richness study on the aquifers that affect safe production. If the water-richness detection is carried out by geophysical methods or the water pumping test method (drilling hydrogeological holes) is used to obtain the unit water yield (q value) to evaluate the water-richness of the aquifer, the workload is large and the cost is high, and the detection area is mostly a small area, which is bound to significantly increase the production cost of the mine. If the technical solution of the present invention is transformed, it can be widely used in the water-richness evaluation of aquifers in various mines in multiple mining areas in my country, saving the high costs required for geophysical work, water pumping test work, etc., effectively reducing the production cost of coal mines, and having great commercial value.

[0034] (2) The technical solution of the present invention fills a technological gap in the industry both domestically and internationally: Previous research on aquifer water-yield evaluation models based on multi-factor fusion analysis has mostly focused on fractured aquifers in bedrock, while relatively little research has been conducted on the refined evaluation of water-yield in loose porous aquifers. The present invention proposes a method suitable for the refined evaluation of water-yield in loose porous aquifers.

[0035] (3) Does the technical solution of the present invention solve the technical problems that people have been eager to solve but have never been able to solve successfully? In previous studies, the quantitative evaluation of the water-richness of aquifers mostly used some methods to calculate the weights and comprehensive values ​​of the participating indicators and establish a linear relationship between water-richness and the evaluation value, but ignored the useful information of the nonlinear part between water-richness and the evaluation indicators, which affected the accuracy of the evaluation results. At the same time, the traditional multi-factor comprehensive analysis method of water-richness is based on the assumption that the indicator data as a whole obeys the normal distribution. However, the evaluation indicator data that affects water-richness has the characteristics of high dimension, nonlinearity and not necessarily conforming to the normal distribution, resulting in many excellent models and methods not being applicable to the quantitative evaluation calculation of water-richness. The present invention establishes an evaluation model and method that can effectively reflect the nonlinear characteristics between water-richness and evaluation indicators and has a high tolerance for data structure characteristics.

[0036] Fourth, the present invention provides a refined zoning evaluation method for assessing the water richness of loose aquifers in mines. Its main features include the following steps:

[0037] S101: Develop an evaluation matrix based on the aquifer thickness, gravel layer thickness, lithology-thickness index, sand-to-ground ratio, and number of gravel layers. Establishing this matrix is ​​the first step in assessing the water-richness of a mine. This includes assessing key indicators such as aquifer thickness, gravel layer thickness, the relationship between lithology and thickness, and sand-to-ground ratio and number of gravel layers.

[0038] S102: Unify the ranges of variation for different evaluation indicators, normalize indicators with larger values ​​and those with smaller values, and obtain a standardized evaluation matrix to construct a "bottom-water" water-richness evaluation index system. This step primarily normalizes the different evaluation indicators to unify their ranges. This allows for comparison of each indicator on a consistent scale, leading to a better comprehensive evaluation.

[0039] S103: Integrating the established bottom-water-yield evaluation index system and using the projection pursuit model, a water-yield evaluation model was established using DPS software. Using the projection pursuit model and DPS software, an evaluation model was constructed. This model, combined with the evaluation index system, enables more accurate aquifer water-yield assessment.

[0040] S104: Perform linear projection and construct projection objective function in the aquifer water richness evaluation model.

[0041] In the evaluation model, linear projection and projection objective function are used to determine the best evaluation strategy.

[0042] S105: Optimize the projection objective function to determine the optimal projection direction. By optimizing the projection objective function, the optimal projection direction is determined. This direction can best reflect the information of the evaluation index.

[0043] S106: Substitute the optimal projection direction back into the linear projection expression to obtain projection eigenvalues ​​that reflect information about each evaluation indicator. This step involves substituting the optimal projection direction back into the linear projection expression to obtain projection eigenvalues ​​that reflect information about each evaluation indicator. These eigenvalues ​​can quantify the importance of each evaluation indicator.

[0044] S107: Comprehensively evaluate the evaluation object based on the distribution characteristics and magnitude of the projected eigenvalues. The final step is to comprehensively evaluate the evaluation object based on the distribution characteristics and magnitude of the projected eigenvalues. This step integrates various evaluation indicators to obtain the final aquifer water-richness evaluation result.

[0045] The significant technical advancement of this method is that it combines multiple evaluation indicators and obtains more accurate aquifer water richness assessment results through normalization processing and projection pursuit model. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] FIG1 is a flow chart of a method for evaluating fine-grained zoning of water-richness of loose aquifers in mines provided by an embodiment of the present invention.

[0047] FIG2 is a structural diagram of an evaluation system for fine-grained zoning of water-richness of loose aquifers in mines provided by an embodiment of the present invention.

[0048] FIG3 is an optimal projection direction of the “bottom-containing” water-rich projection pursuit model provided by an embodiment of the present invention.

[0049] FIG4 is a contour map of the projection characteristic values ​​of the “bottom-containing” water-richness of the study area provided by an embodiment of the present invention.

[0050] FIG5 is a “bottom-containing” water-rich zoning map of the study area provided by an embodiment of the present invention.

[0051] FIG6 is a diagram showing the distribution of the “bottom-containing” sedimentary phases of the Xutuan mine provided by an embodiment of the present invention. Implementation Method

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0053] As shown in Figure 1, the evaluation method for fine zoning of water-richness of loose aquifers in mines includes the following steps:

[0054] S101, the evaluation index matrix is ​​established by combining the thickness of the aquifer, the thickness of the gravel layer, the lithology-thickness index, the sand-to-ground ratio, and the number of gravel layers;

[0055] S102, unifying the variation ranges of different evaluation index data, normalizing the evaluation indexes with larger values ​​and those with smaller values, obtaining a normalized evaluation matrix, and constructing a "bottom-content" water-richness evaluation index system;

[0056] S103, combining the constructed “bottom-containing” water-richness evaluation index system and based on the principle of projection pursuit model, establish the aquifer water-richness evaluation model through DPS software;

[0057] S104, performing linear projection and constructing a projection objective function in the aquifer water-richness evaluation model;

[0058] S105, optimizing the projection objective function to determine the optimal projection direction;

[0059] S106, bringing the optimal projection direction back into the linear projection expression to obtain projection eigenvalues ​​that can reflect the information of each evaluation index;

[0060] S107: Comprehensively evaluate the evaluation object based on the distribution characteristics and size of the projection eigenvalues.

[0061] The specific algorithm steps of the evaluation method for fine zoning of water-richness of loose aquifers in mines provided by the present invention are as follows:

[0062] (1) Establish an evaluation index matrix

[0063] Assume that the sample size of the sample set to be evaluated is n, the number of indicators in the evaluation indicator set is p, and the j-th evaluation indicator value of the i-th sample is , then all the indicator data of all samples to be evaluated can be represented by the n×p matrix X*:

[0064] (1)

[0065] (2) Normalization

[0066] In order to eliminate the influence of different dimensions and unify the range of variation of different evaluation index data, the evaluation indexes with larger values ​​are better and the indexes with smaller values ​​are better are normalized according to formula (2) and formula (3) respectively:

[0067] (2)

[0068] (3)

[0069] Where: 、 are the maximum and minimum values ​​of the jth index respectively; x ij The data are normalized.

[0070] Thus, the normalized n×p standardized evaluation matrix X is obtained:

[0071] (4)

[0072] (3) Linear projection

[0073] By observing the sample data from different angles, we can find the best projection direction that can fully mine the data information and reflect the data characteristics to the greatest extent, and reduce the high-dimensional data to low dimensions. Synthesize into a 1-dimensional vector Projection value z in the projection direction i :

[0074] (5)

[0075] The present invention utilizes a projection pursuit model. Although it is based on the linear projection of data, it seeks the nonlinear structure in the linear projection, so it can be used to solve nonlinear problems.

[0076] (4) Constructing the projection objective function

[0077] According to the classification principle, the distribution characteristics of the projection value should satisfy the following conditions as much as possible: the local projection points should be as dense as possible, preferably condensed into several point clusters; the overall projection point clusters should be as scattered as possible, that is, the inter-class distance S of the p-dimensional data in the 1-dimensional space should be z and intra-class density D z At the same time, the maximum value is reached and the projection objective function Q(a) is expressed as S z With D z The product of is shown in formula (6):

[0078] (6)

[0079] Where: S z Represents the distance between classes, which is numerically equal to z i The standard deviation of D z Represents z i The local density or intra-class density. z and D z Calculate according to formula (7) and formula (8) respectively:

[0080] (7)

[0081] (8)

[0082] Where: For sequence The mean of n is the sample size; R is the radius of the local density window. The value requirement is to make the average number of projection points contained in the window not too small to avoid excessive deviation of the sliding average, and at the same time, it cannot increase too much with the increase of n. In actual operation, , α is based on the projection point z i The distribution among regions is appropriately adjusted, and can be 0.1, 0.01, 0.001, etc., with 0.1 being the most common; r ij =|z i -z j |, is the distance between samples; u t is a unit step function, which changes with r ij The monotone density function decreases with the increase of ij )≥0, its value is equal to 1; when t=(Rr ij )<0, its value is equal to 0.

[0083] (5) Optimize the projection objective function and determine the optimal projection direction

[0084] Different projection directions indicate different data structure characteristics. The optimal projection direction is the one that can best mine data information and expose a certain type of characteristic structure of high-dimensional data to the greatest extent. For a given sample set index value, Q(a) is only related to the change in projection direction. Therefore, under certain constraints (st), the objective function can be maximized to optimize it and estimate the optimal projection direction. The maximization objective function is shown in Equation (9):

[0085] (9)

[0086] Formula (9) is a j To solve the complex nonlinear optimization problem of optimizing variables, the corresponding modules in current mathematical software such as MATLAB and DPS can realize the optimization of the projection objective function. The present invention uses DPS software to solve the optimal projection direction and projection value through a nonlinear optimization algorithm.

[0087] (6) Establish a comprehensive evaluation model

[0088] According to formula (9), the optimal projection direction can be obtained , The value can reflect the contribution of each evaluation index to the research object; Substituting into formula (5) we can get the optimal projection value of each sample , according to the projection value The original evaluation object is comprehensively evaluated based on its distribution characteristics and size.

[0089] Fine classification of aquifer water richness levels

[0090] According to the projection characteristic values ​​obtained by the comprehensive evaluation model, the natural break classification method is used to grade and evaluate the projection characteristic values, fully considering the extreme heterogeneity of the "bottom-content" water-richness, and further subdividing the weak water-richness level into relatively weak water-richness, slightly water-richness, and extremely weak water-richness. That is, the natural break classification method is used to divide the projection characteristic values ​​into 4 natural levels, namely, medium water-rich area, relatively weak water-rich area, slightly water-rich area, and extremely weak water-rich area.

[0091] In S101, this embodiment of the present invention objectively reflects and evaluates the comprehensive water-richness of the "bottom aquifer." Based on an analysis of sedimentary characteristics and facies, such as the thickness and lithologic composition of the "bottom aquifer," and in accordance with the laws governing water control by sedimentation, aquifer thickness, gravel layer thickness, lithologic-thickness index, sand-to-ground ratio, and the number of gravel layers (total number of gravel and sand layers) are selected as the primary controlling factors of the "bottom aquifer" water-richness. An evaluation index system composed of these five primary controlling factors is constructed to quantitatively evaluate the "bottom aquifer" water-richness. The relationship between each primary controlling factor and the aquifer's water-richness is described below. Aquifer thickness, gravel layer thickness, lithologic-texture index, and sand-to-ground ratio are all positive indicators of aquifer water-richness, while the number of gravel layers is a negative indicator.

[0092] In S105, the projection objective function is optimized by using a composite method. The composite method is the most effective optimization algorithm for processing and solving medium- and small-scale optimization problems. It has global convergence and can quickly calculate the optimal projection direction value. It can serve as an auxiliary optimization tool for the projection pursuit method.

[0093] In S106, the absolute value of each component in the optimal projection direction essentially reflects the contribution (influence) of each evaluation index to the evaluation result of the "bottom content" water-richness. The larger the absolute value of each component, the greater the influence of the corresponding evaluation index on the "bottom content" water-richness, and vice versa.

[0094] In S107, the projected eigenvalues ​​are visualized to produce a contour map of the projected eigenvalues ​​of the bottom water content. The magnitude of the projected eigenvalues ​​reflects the water content of the bottom water content; that is, the larger the projected eigenvalue, the greater the water content. Using the projected eigenvalues ​​as a basis for evaluating the water content of the bottom water content allows for intuitive comparison of the water content of various regions, demonstrating both scientific validity and practicality.

[0095] As shown in Figure 2, the evaluation system for fine-grained zoning of water-richness in loose aquifers in mines includes:

[0096] Evaluation index module, used to determine evaluation index;

[0097] An evaluation index system establishment module, connected with the evaluation index module, is used to establish an evaluation index system;

[0098] Projection pursuit module, used to provide a projection pursuit model;

[0099] The aquifer water-richness evaluation module is connected to the evaluation index system establishment module and the projection pursuit module, and is used to establish an aquifer water-richness evaluation model by combining the data of the evaluation index system establishment module and the model of the projection pursuit module through DPS software.

[0100] The evaluation method for fine-grained zoning of water-richness of loose aquifers in mines provided by an application embodiment of the present invention is applied to a computer device, wherein the computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the evaluation method for fine-grained zoning of water-richness of loose aquifers in mines.

[0101] The evaluation method for fine-grained zoning of water-richness of loose aquifers in mines provided by an application embodiment of the present invention is applied to an information data processing terminal, which is used to implement an evaluation system for fine-grained zoning of water-richness of loose aquifers in mines.

[0102] It should be noted that the embodiments of the present invention can be implemented by hardware, software, or a combination of software and hardware. The hardware portion can be implemented using dedicated logic; the software portion can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated design hardware. Those skilled in the art will appreciate that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or contained in processor control code, for example, such as a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and modules of the present invention can be implemented by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., can also be implemented by software executed by various types of processors, or can be implemented by a combination of the above-mentioned hardware circuits and software, such as firmware.

[0103] Example 1:

[0104] The present invention was used to evaluate the water-richness of aquifers in a large study area.

[0105] Model establishment:

[0106] Given the high-dimensional, nonlinear, and non-normal distribution of aquifer water-yield evaluation index data, a projection pursuit model was introduced to comprehensively evaluate the bottom-water-yield. This model effectively reflects the nonlinear relationship between water-yield and evaluation indexes while being highly inclusive of data structural characteristics. This model overcomes the limitation of traditional comprehensive evaluation methods, which are limited to processing normally distributed data, while also overcoming the significant subjectivity inherent in weight determination in conventional evaluation methods. The projection pursuit model fully reflects the inherent structural characteristics of the original data, eliminating interference from irrelevant variables and the influence of subjective factors, resulting in more objective evaluation results.

[0107] Integrating the previously constructed "bottom-containment" water-richness evaluation index system and following the principles of the projection pursuit model, 384 sets of borehole data from the study area were normalized and then a water-richness evaluation model was established using DPS software. The composite form method was used to optimize and solve the projection objective function. The composite form method is the most effective optimization algorithm for medium- and small-scale optimization problems. It exhibits global convergence and can quickly calculate the optimal projection direction, making it a suitable auxiliary optimization tool for the projection pursuit method.

[0108] After optimization, the maximum projection index function is obtained , the best projection direction is , as shown in Figure 3. The absolute value of each component in the optimal projection direction essentially reflects the contribution (degree of influence) of each evaluation index to the evaluation result of the "bottom-containment" water-richness. The larger the absolute value of each component, the greater the influence of the corresponding evaluation index on the "bottom-containment" water-richness, and vice versa. Although the optimal projection direction and the weight value of the evaluation index are essentially different, there are also connections and commonalities between the two. That is, both reflect the degree of influence of each evaluation index on the result and the inherent connection of the original data itself. The difference between the two is that the sum of the optimal projection direction values ​​of each evaluation index is not 1, while the sum of the weight values ​​of each evaluation index is 1.

[0109] The best projection direction of the present invention , indicating that the influence of gravel layer thickness, lithology-thickness index, aquifer thickness, sand-to-ground ratio and number of gravel layers on the "bottom-containing" water-richness of the study area decreases in turn, among which gravel layer thickness, lithology-thickness index and aquifer thickness are the most important factors affecting the "bottom-containing" water-richness of the study area, and the influence of the number of gravel layers is the smallest.

[0110] Substitute the obtained optimal projection direction a into formula (5) to obtain the projection eigenvalue z that can reflect the information of each evaluation index i , as shown in formula (10), and z iThe difference level of the bottom water content is used as the evaluation basis, and the evaluation results are determined by analysis. By visualizing the projection eigenvalues, the contour map of the projection eigenvalues ​​of the "bottom water content" is obtained, as shown in Figure 4. The size of the projection eigenvalue reflects the degree of water content of the "bottom water content", that is, the larger the projection eigenvalue, the stronger the water content of the "bottom water content". Using the projection eigenvalue as the basis for evaluating the degree of water content of the "bottom water content" can intuitively compare the water content of each area in the study area, which has certain scientificity and operability. As shown in Figure 4, the projection eigenvalue of the "bottom water content" in the northern part of the study area is larger, and that in the southern part is smaller, and the overall trend is gradually decreasing from south to north.

[0111]

[0112] Water-rich zoning evaluation:

[0113] The projection characteristic values ​​of the "bottom-containing" water-richness in the study area vary greatly in size, indicating that the "bottom-containing" water-richness is unevenly distributed and varies greatly. In order to further objectively and quantitatively characterize the spatial distribution characteristics of the "bottom-containing" water-richness, the natural break classification method is used to grade and evaluate the projection characteristic values. This is conducive to finely characterizing the strength and weakness of the aquifer water-richness, refining the internal differences in the aquifer water-richness, and identifying the key target areas for the prevention and control of "bottom-containing" water hazards in mines. The projection characteristic values ​​that characterize the water-richness of the aquifer reflect the natural properties of the aquifer. The natural break classification method is based on the inherent natural grouping of the data and identifies natural breakpoints by selecting the maximum similarity within the group or the maximum difference between groups. It emphasizes natural breakpoints and groupings, reducing the impact of human intervention. Therefore, it is reasonable and scientific to use the natural breakpoint classification method for classification.

[0114] In mine production practice, the water-richness of mine aquifers is generally classified based on the unit water yield of the borehole, in accordance with the classification standards in the "Detailed Rules for Coal Mine Water Prevention and Control." Analysis of data from 15 "bottom-containing" pumping (injection) test boreholes within the study area revealed that the unit water yield ranged from 0.00062 to 0.33580 L / (s·m), indicating weak to moderate water-richness. 66.67% of the boreholes showed weak water-richness, while 33.33% showed moderate water-richness. No data from extremely or strongly water-rich boreholes were found. However, the unit water yield values ​​varied widely, by several hundred times. In order to more objectively reflect the relative strength of the "bottom-containing" water-richness, and at the same time to make the water-richness zoning prediction results more convenient for mine water prevention and control work, the extreme heterogeneity of the "bottom-containing" water-richness is fully considered, and the weak water-richness level is further subdivided into relatively weak water-richness, weak water-richness, and extremely weak water-richness. That is, the natural break classification method is used to divide the projection eigenvalue into four natural levels, namely, medium water-rich area (projection eigenvalue > 0.5101), relatively weak water-rich area (projection eigenvalue 0.3155~0.5101), weak water-rich area (projection eigenvalue 0.1821~0.3155), and extremely weak water-rich area (projection eigenvalue <0.1821). A "bottom-containing" water-richness zoning map of the study area was drawn, see Figure 5. As can be seen from Figure 5, the overall water-richness of the "bottom content" in the study area shows a trend of gradually weakening from north to south. Among them, the moderately water-rich area is distributed in the northern part of the mine and near the south side of the Xutuan fault. The middle part of the mine is mainly a relatively weak water-rich area and a slightly water-rich area, and the southern part of the mine is mainly an extremely weak water-rich area.

[0115] Combined with sedimentary facies analysis (Figure 6), the southern "bottom-containment" is primarily composed of colliform-eluvial sediments, with very weak water-richness. The central and northern "bottom-containment" is primarily composed of multi-stage alluvial fan-facies sediments, gradually retrograded from east to west and from south to north. Its overall water-richness is stronger than that of the colliform-eluvial sediments. However, due to the superposition of different sedimentary subfacies / microfacies, the horizontal and vertical phase transitions are rapid, the lithologic structure is complex, and the water-richness is heterogeneous, with weak to moderate water-richness. The results of the water-richness evaluation zoning reflect the control of sedimentary phases on the water-richness of the "bottom-containment".

[0116] Validation of the water-richness evaluation model:

[0117] The unit yield is the most direct indicator of aquifer water-yield strength. The reliability of this water-yield assessment model was tested using the unit yield values ​​from 15 "bottom-water" pumping (injection) test boreholes within the study area. The distribution of pumping test boreholes corresponding to each water-rich zone is shown in Table 1. There are 5 unit water yield boreholes in the medium water-rich zone, among which the unit water yield of 62-4, 10-Shui1, 11-Shui2, and 66-68-4 ranges from 0.13690 to 0.30330 L / (s·m). According to the relevant standards for water-richness classification, it should be classified as medium water-rich, which is consistent with the results of this judgment. Only the 06-Guan2 hole (q is 0.01111 L / (s·m)) is incorrectly judged. There are 6 unit water yield boreholes in the weak water-rich zone, among which the unit water yield of 06-Guan1, 10-Shui2, 2019-Shui1, 11-Shui1, and 2013-Shui1 ranges from 0.00752 to 0.03317 The q value is significantly lower than that in the moderately water-rich area. Only the 70-71-3 well (q is 0.33580 L / (s·m)) is misjudged. There are two unit water yield boreholes in the weakly water-rich area, namely Jian 2 and 2015-Guan 2, with unit water yields ranging from 0.00349 to 0.00600 L / (s·m), and their q values ​​are both lower than those in the weakly water-rich area. There are two unit water yield boreholes in the very weakly water-rich area, namely 2019-Shui 2 and 74-75-6, with unit water yields ranging from 0.00062 to 0.00253 L / (s·m), and their q values ​​are both lower than those in the weakly water-rich area.

[0118] Table 1 Evaluation zoning results and pumping test borehole distribution discrimination

[0119] Verification of drilling data from pumping (injection) tests revealed quantitative differences in the unit water yield values ​​of the pumping (injection) boreholes within the moderately water-rich, weakly water-rich, slightly water-rich, and extremely weakly water-rich zones, respectively, as defined by this method (only holes 70-71-3 and 06-282 were incorrectly identified). The q value for the moderately water-rich zone was >0.1 L / (s·m), while the q values ​​for the weakly water-rich, slightly water-rich, and extremely weakly water-rich zones were all <0.1 L / (s·m). Furthermore, the q value for the extremely weakly water-rich zone was < that for the slightly water-rich zone, and < that for the weakly water-rich zone. The evaluation model's results for relative water-richness were highly consistent with the water-richness grades determined based on unit water yield, and further detailed classification of weak water-rich grades was provided. The water-richness evaluation results were consistent with the aforementioned "bottom-containing" sedimentary facies distribution and the laws governing sedimentary water control. The entire process avoided the influence of human factors, resulting in relatively ideal model evaluation results.

[0120] Taking the "bottom content" of Xutuan Mine in Huaibei Mining Area as an example, the present invention was applied to the water-richness evaluation of the "bottom content" of Xutuan Mine, and the water-richness grades and zones of the "bottom content" were finely divided. The key target areas for the prevention and control of water hazards in the "bottom content" of Xutuan Mine were identified, ensuring the safe recovery of 32 coal seams affected by water hazards in the "bottom content" of Xutuan Mine, and effectively reducing the cost of preventing and controlling water hazards in the mine.

[0121] The following are two specific embodiments and implementation schemes based on the above-mentioned evaluation method for the refined zoning of water-richness of loose aquifers in mines:

[0122] Example 1:

[0123] Evaluation object: Aquifer of a mine

[0124] Step 1: Establish an evaluation indicator matrix

[0125] Aquifer Thickness: Measures the actual thickness of the aquifer in meters.

[0126] Gravel layer thickness: measures the actual thickness of the gravel layer (unit: meter).

[0127] Lithology-thickness index: A parameter used to describe lithology, which is related to the thickness of the aquifer.

[0128] Sand-to-ground ratio: the ratio of the thickness of the gravel layer to the total thickness of the aquifer.

[0129] Number of gravel layers: Count the number of gravel layers.

[0130] Step 2: Normalization

[0131] Each evaluation indicator is normalized to ensure that the indicators with larger values ​​are better and the indicators with smaller values ​​are better are on the same scale.

[0132] Step 3: Establish a water-richness evaluation model

[0133] Using DPS software, based on the principle of projection pursuit model, an aquifer water-richness evaluation model was established.

[0134] Perform linear projection in the model and construct the projection objective function.

[0135] Step 4: Optimize the projection objective function

[0136] By optimizing the projection objective function, the optimal projection direction is determined to reflect the information of each evaluation index to the greatest extent.

[0137] Step 5: Calculate the projected eigenvalues

[0138] Bring the optimal projection direction back into the expression of linear projection and calculate the projection eigenvalue that can reflect the information of each evaluation index.

[0139] Step 6: Comprehensive Evaluation

[0140] Based on the distribution and magnitude of the projected eigenvalues, a comprehensive evaluation of the mine's aquifer is performed. The evaluation results include a water-richness grade or score, helping mine managers better understand the aquifer's hydrological characteristics.

[0141] Example 2:

[0142] Evaluation object: Aquifer of another mine

[0143] Step 1: Establish an evaluation indicator matrix

[0144] The same evaluation indicators as those in Example 1 were used, including aquifer thickness, gravel layer thickness, lithology-thickness index, sand-to-ground ratio, and number of gravel layers.

[0145] Step 2: Normalization

[0146] Each evaluation indicator is normalized to ensure that the indicators with larger values ​​are better and the indicators with smaller values ​​are better are on the same scale.

[0147] Step 3: Establish a water-richness evaluation model

[0148] Using DPS software, based on the principle of projection pursuit model, an aquifer water-richness evaluation model was established.

[0149] Perform linear projection in the model and construct the projection objective function.

[0150] Step 4: Optimize the projection objective function

[0151] By optimizing the projection objective function, the optimal projection direction is determined to reflect the information of each evaluation index to the greatest extent.

[0152] Step 5: Calculate the projected eigenvalues

[0153] Bring the optimal projection direction back into the expression of linear projection and calculate the projection eigenvalue that can reflect the information of each evaluation index.

[0154] Step 6: Comprehensive Evaluation

[0155] Based on the distribution characteristics and size of the projection eigenvalues, a comprehensive evaluation of the aquifer in the mine was conducted to provide information on its water richness.

[0156] These two examples demonstrate how to use the same method to assess water richness in different mines and aquifers, but adjust the evaluation indicators and parameters based on actual conditions. This approach helps to more accurately understand the hydrological conditions of unconsolidated aquifers.

[0157] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications, equivalent substitutions and improvements made by any technician familiar with this technical field within the technical scope disclosed by the present invention and within the spirit and principles of the present invention should be covered by the scope of protection of the present invention.

Claims

1. A method for evaluating the water-richness of loose aquifers in mines by fine zoning, characterized in that: A standardized evaluation matrix was established by combining multiple evaluation indicators, and the water-richness evaluation model was constructed through the DPS software using the principle of projection pursuit model. The optimal projection direction was determined by optimizing linear projection and projection objective function, thereby accurately reflecting the projection eigenvalues ​​of each evaluation indicator information, providing a more accurate and systematic method for comprehensive evaluation. 2..The method for evaluating the fine zoning of water-richness of loose aquifers in mines as claimed in claim 1 is characterized in that: The following steps are involved: S101, the evaluation index matrix is ​​established by combining the evaluation indexes of aquifer thickness, gravel layer thickness, lithology-thickness index, sand-to-ground ratio, and number of gravel layers; S102, unifying the variation range of different evaluation index data, normalizing the evaluation indexes with larger values ​​and the indexes with smaller values, obtaining a standardized evaluation matrix after normalization, and constructing a "bottom-content" water-richness evaluation index system; S103, combining the constructed bottom water-bearing richness evaluation index system and according to the principle of projection pursuit model, an aquifer water-bearing richness evaluation model is established by using DPS software; S104, performing linear projection and constructing a projection objective function in an aquifer water-richness evaluation model; S105, optimizing the projection objective function to determine the best projection direction; S106, bringing the optimal projection direction back into the expression of linear projection to obtain projection characteristic values ​​that can reflect information of each evaluation index; S107, comprehensively evaluating the evaluation object according to the distribution characteristics and size of the projection eigenvalues.

3. The method for evaluating the water-richness of loose aquifers in mines according to claim 2 is characterized in that: The thickness of the aquifer, the thickness of the gravel layer, the lithology structure index, and the sand-to-ground ratio in S101 are all positive indicators that affect the water-richness of the aquifer, and the number of gravel layers is a negative indicator that affects the water-richness.

4. The method for evaluating the water-richness of loose aquifers in mines according to claim 2, characterized in that: In S105, the projection objective function is optimized by using a composite method to optimize and solve the projection objective function.

5. The method for evaluating the water-richness of loose aquifers in mines according to claim 2, characterized in that: In S106, the absolute value of each component in the optimal projection direction essentially reflects the contribution of each evaluation index to the bottom water-richness evaluation result. The larger the absolute value of each component, the greater the influence of the corresponding evaluation index on the bottom water-richness, and vice versa.

6. The method for evaluating the water-richness of loose aquifers in mines according to claim 2, characterized in that: The projection characteristic values ​​in S107 can be visualized to obtain a contour map of the bottom water-rich projection characteristic values.

7. The method for evaluating the water-richness of loose aquifers in mines according to claim 2, characterized in that: The size of the projection characteristic value in S107 reflects the water-richness of the bottom content, that is, the larger the projection characteristic value, the stronger the water-richness of the bottom content.

8. An evaluation system for fine zoning of water-richness of loose aquifers in mines, which implements the evaluation method for fine zoning of water-richness of loose aquifers in mines as claimed in any one of claims 1 to 7, characterized in that: The evaluation system for fine zoning of water-richness of loose aquifers in mines includes: An evaluation index module, used to determine the evaluation index; An evaluation index system establishment module, connected with the evaluation index module, is used to establish an evaluation index system; A projection pursuit module, used for providing a projection pursuit model; The aquifer water-richness evaluation module is connected with the evaluation index system establishment module and the projection pursuit module, and is used to establish an aquifer water-richness evaluation model by combining the data of the evaluation index system establishment module and the model of the projection pursuit module through the DPS software.

9. A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, the processor executes the steps of the method for evaluating the fine zoning of water-richness of loose aquifers in mines as described in any one of claims 1 to 7.

10. An information data processing terminal, characterized in that: The information data processing terminal is used to implement the evaluation system for refined zoning of water-richness of loose aquifers in mines as described in claim 8.

Images