Sarn type polymetallic deposit prospecting exploration method and system based on three-dimensional modeling

Abstract

The invention provides a skarn type polymetallic deposit prospecting exploration method and system based on three-dimensional modeling, and relates to the field of geological prospecting, and the method comprises the steps: obtaining deposit data of an area where a deposit is located, and carrying out the preprocessing of the data of the mined skarn type polymetallic deposit; selecting a multi-scale convolutional neural network as a deep learning model for training so as to predict the position where an ore deposit possibly exists in the unmined area; generating a geologic feature coefficient according to the predicted geologic features of the unmined area; generating a mineral characteristic coefficient through the mineral characteristic data of the three-dimensional model; comparing the mineral characteristic coefficient with a preset weight, and judging whether tungsten-tin ore exists or not; generating a mineral type judgment coefficient according to the content of mineral elements in the sample; generating a comprehensive prediction value through the geologic characteristic coefficient and the mineral characteristic coefficient, and judging the type of the ore deposit according to the comprehensive prediction value and the type judgment coefficient. The method improves the accuracy and efficiency of ore deposit prediction.

Description

Technical Field

[0001] The present invention relates to the field of geological prospecting technology, and in particular to a skarn-type polymetallic ore deposit prospecting method and system based on three-dimensional modeling. Background Art

[0002] For decades, geological surveys have been a key pillar of energy development. However, traditional mineral exploration methods no longer meet the requirements for efficiency and accuracy, especially for complex skarn-type polymetallic deposits. Traditional methods face numerous challenges in data processing, accuracy, and resource investment. Existing technologies typically rely on two-dimensional geological maps for exploration, which limits a comprehensive understanding of the spatial structure of the deposit and results inaccurate.

[0003] In the prior art, publication number CN118363087B discloses a skarn-type iron-copper-gold polymetallic mineral exploration method and system, which constructs a reference model based on the prior information of the target ore deposit; performs three-dimensional physical property inversion of gravity and magnetism with the reference model as a constraint condition, obtains the correspondence between lithology and physical properties and performs lithology mapping, establishes a two-dimensional to three-dimensional model, calculates the connectivity of the prospecting units from the three-dimensional model, and finds the location of the ore target area based on the connectivity. However, this limits the comprehensive understanding of the spatial structure of the ore deposit, resulting in insufficient accuracy of the exploration results. In addition, in data processing, advanced data processing technologies and models are not fully utilized, and the distribution of ore deposits in unmined areas cannot be effectively predicted. Therefore, there is a need for an advanced exploration method based on three-dimensional modeling that can more accurately process and analyze ore deposit data, improve the accuracy and efficiency of ore deposit prediction, and thus better guide the rational development and utilization of mineral resources.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a skarn-type polymetallic ore prospecting and exploration method and system based on three-dimensional modeling to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A skarn-type polymetallic mineral deposit prospecting method based on three-dimensional modeling, comprising the following steps:

[0008] Step 1: Obtain ore deposit data for the area where tungsten-tin ore and skarn-type polymetallic deposits are located. The ore deposit data includes geological characteristic data and mineral characteristic data. The data is divided into mined areas and unmined areas. A three-dimensional geological model is constructed to determine the sampling area. The geological characteristic coefficients are generated based on the content data of tungsten ore, tin ore, and skarn in the mined area.

[0009] Step 2: Build a deep learning model, use the geological characteristic data of the mined area as the training set, and the geological characteristic coefficients of the mined area as labels to train the deep learning model; input the geological characteristic data of the unmined area into the trained deep learning model to obtain the predicted geological characteristic coefficients of the unmined area;

[0010] Step 3: Calculate the skewness and kurtosis values ​​based on the mineral characteristic data in the ore deposit data, generate the mineral characteristic coefficient, and compare the mineral characteristic coefficient with the preset weight to determine whether tungsten-tin ore exists;

[0011] Step 4: Generate a mineral type judgment coefficient based on the content of silicon, aluminum, sulfur, and iron in the mineral characteristic data; generate a comprehensive prediction value through the geological characteristic coefficient and the mineral characteristic coefficient, and judge the type of the ore deposit based on the comprehensive prediction value and the type judgment coefficient.

[0012] Furthermore, the skarn-type polymetallic mineral deposit prospecting and exploration method based on three-dimensional modeling is characterized in that the geological characteristic data includes: lithological characteristic data, stratigraphic characteristic data, and the mineral characteristic data includes: data on the content of tungsten, tin, antimony, molybdenum, silicon, aluminum, sulfur, and iron in soil and core samples.

[0013] Furthermore, the skarn-type polymetallic mineral deposit prospecting method based on three-dimensional modeling is characterized in that the logic of setting the data set and label of the mined area data is:

[0014] The entire target area is divided into several sub-areas, each sub-area is represented by a sample, and each sample contains all the geological characteristic data of the sub-area. All geological characteristic data are set as a dataset, including lithologic characteristic data and stratigraphic characteristic data; the dataset is divided into a training set and a validation set with a ratio of 7:3.

[0015] Furthermore, the skarn-type polymetallic ore prospecting method based on three-dimensional modeling is characterized in that the calculation formula of the geological characteristic coefficient is:

[0016]

[0017] Among them, φ is the geological characteristic coefficient, f iis the mineral content in the i-th mined area, including the content data of tungsten ore, tin ore and skarn; i=1 is tungsten ore, i=2 is tin ore, i=3 is skarn, is the preset weight corresponding to the i-th mineral, and are greater than 0, and

[0018] Furthermore, the formula for calculating the skewness and kurtosis values ​​based on the data of tungsten, tin, antimony, and molybdenum content in soil and core samples is:

[0019]

[0020]

[0021] Where β is the skewness value of the element, γ is the kurtosis value of the element, j is the index number of the sample, j = 1, 2, ..., n, where n is the total number of samples, x j is the element content data of the jth sample, μ is the mean, and σ is the standard deviation.

[0022] Furthermore, the calculation formula for the generated mineral characteristic coefficient is:

[0023] ψ=log(α(β W +β Sn )+ρ(β Sb +β Mo )+γ W +γ Sn +γ Sb +γ Mo )

[0024] Among them, ψ is the mineral characteristic coefficient, α and ρ are the weighting coefficients of tungsten and tin and antimony and molybdenum, respectively, where α and ρ are both greater than 0, and α>ρ.

[0025] Furthermore, the calculation formula for the mineral type determination coefficient is:

[0026]

[0027] Among them, [Si], [Al], [S], and [Fe] are the contents of silicon, aluminum, sulfur, and iron in the unmined area, respectively.

[0028] Furthermore, the comprehensive prediction value calculation formula is:

[0029] Ω=φ·ψ 2 +ω1·ψ+ω2·φ

[0030] Among them, Ω is the comprehensive prediction value, ω1 and ω2 are the preset model adjustment parameters, where ω1 and ω2 are both greater than 0, and ω1>Ω2.

[0031] Furthermore, based on the comprehensive prediction value and type judgment coefficient, the logic for judging the type of mineral deposit is as follows:

[0032] According to the preliminary judgment of the comprehensive prediction value, the preset threshold Ω low and threshold Ω high , if Ω<Ω low , then the deposit is low-grade or has no economic value; if Ω low ≤Ω≤Ω high , the deposit is low-grade or has no economic value; if Ω>Ω high , the deposit is low-grade or has no economic value;

[0033] According to the category judgment coefficient, the category is refined and the threshold k is preset. low and threshold k high , if k <k low , it is judged to be a quartz vein type tungsten-tin ore; if k low <k<k high , it is judged to be a silicate-type tungsten-tin ore; if k>k high , it is judged to be sulfide-type tungsten-tin ore.

[0034] The present invention also provides a skarn-type polymetallic deposit prospecting and exploration method and system based on three-dimensional modeling. The distribution method is obtained by executing the above-mentioned skarn-type polymetallic deposit prospecting and exploration method. The specific modules include:

[0035] Acquisition module: Acquires ore deposit data for areas where tungsten-tin ore and skarn-type polymetallic deposits are located. The ore deposit data includes geological and mineral characteristics data. The data is divided into mined and unmined areas, and a three-dimensional geological model is constructed to determine the sampling area. The geological characteristic coefficient is generated based on the content data of tungsten ore, tin ore, and skarn in the mined area.

[0036] Training module: Build a deep learning model, use the geological characteristic data of the mined area as the training set, and the geological characteristic coefficients of the mined area as labels to train the deep learning model; input the geological characteristic data of the unmined area into the trained deep learning model to obtain the predicted geological characteristic coefficients of the unmined area;

[0037] Mineral detection module: Calculate the skewness and kurtosis values ​​based on the mineral characteristic data in the ore deposit data, generate the mineral characteristic coefficient, compare the mineral characteristic coefficient with the preset weight, and determine whether there is tungsten-tin ore;

[0038] Mineral type judgment module: Generates mineral type judgment coefficients based on the content of silicon, aluminum, sulfur, and iron in the mineral characteristic data; generates a comprehensive prediction value through the geological characteristic coefficient and the mineral characteristic coefficient, and judges the type of the ore deposit based on the comprehensive prediction value and the type judgment coefficient.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention significantly improves the exploration accuracy and efficiency of tungsten-tin ore and skarn-type polymetallic deposits by introducing deep learning technology based on three-dimensional modeling. Specifically, the method first systematically processes the deposit data and carefully labels the data of mined and unmined areas, providing a solid data foundation for subsequent analysis. Through deep learning model training of multi-scale convolutional neural networks, the system can accurately predict the distribution of deposits in unmined areas, making full use of the geological characteristics and mineral characteristic information in historical data, greatly improving the efficiency and accuracy of resource exploration. In addition, the generated geological characteristic coefficients are compared with the mineral characteristic coefficients to further enhance the ability to judge the type of ore deposit, effectively reducing the risk of misjudgment caused by insufficient information in traditional exploration methods. Compared with the existing technology, this solution provides a more forward-looking solution for the rational development of mineral resources, which not only improves the reliability of exploration results, but also promotes technological progress in the mining field. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 Schematic diagram of the overall method flow of the present invention;

[0042] Figure 2 Schematic diagram of the overall system structure of the present invention. DETAILED DESCRIPTION

[0043] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to specific embodiments.

[0044] It should be noted that, unless otherwise defined, the technical or scientific terms used in the present invention should have the usual meanings understood by people with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0045] Example:

[0046] See also Figure 1, the present invention provides a technical solution:

[0047] A skarn-type polymetallic mineral deposit prospecting method based on three-dimensional modeling, comprising the following steps:

[0048] Step 1: Obtain ore deposit data for the area where tungsten-tin ore and skarn-type polymetallic deposits are located. The ore deposit data includes geological characteristic data and mineral characteristic data. The data is divided into mined areas and unmined areas. A three-dimensional geological model is constructed to determine the sampling area. The geological characteristic coefficients are generated based on the content data of tungsten ore, tin ore, and skarn in the mined area.

[0049] Using drones and automated sampling equipment, high-precision data collection is conducted within the target area. The collected data includes: lithologic data (e.g., distribution of volcanic, sedimentary, and metamorphic rock types); stratigraphic data (e.g., distribution and sequence of different strata); and data on the concentrations of tungsten, tin, antimony, molybdenum, silicon, aluminum, sulfur, and iron in soil and core samples.

[0050] Import the collected data into GoCAD to construct a 3D geological model of the target area. The specific locations of existing tungsten-tin and skarn-type polymetallic deposits were marked. The entire target area was divided into several sub-areas, each represented by a sample containing all characteristic data for that sub-area. All characteristic data was set as a dataset, including geological characteristic coefficients: lithologic characteristic data and stratigraphic characteristic data. The preprocessed dataset was divided into a training set and a validation set in a ratio of 7:3. Based on known mineral exploration data, each sub-area was labeled with the type of tungsten ore, tin ore, and skarn-type polymetallic deposit.

[0051] Step 2: Build a deep learning model, use the geological characteristic data of the mined area as the training set, and the geological characteristic coefficients of the mined area as labels to train the deep learning model; input the geological characteristic data of the unmined area into the trained deep learning model to obtain the predicted geological characteristic coefficients of the unmined area;

[0052] A multi-scale convolutional neural network was selected as the deep learning model. This model can process geological features at different scales and improve the ability to identify complex geological structures. A time series analysis module was introduced into the multi-scale convolutional neural network to analyze the temporal characteristics of geological evolution and enhance the timeliness and robustness of the model.

[0053] The lithologic and stratigraphic data for each region in the training set is input. An adaptive optimization algorithm is used to dynamically adjust the model's weights and parameters, and a backpropagation algorithm is used to minimize the error between the output and the sample labels. The trained model is evaluated on the validation set until the accuracy and precision of the evaluation results meet the set requirements. The validated deep learning model is then output.

[0054] Step 3: Calculate the skewness and kurtosis values ​​based on the mineral characteristic data in the ore deposit data, generate the mineral characteristic coefficient, and compare the mineral characteristic coefficient with the preset weight to determine whether tungsten-tin ore exists;

[0055] The calculation formula of the geological characteristic coefficient is:

[0056]

[0057] Among them, φ is the geological characteristic coefficient, f i is the mineral content in the i-th mined area, including the content data of tungsten ore, tin ore and skarn; i=1 is tungsten ore, i=2 is tin ore, i=3 is skarn, is the preset weight corresponding to the i-th mineral, and are greater than 0, and Weight Corresponding to the influence of tungsten ore content, weight Corresponding to the impact of tin ore content, weight The relative size of the weights corresponding to the influence of skarn content determines the relative importance of these three factors in the priority calculation. and The setting ensures that the tungsten content has a greater impact on the priority, because the tungsten content is usually a more critical factor. Even if the tin content is sufficient during the mineral exploration process, if the tungsten content is too low, its priority may not be very high. Although the tungsten content is more critical, tin cannot be ignored. A large enough tin content can improve the efficiency of geological identification, so it also has a certain weight, but it is relatively small compared to the tungsten content. Skarn content is an important factor in identifying mineral deposit geology, but compared with tungsten and tin, its economic value is lower, so the weight given is set to be smaller. The value of can be set to 0.1-0.2, indicating that the importance of skarn content is small but not negligible. The value of can be set to 0.3-0.2, indicating that the importance of tin ore content is limited but not negligible. The setting of 0.4-0.5 indicates that the tungsten content is more important. This range is selected based on experience and actual operational needs to ensure that in most cases, the formula can reflect the reasonable priority ranking.

[0058] The formula for calculating the skewness and kurtosis values ​​based on the data of tungsten, tin, antimony and molybdenum content in soil and core samples is:

[0059]

[0060]

[0061] Where β is the skewness value of the element, γ is the kurtosis value of the element, j is the index number of the sample, j = 1, 2, ..., n, where n is the total number of samples, x j is the element content data for the jth sample, μ is the mean, σ is the standard deviation, and the skewness of an element reflects the symmetry of the distribution. A skewness of 0 indicates a symmetrical distribution; a positive skewness indicates a right-skewed distribution; a negative skewness indicates a left-skewed distribution. The cubic function is used because it amplifies the influence of samples above the mean and samples below the mean. The kurtosis of an element reflects the kurtosis of the distribution. A high kurtosis indicates a high concentration of data around the mean, while a low kurtosis indicates a flatter distribution. The quartic function is used because it amplifies the influence of extreme values, effectively describing the thickness of the tail. To normalize the kurtosis of the element, the kurtosis of the normal distribution is set to 0. The kurtosis of the normal distribution is 3, so 3 is subtracted.

[0062] The calculation formula for the generated mineral characteristic coefficient is:

[0063] ψ=log(α(β W +β Sn )+ρ(β Sb +β Mo )+γ W +γ Sn +γ Sb +γ Mo )

[0064] Among them, ψ is the mineral characteristic coefficient. The calculation of ψ takes into account the skewness and kurtosis values ​​of the four mineral elements tungsten, tin, antimony, and molybdenum, reflecting the comprehensive characteristics of the deposit, especially the dominant role of tungsten and tin. The mineral characteristic coefficient can provide a quantitative basis for deposit evaluation, help geological prospectors formulate more effective exploration strategies, and improve the efficiency and economic benefits of resource development. α and ρ are the weighting coefficients for tungsten and tin and antimony and molybdenum, respectively, where α and ρ are both greater than 0, and α>ρ. The weight α corresponds to the influence of the skewness values ​​of tungsten and tin elements, and the weight ρ corresponds to the influence of the skewness values ​​of antimony and molybdenum elements. The relative size of the weights determines the relative importance of these two factors in the priority calculation. The setting of α and ρ ensures that the skewness values ​​of tungsten and tin elements have a greater impact on the priority, because the skewness values ​​of tungsten and tin elements are usually more critical factors. During the mineral deposit exploration process, even if the skewness values ​​of antimony and molybdenum elements are sufficient, if the skewness values ​​of tungsten and tin elements are too small, their priority may not be high. Although the skewness values ​​of tungsten and tin elements are more critical, the skewness values ​​of antimony and molybdenum elements cannot be ignored. Large enough skewness values ​​for antimony and molybdenum can improve the efficiency of mineral identification. Therefore, they are also given a certain weight, albeit a relatively small one. A value of ρ can be set between 1.0 and 2.0, indicating that the skewness values ​​of antimony and molybdenum are of limited but not negligible importance, while a setting of α between 2.0 and 3.0 indicates that the skewness values ​​of tungsten and tin are more important. This range is selected based on experience and practical operational needs to ensure that the formula reflects the appropriate priority in most cases.

[0065] Step 4: Compare the mineral characteristic coefficient with the preset weight to determine whether tungsten-tin ore exists, and generate a mineral type judgment coefficient based on the silicon, aluminum, sulfur, and iron contents in the sample; generate a comprehensive prediction value through the geological characteristic coefficient and the mineral characteristic coefficient, and judge the type of the ore deposit based on the comprehensive prediction value and the type judgment coefficient.

[0066] The mineral characteristic coefficient ψ is compared with the preset weight threshold n. According to experience and geological characteristics, the value of n can be set to 2.0-3.0. If ψ<2.4, it is predicted that there is no tungsten-tin ore in the area, otherwise it is judged that there is tungsten-tin ore.

[0067] The calculation formula for the mineral type judgment coefficient is:

[0068]

[0069] Among them, [Si], [Al], [S], and [Fe] are the contents of silicon, aluminum, sulfur, and iron in the unmined area, respectively.

[0070] Elements such as silicon and aluminum are commonly associated with silicate minerals in skarn-type deposits. The silicon and aluminum content can help identify and differentiate between different types of silicate minerals, such as feldspar and mica, which are important components of skarns.

[0071] Sulfur and iron are the main components of many sulfide minerals. Iron often forms pyrite and pyrrhotite with sulfur. These sulfide minerals are also very common in skarn-type deposits.

[0072] The comprehensive prediction value calculation formula is:

[0073] Ω=φ·ψ 2 +ψ1·ψ+ω2·φ

[0074] Among them, Ω is a comprehensive prediction value, which is used to reflect the potential value and enrichment of the deposit. It is obtained by combining the mineral characteristic coefficient and the geological characteristic coefficient. The purpose is to evaluate the economic value or development potential of the deposit. It can help geological prospectors to select exploration areas more scientifically, thereby increasing the discovery rate of mineral resources and saving time and costs. ψ 2 Emphasizing the quadratic effect of ψ may increase the influence of elemental content on the comprehensive prediction. ω1·ψ and ω2·φ have a linear effect on the comprehensive prediction value. When φ or ψ increases, the comprehensive prediction value Ω generally also increases. This indicates that improvements in the deposit's characteristic quantities and mineralogy coefficients are positively correlated with the comprehensive prediction value, reflecting enhanced deposit potential. ω1 and ω2 are preset model adjustment parameters, where both ω1 and ω2 are greater than 0, and ω1>ω2. The weight ω1 corresponds to the influence of the mineralogy coefficient, while the weight ω2 corresponds to the influence of the squared geological coefficient. The relative size of the weights determines the relative importance of these two factors in the priority calculation. The setting of ω1 and ω2 ensures that the mineralogy coefficient has a greater influence on the priority, as the mineralogy coefficient is generally the more critical factor. During mineral exploration, even if the geological coefficient is sufficient, if the mineralogy coefficient is too small, its priority may not be high. Although the mineralogy coefficient is more critical, the geological coefficient cannot be ignored. A sufficiently large geological coefficient can improve exploration efficiency. Therefore, there are also certain weights, but they are relatively small. The value of ω2 can be set to 0.5-3.0, indicating that the importance of geological characteristics is limited but not negligible, while the value of ω1 is set to 5.0-10.0, indicating that the importance of mineral characteristics is higher. This range is selected based on experience and actual operational needs to ensure that the formula reflects reasonable priority in most cases.

[0075] Based on the comprehensive prediction value and type judgment coefficient, the logic for judging the type of mineral deposit is as follows:

[0076] According to the preliminary judgment of the comprehensive prediction value, the preset threshold Ω low and threshold Ω high , if Ω<Ω low , then the deposit is low-grade or has no economic value; if Ω low ≤Ω≤Ω high , the deposit is low-grade or has no economic value; if Ω>Ω high, the deposit is low-grade or has no economic value;

[0077] According to the category judgment coefficient, the category is refined and the threshold k is preset. low and threshold k high , if k <k low , it is judged to be a quartz vein type tungsten-tin ore; if k low <k<k high , it is judged to be a silicate-type tungsten-tin ore; if k>k high , it is judged to be sulfide-type tungsten-tin ore.

[0078]

[0079] See also Figure 2 The present invention also provides a skarn-type polymetallic deposit prospecting and exploration system based on three-dimensional modeling. The distribution method is obtained by executing the above-mentioned skarn-type polymetallic deposit prospecting and exploration method. The specific modules include:

[0080] Acquisition module: Acquires ore deposit data for areas where tungsten-tin ore and skarn-type polymetallic deposits are located. The ore deposit data includes geological and mineral characteristics data. The data is divided into mined and unmined areas, and a three-dimensional geological model is constructed to determine the sampling area. The geological characteristic coefficient is generated based on the content data of tungsten ore, tin ore, and skarn in the mined area.

[0081] Training module: Build a deep learning model, use the geological characteristic data of the mined area as the training set, and the geological characteristic coefficients of the mined area as labels to train the deep learning model; input the geological characteristic data of the unmined area into the trained deep learning model to obtain the predicted geological characteristic coefficients of the unmined area;

[0082] Mineral detection module: Calculate the skewness and kurtosis values ​​based on the mineral characteristic data in the ore deposit data, generate the mineral characteristic coefficient, compare the mineral characteristic coefficient with the preset weight, and determine whether there is tungsten-tin ore;

[0083] Mineral type judgment module: Generates mineral type judgment coefficients based on the content of silicon, aluminum, sulfur, and iron in the mineral characteristic data; generates a comprehensive prediction value through the geological characteristic coefficient and the mineral characteristic coefficient, and judges the type of the ore deposit based on the comprehensive prediction value and the type judgment coefficient.

[0084] The above formulas are all dimensionless and numerical calculations. The formulas are obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0085] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0086] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, and may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment as needed.

[0087] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. A skarn-type polymetallic ore deposit prospecting method based on three-dimensional modeling, characterized in that: The specific steps include: Step 1: Obtain the deposit data of the area where the tungsten-tin ore and skarn-type polymetallic deposits are located. The deposit data includes geological characteristic data and mineral characteristic data. The data is divided into mined areas and unmined areas, a three-dimensional geological model is constructed, the sampling area is determined, and the geological characteristic coefficient is generated according to the content data of tungsten ore, tin ore and skarn in the mined area; Step 2: Build a deep learning model, use the geological characteristic data of the mined area as the training set, and the geological characteristic coefficients of the mined area as labels to train the deep learning model; input the geological characteristic data of the unmined area into the trained deep learning model to obtain the predicted geological characteristic coefficients of the unmined area; Step 3: According to the mineral characteristic data in the ore deposit data, calculate the skewness value and the kurtosis value, generate the mineral characteristic coefficient, compare the mineral characteristic coefficient with the preset weight, and determine whether there is tungsten-tin ore; Step 4: Generate mineral type judgment coefficients based on the contents of silicon, aluminum, sulfur, and iron in the mineral characteristic data; generate comprehensive prediction values ​​through geological characteristic coefficients and mineral characteristic coefficients, and judge the type of the ore deposit based on the comprehensive prediction values ​​and type judgment coefficients.

2. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: The geological characteristic data include: lithological characteristic data, stratigraphic characteristic data, and mineral characteristic data include: data on the content of tungsten, tin, antimony, molybdenum, silicon, aluminum, sulfur, and iron in soil and core samples.

3. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: The logic for setting the data set and label for the mined area data is: The entire target area is divided into several sub-areas, each sub-area is represented by a sample, each sample contains all geological characteristic data of the sub-area, and all geological characteristic data are set as a data set, including lithology characteristic data and stratigraphic characteristic data; The dataset is divided into a training set and a validation set with a ratio of 7:

3.

4. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: The calculation formula of the geological characteristic coefficient is: Among them, φ is the geological characteristic coefficient, f i is the mineral content in the i-th mined area, including the content data of tungsten ore, tin ore and skarn; i=1 is tungsten ore, i=2 is tin ore, i=3 is skarn, is the preset weight corresponding to the i-th mineral, and are greater than 0, and 5. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: The formula for calculating the skewness and kurtosis values ​​based on the data of tungsten, tin, antimony and molybdenum content in soil and core samples is: Where β is the skewness value of the element, γ is the kurtosis value of the element, j is the index number of the sample, j = 1, 2, ..., n, where n is the total number of samples, x j is the element content data of the jth sample, μ is the mean, and σ is the standard deviation.

6. A skarn-type polymetallic ore deposit prospecting and exploration method based on three-dimensional modeling according to claim 5, characterized in that: The calculation formula of the generated mineral characteristic coefficient is: ψ=log(α(β W +b Sn )+ρ(β Sb +b Mo )+c W +g Sn +g Sb +g Mo ) Among them, ψ is the mineral characteristic coefficient, α and ρ are the weighting coefficients of tungsten and tin and antimony and molybdenum, respectively, where α and ρ are both greater than 0, and α>ρ.

7. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: The calculation formula of the mineral type determination coefficient is: Among them, [Si], [Al], [S], and [Fe] are the contents of silicon, aluminum, sulfur, and iron in the unmined areas, respectively.

8. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: The comprehensive prediction value calculation formula is: Ω=φ·ψ 2 +ω1·ψ+ω2·φ Among them, Ω is the comprehensive prediction value, ω1 and ω2 are the preset model adjustment parameters, where ω1 and ω2 are both greater than 0, and ω1>ω2.

9. The method for prospecting and exploration of skarn-type polymetallic deposits based on three-dimensional modeling according to claim 1, characterized in that: According to the comprehensive prediction value and type judgment coefficient, the logic for judging the type of mineral deposit is as follows: According to the preliminary judgment of the comprehensive prediction value, the preset threshold Ω low and threshold Ω high , if Ω<Ω low , then the deposit is low-grade or has no economic value; if Ω iow ≤Ω≤Ω high , the deposit is low-grade or has no economic value; if Ω>Ω high , then the deposit is low-grade or has no economic value; According to the category judgment coefficient, the category is refined and the threshold k is preset low and threshold k high , if k <k low , it is judged to be a quartz vein type tungsten-tin ore; if k low <k<k high , it is judged to be a silicate tungsten-tin ore; if k>k high , it is judged to be a sulfide-type tungsten-tin ore.

10. A skarn-type polymetallic ore deposit prospecting and exploration system based on three-dimensional modeling, characterized in that: The exploration system is used to execute the skarn-type polymetallic ore deposit prospecting method according to any one of claims 1 to 9, comprising: The acquisition module is used to obtain the deposit data of the area where the tungsten-tin ore and skarn-type polymetallic deposits are located. The deposit data includes geological characteristic data and mineral characteristic data. The data is divided into mined areas and unmined areas, a three-dimensional geological model is constructed, the sampling area is determined, and the geological characteristic coefficient is generated according to the content data of tungsten ore, tin ore and skarn in the mined area; The training module is used to construct a deep learning model, and the geological characteristic data of the mined area is used as a training set, and the geological characteristic coefficients of the mined area are used as labels to train the deep learning model; the geological characteristic data of the unmined area is input into the trained deep learning model to obtain the predicted geological characteristic coefficients of the unmined area; The mineral detection module is used to calculate the skewness value and the kurtosis value according to the mineral characteristic data in the ore deposit data, generate the mineral characteristic coefficient, compare the mineral characteristic coefficient with the preset weight, and determine whether there is tungsten-tin ore; The mineral type judgment module is used to generate mineral type judgment coefficients based on the content of silicon, aluminum, sulfur and iron in the mineral characteristic data; generate comprehensive prediction values ​​through geological characteristic coefficients and mineral characteristic coefficients, and judge the type of ore deposit based on the comprehensive prediction values ​​and type judgment coefficients.

Images