Metallization prediction method and system based on geological big data

Abstract

The invention relates to the technical field of mineralization prediction methods, in particular to a mineralization prediction method and system based on geological big data, and the method comprises the steps: obtaining multi-source geological data, carrying out the rasterization of the multi-source geological data, obtaining a plurality of types of data channels, and calculating the spatial change rate vector of each position in the data channels; for each position, dividing the local area into two sub-areas along a plurality of test directions, calculating the vector sum of all space change rate vector fields in each sub-area, calculating the direction saliency of the test directions by using the vector sum of the two sub-areas, and taking the test direction corresponding to the maximum direction saliency as the construction direction of the position; and performing direction enhancement on the data channel to obtain a direction enhanced data channel, and calculating a mineralization potential index of each position based on the direction enhanced data channel so as to evaluate the mineralization possibility of the corresponding position. According to the invention, the accuracy and reliability of finding the metallogenic area are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ore-forming prediction methods, and particularly relates to an ore-forming prediction method and system based on geological big data. BACKGROUND

[0002] In modern mineral resource exploration, the ore-forming prediction method based on geological big data plays an important role. This kind of method comprehensively analyzes data of different sources such as geochemistry, geophysics, remote sensing images, and regional geological structure, uses statistical or machine learning models to mine the hidden ore-forming rules in the data, then quantitatively evaluates the ore-forming potential of unknown areas through the construction of a prediction model, and finally generates an ore-forming data channel. This technical means can combine the experience and knowledge of geologists with the objective rules of massive data, effectively reduce the scope of the prospecting target area, significantly reduce the exploration cost and risk, and thus improve the overall efficiency and success rate of mineral exploration.

[0003] However, the existing ore-forming prediction model fails to fully consider the directionality and multiscale nature of the geological anomaly field. The ore-forming process is often controlled by specific trending geological structures, such as fault zones and fold axes, resulting in the strip-shaped and linear distribution characteristics of geochemical or geophysical anomalies along the structure, which is known as directionality. However, traditional models often use symmetric square analysis windows for feature extraction, which treats all directions equally and cannot effectively identify and enhance continuous anomalies in specific directions, easily misjudging real structural anomalies as background noise. At the same time, the structure system controlling ore formation has multiscale nature, from large-scale deep faults to local small fractures, and different scales of structures jointly constitute the favorable space for ore body formation. Existing methods usually analyze data at a single spatial scale or use simple data channel scaling for rough fusion, lacking a mechanism to systematically integrate directional features at different scales, ignoring the prior knowledge of geological structure, and reducing the accuracy of the prediction results of the ore-forming prediction model. SUMMARY

[0004] To solve the problem of low accuracy of the prediction results of the existing ore-forming prediction model, the present application provides an ore-forming prediction method and system based on geological big data.

[0005] In the first aspect, the present application provides an ore-forming prediction method based on geological big data, which adopts the following technical scheme: The multi-source geological data is acquired and rasterized to obtain multiple types of data channels, a spatial variation rate vector of each position in the data channel is calculated, and all spatial variation rate vectors of the same position are superimposed to obtain a spatial variation rate vector field; for each position, the local area is divided into two sub-areas along multiple test directions, the vector sum of all spatial variation rate vectors in each sub-area is calculated, the directional significance of the test direction is calculated by using the vector sum of the two sub-areas, the directional significance is negatively correlated with the dot product of the two vector sums, and the test direction corresponding to the maximum directional significance is taken as the structural direction of the position; the data channel is directionally enhanced to obtain a directionally enhanced data channel, and a metallogenic potential index of each position is calculated based on the directionally enhanced data channel, so as to evaluate the metallogenic possibility of the corresponding position.

[0006] By dividing the area in the local area along multiple test directions, the vector sum of the spatial variation rate vector field is calculated and compared, so that the directional significance is defined and the most possible structural direction is determined. Based on the structural direction, the data channel is enhanced, which can effectively amplify the linear abnormal signal along the specific geological structural direction, while suppressing the background noise randomly distributed. The directionality of the geological anomaly field is quantified and utilized as a key parameter, so that the model can simulate the control effect of the geological structure on the mineralization, thereby significantly improving the sensitivity and accuracy of identifying real structural anomalies from massive geological data, and improving the reliability of the metallogenic potential prediction.

[0007] Preferably, the Sobel operator is used to calculate the spatial variation rate vector.

[0008] The Sobel operator not only considers the influence of adjacent pixel points, but also gives a greater weight to the grid close to the center point, which can better smooth the noise and provide more stable and reliable basic data for subsequent calculation of directional significance, thereby ensuring the effectiveness of the spatial variation rate calculation.

[0009] Preferably, the expression of the directional significance is: ; Wherein, represents the directional significance of the direction at the position , represents the angle of the test direction, , respectively represent all adjacent positions on one side and the other side of the straight line with as the center and as the direction in the local area, represents the sum of the spatial variation rate vectors of all adjacent positions on one side in the local area, This represents the sum of the spatial rate of change vectors of all neighboring locations on the other side of the local region. This represents the magnitude of the vector.

[0010] By utilizing the properties of the vector dot product, the dot product of the vector sums of two sub-regions is negative, ensuring a positive and large significance when the two vector sums are in opposite directions. Simultaneously, multiplying the formula by the magnitude of the two vector sums ensures a large value is only obtained when significant changes exist on both sides, avoiding misjudgments caused by weak changes on one side. This allows for precise quantification and identification of potential geological structural directions.

[0011] The preferred expression for the directional enhancement data channel is: ; in, Indicates the location Enhance the channel in the direction of data channel k. Indicated by The local area centered on Indicates the location Data channel k at location, , These represent the preset distance weights and direction weights, respectively. , Indicates the position coordinates.

[0012] By combining the original data, distance weights, and direction weights, selective amplification of information in specific directions is achieved, enhancing the continuity and intensity of linear anomaly signals along the structural direction. This makes the characteristics of mineral exploration targets more prominent and provides a data foundation with a higher signal-to-noise ratio for the subsequent calculation of the mineralization potential index.

[0013] Preferably, the expressions for distance weight and direction weight are: ; ; in, Indicates distance weight, This represents the standard deviation of the squared Euclidean distances from all locations within a local region to the center location. Indicates directional weights. This represents the distance from all locations within a local area to the center location. standard deviation Indicates from point to The direction angle, Indicates position The structural direction, , The coordinates represent the position coordinates, and exp represents an exponential function with base e.

[0014] The distance weight adopts a Gaussian function, ensuring locality, i.e. the closer the points, the greater the influence, in line with the principle of spatial autocorrelation. The direction weight also adopts the form of a Gaussian function, the variable of which is the included angle between the direction of the neighboring point and the construction direction, which makes the weight distribution form a sharp peak in the construction direction, ensuring that only information consistent with the construction direction can be effectively enhanced, greatly improving the accuracy and effect of direction enhancement.

[0015] Preferably, before calculating the metallogenic potential index, the following steps are further included: multi-level down-sampling of the same type of data channel to form data of different scales, direction enhancement of the data of different scales to obtain direction enhancement channels of different scales, enlarging the direction enhancement channel obtained at a low scale to the original scale through an interpolation method, and merging the direction enhancement data channels of all scales according to the type of the data channel to obtain a comprehensive data set, the comprehensive data set including multiple types of features.

[0016] By multi-level down-sampling of the data, repeated direction identification and enhancement are performed at different scales, and finally the results of each scale are fused. This can capture tectonic information of different scales at the same time, comprehensively depict the favorable space for mineralization, overcome the limitations of single-scale analysis, and enable the prediction model to consider the influence of multi-level geological structures, thereby obtaining more comprehensive and accurate prediction results than the prior art.

[0017] Preferably, the expression of the metallogenic potential index is: ; Wherein, represents the metallogenic potential index of the position , represents the value of the type feature in the position in the comprehensive data set after merging, and represents the weight of the type feature.

[0018] The continuous product model is adopted, and each feature value is first added 1 and then subjected to weighted exponential operation, reflecting the principle that mineralization is a multi-factor coupling process, i.e. multiple favorable conditions must exist at the same time to form a high-potential area, and the absence of any key feature will significantly lower the final potential index. Using weights as indices makes the features more closely related to mineralization have an exponential growth in the influence on the final result, which is more in line with the dominant role of key ore-controlling factors than simple linear weighting, making the final potential evaluation more in line with the geological mineralization regularity.

[0019] Preferably, the ratio of the average value of the ore point area feature to the average value of the feature of the non-ore point area is taken as the weight of the feature. ​

[0020] Preferably, the local area for calculating the structural direction is a window area of MxN.

[0021] In a second aspect, the present application provides a mineralization prediction system based on geological big data, which adopts the following technical scheme: A mineralization prediction system based on geological big data comprises a processor and a memory, and the memory stores computer program instructions, which, when executed by the processor, implement a mineralization prediction method based on geological big data according to the above.

[0022] The mineralization prediction method based on geological big data is generated into a computer program and stored in the memory to be loaded and executed by the processor, so that the system is convenient to use according to the memory and the processor.

[0023] The present application has the following technical effects: The present application automatically identifies and quantifies the linear structural direction related to mineralization from multi-source geological data, and selectively enhances different scale data using the direction, thereby effectively amplifying the prospecting information in the data background, overcoming the fundamental defect of the traditional method that the prediction accuracy is not high due to ignoring the directionality, and improving the accuracy and reliability of finding the mineralization area. BRIEF DESCRIPTION OF DRAWINGS

[0024] Figure 1 is a flowchart of a mineralization prediction method based on geological big data of the present application. DETAILED DESCRIPTION

[0025] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0026] The present application discloses a mineralization prediction method based on geological big data, which refers to Figure 1 and comprises the following steps: S1: Obtain multi-source geological data and calculate the spatial variation rate field.

[0027] When a study area is to be predicted for its metallogenic potential, firstly, multi-source geological data in the study area are collected from a geological database, for example, the multi-source geological data include geochemical data, aeromagnetic anomaly data, gravity anomaly data, and remote sensing alteration information, etc., and the study area is divided into multiple grid regions. In order to facilitate unified processing, the collected multi-source geological data are subjected to coordinate registration and spatial gridding processing, so as to ensure that all the data are aligned on the same coordinate grid to form a group of data channels. The data at each position in each data channel are subjected to normalization processing to map the value range thereof to an interval, and finally a three-dimensional data is formed, denoted as , wherein, represents a grid position in a geographical space, represents a data channel type, for example, gold element content, magnetic anomaly value, etc.

[0028] Subsequently, for each data channel , a operator is applied to calculate a spatial variation rate vector of each position . Then, the spatial variation rate vectors of different data channels at the same position are subjected to vector superposition to generate a total spatial variation rate vector field . The vector field comprehensively reflects the overall variation trend and intensity of all geological variables at the corresponding position.

[0029] S2: calculate the direction significance of each position in the spatial variation rate vector field to obtain a structural direction.

[0030] Linear structures of fracture zones or fold axes are usually decisive factors affecting the formation of ore deposits. The presence of linear structures makes it more likely to form ore deposits. Linear structures often exhibit obvious linear or zonal characteristics. For example, a fracture zone often exhibits a ridge shape on an abnormal data channel, and the spatial variation rate directions on both sides of the ridge are opposite and point away from the ridge. It should be noted that the ridge here refers to a continuous region exhibiting linear or curved characteristics and having a local high value in a two-dimensional data field. Similarly, a continuous region exhibiting linear or curved characteristics and having a local low value in a two-dimensional data field is referred to as a valley.

[0031] In addition, in step S1, although the types of the collected data are different, the linear structures of the fracture zones or fold axes exhibit obvious linear or zonal characteristics in different data channels. The reason is that, for the data channel of aeromagnetic anomaly, when a fracture cuts different magnetic rock bodies, or the fracture zone itself is intruded by a later rock vein, a very clear linear or zonal anomaly will be formed on the aeromagnetic data channel. The profile shape of this anomaly is a typical ridge or valley, and the spatial variation rate directions of the magnetic field intensity on both sides of the ridge are necessarily opposite.

[0032] Similarly, for the data channel of gravity anomaly, if the rock density on both sides of the fault is different, or if high-density / low-density material intrudes / alters along the fault zone, a linear gravity high or gravity low anomaly will be formed on the gravity data channel. The morphology of these anomalies is also ridge or valley, and the direction of gravity spatial variation rate on both sides of the ridge or valley is necessarily opposite; for the data channel of geochemical data, when the fault serves as the channel for the migration of ore-forming hydrothermal fluids, specific ore-forming elements or associated indicator elements will be enriched along the fault zone. On the element content contour data channel generated by spatial interpolation of the collected discrete sample point data, this enrichment zone will appear as a clear ridge-shaped high-value area, and the element concentration gradually decreases from the center of the ridge to both sides, so the direction of the spatial variation rate of the concentration is necessarily directed to the low-value areas on both sides, which fully complies with the physical priori.

[0033] For the data channel of remote sensing data, during the migration of ore-forming hydrothermal fluids, reactions with surrounding rocks usually occur to form specific alteration mineral assemblages, such as pyritization, sericitization, and silicification, etc. These alteration zonations can be identified and delineated on remote sensing images through spectral analysis. If the alteration is strictly controlled by fault structures, then in the final generated alteration intensity data channel, the area along the fault zone will show a high-intensity anomaly, and the morphology and spatial variation rate characteristics of this anomaly zone also comply with the ridge model.

[0034] Specifically, for each position on the spatial variation rate vector field , the directional significance is calculated to evaluate whether the direction is suitable as the structural direction of the position, and the expression is as follows:

[0035] wherein represents the directional significance of direction at position , represents the angle of the test direction, which is in the range of , represents all adjacent positions on one side of the straight line with as the center and direction in the local area, represents all adjacent positions on the other side of the straight line, represents the sum of the spatial variation rate vectors of all adjacent positions on one side in the local area, reflecting the total spatial variation rate direction of the side, represents the sum of the spatial variation rate vectors of all adjacent positions on the other side in the local area, reflecting the total spatial variation rate direction of the other side, represents the vector module. The size of the local region can be set according to the research scale, for example, a 7x7 window is constructed with the position as the center, and the window region is the local region of the position .

[0036] For each position , a local region is set around it, and the local region is divided into two parts for each possible direction . The sum of the spatial variation rate vectors of all positions in the two parts is calculated respectively to obtain and , the dot product is calculated and the negative value is taken, when the directions of the two vectors are opposite, the dot product is negative; when the directions of the two vectors are the same, the dot product is positive. After taking the negative sign, when the directions of the two vectors are opposite, the item is positive and the value is larger, indicating that there is a significant geological anomaly in that direction. is the product of the total spatial variation rate intensity on both sides, when there is a strong spatial variation rate change on both sides, the product is larger, which ensures that the direction significance not only considers the opposite direction, but also considers the intensity of the spatial variation rate on both sides, thereby avoiding false judgment due to weak spatial variation rate on one side.

[0037] The negative value of the dot product is multiplied by the product of the lengths of the two vectors to obtain the direction significance. For a fixed position , different outputs different direction significance, the larger the direction significance, the more likely it is that the position belongs to the linear structure in that direction, and the greater the possibility of mineralization. Further, among all directions, the direction with the largest direction significance is selected as the structural direction of the position, and the structural direction is denoted as , the structural direction reflects the extension direction of the linear structure.

[0038] S3: Direction enhancement is performed on the data channel to obtain a direction-enhanced channel data.

[0039] The formation of ore bodies is often influenced by specific geological structures in certain directions, and the information in these directions is more important than that in other directions. By enhancing the information in these directions, potential ore deposit locations can be more clearly identified, reducing irrelevant information interference and improving the accuracy of ore prediction.

[0040] A direction-enhanced data channel is generated for each data channel, specifically, for each position of each type of data channel in the original data, the expression of the direction-enhanced data channel is: ; ; ; wherein, denotes the directional enhancement channel of data channel k at position , denotes a local region centered at , e.g. a 21x21 window region, denotes data channel k at position , denotes a distance weight, the value of which is higher the closer to the center point, denotes the standard deviation of the squared Euclidean distance of all positions in the local region to the center position, denotes a directional weight, denotes the standard deviation of the of all positions in the local region to the center position, denotes the directional angle from to . denotes the construction direction of position , , denotes the position coordinate, exp denotes the exponential function with base e, if the value of is 0, then the value of is 1, if the value of is 0, then the value of is 1.

[0041] The locality is ensured, i.e. the closer to the center point the higher the weight of the position, which can reduce the influence of irrelevant information far away. The direction perception is realized, i.e. only when the direction of the neighboring points is consistent with the construction direction, a higher weight will be obtained. This makes the value after directional enhancement fuse the information along the construction direction, thereby enhancing the linear abnormal signal. The directional enhancement of the data channel obtains the directional enhancement channel, which reduces the interference of noise.

[0042] S4: Calculate the metallogenic potential index of each position.

[0043] The formation of ore bodies is not only affected by local small structures, but also related to larger-scale geological structures. By extracting directional features at multiple spatial scales and fusing them, the metallogenic environment can be comprehensively understood, thereby making more accurate metallogenic prediction.

[0044] First, the original three-dimensional data structure Multi-level downsampling is performed to generate data at different scales. For example, downsampling can be performed sequentially by 2x, 4x, and 8x to obtain three sets of data at the original scale, mesoscale, and macroscale. Next, for each scale of data, the calculation processes S2 and S3 are repeated to obtain directional enhancement channels at different scales.

[0045] Then, all the orientation enhancement channels calculated at low scale (coarse scale) are amplified back to the original scale using interpolation. The orientation enhancement data channels at all scales are then merged according to their data channel type to form a comprehensive dataset F containing rich multi-scale directional features. Each feature class in the comprehensive dataset F is denoted as... .

[0046] Using the linear normalization algorithm on the comprehensive dataset Each feature in the dataset is normalized, and the location of the mining site is calculated using the known locations of mining sites in historical data. The average value of each characteristic in the mining area and the average value in non-mineralized areas The ratio of the two is used as the weight of the feature. .

[0047] The mineralization potential index for each location is calculated using the following expression: ; in, Indicates position The mineralization potential index Represents all merged feature categories Perform a series of multiplications. This indicates the first element in the merged aggregate dataset. Class features at location The value, Indicates the first Weights of class features.

[0048] For each feature, calculate its weight based on known mineral deposits. For each position Each feature value is incremented by 1, and then multiplied by its weight as an exponent. The results of multiplying all the features together are used to obtain the mineralization potential index for that location. This incorporates the importance of all features; adding 1 prevents a single feature value of 0 from causing the entire product to be zero. Weights are used. As an index, the stronger the positive correlation with mineralization of the feature is, the greater the influence of its numerical size on the final potential index is. This reflects the principle of multi-factor coupling of mineralization, that is, multiple favorable conditions must exist at the same time to form a higher mineralization potential. For each location in the study area, the greater the mineralization potential index is, the more likely there is ore, and vice versa.

[0049] S5: evaluating the mineralization potential.

[0050] According to the mineralization potential index of each location in the study area, the area with the most prospecting potential is obtained, for example, selecting the top 5% of the mineralization potential index, the area with the most prospecting potential in the study area can be directly delineated, thereby providing guidance for subsequent high-cost exploration activities such as field verification and drilling engineering.

[0051] The embodiment of the present application also discloses a mineralization prediction system based on geological big data, comprising a processor and a memory, and the memory stores computer program instructions, which realize the mineralization prediction method based on geological big data according to the present application when executed by the processor.

[0052] The above system also includes a communication bus and a communication interface and other components well known to those skilled in the art, the settings and functions of which are known in the art, and therefore will not be repeated here.

[0053] The above are preferred embodiments of the present application, which do not limit the protection scope of the present application, therefore: any equivalent changes made in the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. A method for metallogenic prediction based on geological big data, characterized in that, Including the following steps: Multi-source geological data is acquired and rasterized to obtain multiple data channels. The spatial rate of change vector at each location in the data channel is calculated, and all spatial rate of change vectors at the same location are superimposed to obtain a spatial rate of change vector field. For each location, the local area is divided into two sub-regions along multiple test directions. The vector sum of all spatial rate of change vector fields in each sub-region is calculated. The directional significance of the test direction is calculated using the vector sum of the two sub-regions. The directional significance is negatively correlated with the dot product of the two vector sums. The test direction corresponding to the maximum directional significance is taken as the structural direction of the location. The data channels are directionally enhanced to obtain directionally enhanced data channels. Based on the directionally enhanced data channels, the mineralization potential index of each location is calculated to assess the mineralization probability of the corresponding location. 2.The metallogenic prognosis method based on geological big data according to claim 1, characterized in that, The spatial rate of change vector is calculated using the Sobel operator. 3.The metallogenic prognosis method based on geological big data according to claim 1, characterized in that, The expression for directional significance is: ； in, Indicates the location Upward directional salience, Angle indicating the test direction , They represent local areas as follows: Centered on, direction is All adjacent locations on one side and the other side of the straight line, This represents the sum of the spatial rate of change vectors of all neighboring locations on one side within a local region. This represents the sum of the spatial rate of change vectors of all neighboring locations on the other side of the local region. This represents the magnitude of the vector.

4. The mineralization prediction method based on geological big data according to claim 1, characterized in that, The expression for the directional enhancement data channel is: ； in, Indicates the location Enhance the channel in the direction of data channel k. Indicated by The local area centered on Indicates the location Data channel k at location, , These represent the preset distance weights and direction weights, respectively. , Indicates the position coordinates.

5. The mineralization prediction method based on geological big data according to claim 4, characterized in that, The expressions for distance weight and orientation weight are: ； ； in, Indicates distance weight, This represents the standard deviation of the squared Euclidean distances from all locations within a local region to the center location. Indicates directional weights. This represents the distance from all locations within a local area to the center location. standard deviation Indicates from point to The direction angle, Indicates position The structural direction, , The coordinates represent the position coordinates, and exp represents an exponential function with base e.

6. The mineralization prediction method based on geological big data according to claim 1, characterized in that, Before calculating the mineralization potential index, the following steps are also taken: multi-level downsampling of data channels of the same type to form data at different scales; directional enhancement of data at different scales to obtain directional enhancement channels at different scales; directional enhancement channels calculated at low scales to be amplified back to the original scale through interpolation; and directional enhancement data channels at all scales to be merged according to data channel type to obtain a comprehensive dataset, which includes multiple types of features.

7. The mineralization prediction method based on geological big data according to claim 6, characterized in that, The expression for the mineralization potential index is: ； in, Indicates position The mineralization potential index This indicates the first element in the merged aggregate dataset. Class features at location The value, Indicates the first Weights of class features.

8. The mineralization prediction method based on geological big data according to claim 7, characterized in that, The ratio of the average value of the features in the mining area to the average value of the features in the non-mining area is used as the weight of the feature.

9. The mineralization prediction method based on geological big data according to claim 1, characterized in that, A window region of M×N is used to calculate the local region of the construction direction.

10. A mineralization prediction system based on geological big data, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a mineralization prediction method based on geological big data according to any one of claims 1-9.

Images