Intelligent mineralization prediction method based on geological big data

Abstract

The present invention relates to the field of mineralization prediction technology, and in particular to an intelligent mineralization prediction method based on geological big data, including establishing a mapping relationship between mineralization data features, detecting a mining area to be explored, and when a mineralization data feature is detected in the mining area to be explored, determining whether a first mineralization data feature has an associated mineralization data feature, and if so, recording the associated mineralization data feature and the corresponding weight information to form a first associated comparison data feature set, calculating a mineralization characterization value to determine the mineralization grade of the mining area to be explored, and whether the mineralization characterization value needs to be corrected to determine the mineralization grade of the mining area to be explored for a second time, and if so, calculating a reference correction coefficient to correct the mineralization characterization value, and re-determining the mineralization grade of the mining area to be explored. The present invention improves the accuracy and reliability of mineralization prediction.

Description

Technical Field

[0001] The present invention relates to the technical field of mineralization prediction, and in particular to an intelligent mineralization prediction method based on geological big data. Background Art

[0002] Mineralization prediction is a comprehensive research work carried out to improve the effectiveness and foresight of mineral exploration. Its main process is to comprehensively analyze the geological characteristics of the area and the types, scales and co-existence of various minerals with geological structures in time and space based on the actual data of various geological minerals and geophysical and geochemical exploration in the working area, clarify their mineralization laws, and then predict the favorable areas and controlling conditions where minerals may be discovered in the area, point out the direction, sequence and content of further work, etc., to provide a basis for the ongoing or next stage of general prospecting work.

[0003] Chinese patent publication number: CN104865613B discloses a mineralization prediction method, which relates to the field of mineral resource exploration. It is invented to solve the problem of low accuracy of mineralization prediction. It includes: S10, full-hole sampling from the surface to the underground of the prospecting mining area according to the preset spacing to obtain a group of sampled samples; S20, obtaining the three-dimensional spatial position of each sampled sample in a group of sampled samples; S30, respectively determining the main mineralization elements of all sampled samples and the fluid inclusion parameters of each sampled sample; S40, according to the three-dimensional spatial position of each sampled sample, the main mineralization elements are interpolated by the three-dimensional distance inverse weighted method to obtain the three-dimensional spatial distribution model of the main mineralization elements; S50, according to the three-dimensional spatial position of each sampled sample, the fluid inclusion parameters are interpolated by the three-dimensional distance inverse weighted method to obtain the three-dimensional spatial distribution model of the fluid inclusion parameters; S60, according to the three-dimensional spatial distribution model of the main mineralization elements and fluid inclusion parameters, determine the prospecting direction of the main mineralization elements.

[0004] However, the influence weights of various metallogenic factors on the metallogenic results are different, and there are mutual correlations and interactions among the metallogenic factors. The existing technology usually predicts the probability of mineralization by metallogenic factors without considering the interactions between the metallogenic factors, resulting in the need to improve the accuracy of mineralization prediction. Summary of the invention

[0005] To this end, the present invention provides an intelligent mineralization prediction method based on geological big data, so as to overcome the problem that the probability of mineralization is usually predicted by mineralization factors in the prior art without considering the interaction between the mineralization factors, resulting in the accuracy of mineralization prediction needs to be improved.

[0006] To achieve the above object, the present invention provides an intelligent mineralization prediction method based on geological big data, comprising:

[0007] Establishing a mineralization data feature set and a weak mineralization data feature set by analyzing geological big data of the mineralization area, and establishing a mapping relationship between the mineralization data features, wherein the mapping relationship includes an association relationship between the mineralization data features and a weight value between the association relationships;

[0008] Detecting geological data of the mining area to be explored, when a mineralization data feature is detected in the geological data of the mining area to be explored, marking the mineralization data feature as a first mineralization data feature, determining whether the first mineralization data feature has an associated mineralization data feature, and if so, recording the associated mineralization data feature and the corresponding weight information to form a first associated comparison data feature set;

[0009] Calculating a metallogenic characterization value according to the first associated comparative data feature set, determining the metallogenic grade of the to-be-explored mineral area according to the metallogenic characterization value, and determining whether the metallogenic characterization value needs to be corrected to re-determine the metallogenic grade of the to-be-explored mineral area according to the number of metallogenic data features in the first associated comparative data feature set;

[0010] When the mineralization characterization value is corrected, a reference correction coefficient is calculated according to the weak mineralization data characteristics appearing in the mining area to be explored to correct the mineralization characterization value and re-determine the mineralization grade of the mining area to be explored.

[0011] Furthermore, when analyzing the geological big data of the mineralization area, for any mineralization geological data, if the probability of its occurrence is greater than or equal to the preset probability comparison value, the mineralization geological data is marked as a mineralization data feature and divided into the mineralization data feature set, and the mineralization geological data whose probability of occurrence is less than the preset probability comparison value is marked as a weak mineralization data feature and divided into the weak mineralization data feature set.

[0012] Furthermore, when determining whether the first mineralization data feature has associated mineralization data features, other mineralization data features and weight information associated with the first mineralization data feature are called according to the mapping relationship and recorded as a first associated data feature set, and the geological data of the mining area to be explored is continued to be detected to see whether the mineralization data features in the first associated data feature set appear, and the appearing mineralization data features and corresponding weight information are recorded as a first associated comparison data feature set.

[0013] Further, the mapping relationship is expressed as QB(Qi, Qi j, Qi jk), wherein Qi represents the i-th mineralization data feature, i=1, 2...n, Qi j represents the j-th mineralization data feature associated with the i-th mineralization data feature, Qi jk represents the weight value between the j-th mineralization data feature associated with the i-th mineralization data feature, j=1, 2...n, and the weight value represents the frequency of occurrence of the j-th mineralization data feature when the i-th mineralization data feature appears.

[0014] Further, the metallogenic characterization value is calculated according to the number of metallogenic data features in the first associated comparative data feature set and the weight data between the first metallogenic data features and the first metallogenic data features, and is set as follows:

[0015]

[0016] Among them, F represents the mineralization characterization value, Qi represents the weight value between the association relationship between the mineralization data features in the first associated comparison data feature set and the first mineralization data feature, Qp represents the average value of the weight value between the association relationship between the mineralization data features in the first associated comparison data feature set and the first mineralization data feature, N represents the number of mineralization data features in the first associated comparison data feature set, N>2, and S represents the number of mineralization data features in the first associated comparison data feature set.

[0017] Furthermore, the mineralization characterization value is compared with a preset mineralization reference value to determine the mineralization grade of the mining area to be explored;

[0018] Wherein, when the metallogenic characterization value is less than or equal to the metallogenic reference value, it is determined whether to correct the metallogenic characterization value to secondarily determine the metallogenic grade of the mining area to be explored according to the ratio of the number of metallogenic data features in the first associated comparison data feature set to the number of metallogenic data features in the first associated data feature set;

[0019] When the mineralization characterization value is greater than the mineralization reference value, the mineralization grade of the mining area to be explored is determined to be a high mineralization grade.

[0020] Furthermore, the determination process of whether to modify the mineralization characterization value includes:

[0021] Calculate the quantitative ratio of the number of mineralization data features in the first associated comparison data feature set to the number of mineralization data features in the first associated data feature set; if the quantitative ratio is less than or equal to a preset ratio standard, determine that the mineralization characterization value is not corrected, and determine that the mineralization grade of the mining area to be explored is a low mineralization grade; if the quantitative ratio is greater than the preset ratio standard, correct the mineralization characterization value according to the weak mineralization data features appearing in the mining area to be explored.

[0022] Furthermore, the process of correcting the metallogenic characterization value according to the weak metallogenic data features appearing in the mining area to be explored includes:

[0023] Record the weak mineralization data features that appear in the mining area to be explored. If the number of the recorded weak mineralization data features is less than or equal to the preset number value, do not correct the mineralization characterization value, and judge that the mineralization grade of the mining area to be explored is a low mineralization grade. If the number of the recorded weak mineralization data features is greater than the preset number value, calculate the reference correction coefficient to correct the mineralization characterization value.

[0024] Further, the reference correction coefficient is calculated according to the following formula, setting

[0025] C=m×∑Gl

[0026] Wherein, C is the reference correction coefficient, m is the number of the weak metallogenic data features appearing in the mining area to be explored, Gl is the probability of the occurrence of the lth weak metallogenic data feature, and the value of l is 1 to m.

[0027] Furthermore, the mineralization characterization value is corrected according to the reference correction coefficient, and when the corrected mineralization characterization value is greater than the mineralization reference value, the mineralization grade of the mining area to be explored is determined to be a high mineralization grade.

[0028] Compared with the prior art, the beneficial effect of the present invention lies in that the present invention analyzes the geological big data of the mineralization area, constructs a mapping relationship between the mineralization data features, performs associated feature detection on the mineralization area to be explored, performs associated feature detection and calculates the mineralization characterization value through the mapping relationship, preliminarily determines the mineralization grade of the mineralization area to be explored, and calculates the reference correction coefficient to correct the mineralization characterization value in order to intervene and correct the mineralization grade determination result of the mineralization area to be explored, thereby obtaining a more accurate identification result. By constructing a mapping relationship, the present invention fully considers the influence of the mineralization data features that are difficult to identify or hidden in the mineralization area to be explored on the results of mineralization prediction, reduces missed judgments, and improves the prediction accuracy of the mineralization area to be explored.

[0029] Furthermore, since there is mutual influence between the geological data of mineralization, the present invention constructs a mapping relationship, which includes weights between associated mineralization data features and corresponding features, and assigns weights to the mineralization data features according to big data. The weight values ​​are obtained by model training based on the probability of occurrence of mineralization data features in the geological big data of the mineralization area, thereby improving the prediction accuracy of the mineralization area to be explored through the mineralization data features.

[0030] Furthermore, the present invention calculates the mineralization characterization value by the number of occurrences of mineralization data features and the weight values ​​between the mineralization data features. The mineralization characterization value is a representative parameter of the mineralization probability of the mineral area to be explored, and is used to characterize the magnitude of the mineralization probability of the mineral area to be explored. For the occurrence of multiple related mineralization data features and the greater the weights between the related mineralization data features, the greater the mineralization characterization value is, indicating that the mineralization grade of the mineral area to be explored is higher. The mineralization grade of the mineral area to be explored is predicted based on this judgment, which is representative and has a high prediction accuracy.

[0031] Furthermore, the present invention calculates the quantitative ratio of the number of mineralization data features in the first associated comparison data feature set to the number of mineralization data features in the first associated data feature set for the mining area to be explored with a low mineralization characterization value. The larger the quantitative ratio is, the more mineralization data features with associated relationships appear in the mining area to be explored. At this time, the mineralization characterization value is corrected according to the weak mineralization data features appearing in the mining area to be explored. Through the above technical scheme, the influence of weak mineralization data features on the mineralization grade can be fully considered, the missed judgment is reduced, and the prediction accuracy of the mining area to be explored is thereby improved.

[0032] Furthermore, the present invention records the weak mineralization data features appearing in the mining area to be explored. If the number of recorded weak mineralization data features is greater than a preset value, the mineralization grade of the mining area to be explored is correspondingly improved, even if the influence of the weak mineralization data features on the mineralization prediction of the mining area to be explored is not as great as that of the mineralization data features. However, such hidden mineralization data features are identified, and the mineralization probability of the mining area to be explored is characterized by calculating the reference correction coefficient, and the mineralization characterization value corresponding to the mining area to be explored is corrected based on this, which intervenes in the determination result of the mineralization grade of the mining area to be explored, thereby improving the prediction accuracy of the mining area to be explored. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 The present invention provides a flowchart of an intelligent mineralization prediction method based on geological big data according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0035] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the protection scope of the present invention.

[0036] See also Figure 1 As shown, the intelligent mineralization prediction method based on geological big data of the present invention comprises:

[0037] Step S1, establishing a mineralization data feature set and a weak mineralization data feature set by analyzing geological big data of a mineralization area, and establishing a mapping relationship between mineralization data features, wherein the mapping relationship includes an association relationship between the mineralization data features and a weight value between the association relationships;

[0038] Step S2, detecting geological data of the mining area to be explored, and when a mineralization data feature is detected in the geological data of the mining area to be explored, marking the mineralization data feature as a first mineralization data feature, and determining whether the first mineralization data feature has an associated mineralization data feature, and if so, recording the associated mineralization data feature and the corresponding weight information to form a first associated comparison data feature set;

[0039] Step S3, calculating a metallogenic characterization value according to the first associated comparative data feature set, determining the metallogenic grade of the to-be-explored mineral area according to the metallogenic characterization value, and determining whether the metallogenic characterization value needs to be corrected to re-determine the metallogenic grade of the to-be-explored mineral area according to the number of metallogenic data features in the first associated comparative data feature set;

[0040] Step S4, when correcting the mineralization characterization value, a reference correction coefficient is calculated according to the weak mineralization data characteristics appearing in the mining area to be explored to correct the mineralization characterization value, and the mineralization grade of the mining area to be explored is re-determined.

[0041] The present invention analyzes the geological big data of the mineralization area, constructs a mapping relationship between the mineralization data features, performs associated feature detection on the mineralization area to be explored, performs associated feature detection and calculates the mineralization characterization value through the mapping relationship, preliminarily determines the mineralization grade of the mineralization area to be explored, and calculates the reference correction coefficient to correct the mineralization characterization value for the mineralization grade determination result of the mineralization area to be explored, so as to obtain a more accurate identification result. By constructing the mapping relationship, the present invention fully considers the influence of the mineralization data features that are difficult to identify or hidden in the mineralization area to be explored on the results of mineralization prediction, reduces missed judgments, and improves the prediction accuracy of the mineralization area to be explored.

[0042] Specifically, when analyzing the geological big data of a mineralization area, for any mineralization geological data, if the probability of its occurrence is greater than or equal to a preset probability comparison value, the mineralization geological data will be marked as a mineralization data feature and divided into the mineralization data feature set, and the mineralization geological data whose probability of occurrence is less than the preset probability comparison value will be marked as a weak mineralization data feature and divided into the weak mineralization data feature set.

[0043] In this embodiment, when setting the preset probability comparison value, the geological big data is analyzed to obtain the probability value of any mineralization geological data, and a probability value data set is constructed. The probability value is used as a random variable, and the random variable obeys a probability density function. Then, a normal distribution curve can be constructed according to the probability density function, and the 95% confidence interval of the normal distribution curve is determined. The probability value corresponding to the lower limit of the 95% confidence interval is determined as the preset probability comparison value. Constructing a probability density function and constructing a normal distribution curve based on the probability density function are commonly used technical means in statistics, which will not be repeated here.

[0044] Specifically, when determining whether the first mineralization data feature has associated mineralization data features, other mineralization data features and weight information associated with the first mineralization data feature are called according to the mapping relationship and recorded as a first associated data feature set, and the geological data of the mining area to be explored is continued to be detected to see whether the mineralization data features in the first associated data feature set appear, and the appearing mineralization data features and corresponding weight information are recorded as a first associated comparison data feature set.

[0045] It can be understood that if the first associated comparison data feature set is not formed, it is determined that the mineralization grade of the mining area to be explored is low.

[0046] Specifically, the mapping relationship is expressed as QB(Qi, Qi j, Qi jk), wherein Qi represents the i-th mineralization data feature, i=1, 2...n, Qi j represents the j-th mineralization data feature associated with the i-th mineralization data feature, Qi jk represents the weight value between the j-th mineralization data feature associated with the i-th mineralization data feature, j=1, 2...n, and the weight value represents the frequency of occurrence of the j-th mineralization data feature when the i-th mineralization data feature appears.

[0047] Since there is mutual influence between the geological data of mineralization, the present invention constructs a mapping relationship, which includes weights between associated mineralization data features and corresponding features, and assigns weights to the mineralization data features according to big data. The weight values ​​are obtained by model training based on the probability of occurrence of mineralization data features in the geological big data of the mineralization area, thereby improving the prediction accuracy of the exploration mining area through the mineralization data features.

[0048] Specifically, the metallogenic characterization value is calculated according to the number of metallogenic data features in the first associated comparative data feature set and the weight data between the first metallogenic data features and the first metallogenic data features, and is set as follows:

[0049]

[0050] Among them, F represents the mineralization characterization value, Qi represents the weight value between the association relationship between the mineralization data features in the first associated comparison data feature set and the first mineralization data feature, Qp represents the average value of the weight value between the association relationship between the mineralization data features in the first associated comparison data feature set and the first mineralization data feature, N represents the number of mineralization data features in the first associated comparison data feature set, N>2, and S represents the number of mineralization data features in the first associated comparison data feature set.

[0051] The present invention calculates the mineralization characterization value through the number of occurrences of mineralization data features and the weight values ​​between the mineralization data features. The mineralization characterization value is a representative parameter of the mineralization probability of the mineral area to be explored, and is used to characterize the magnitude of the mineralization probability of the mineral area to be explored. For the occurrence of multiple related mineralization data features and the greater the weights between the related mineralization data features, the greater the mineralization characterization value is, indicating that the mineralization grade of the mineral area to be explored is higher. The mineralization grade of the mineral area to be explored is predicted based on this judgment, which is representative and has high prediction accuracy.

[0052] Specifically, the mineralization characterization value is compared with a preset mineralization reference value to determine the mineralization grade of the mining area to be explored;

[0053] Wherein, when the metallogenic characterization value is less than or equal to the metallogenic reference value, it is determined whether to correct the metallogenic characterization value to secondarily determine the metallogenic grade of the mining area to be explored according to the ratio of the number of metallogenic data features in the first associated comparison data feature set to the number of metallogenic data features in the first associated data feature set;

[0054] When the mineralization characterization value is greater than the mineralization reference value, the mineralization grade of the mining area to be explored is determined to be a high mineralization grade.

[0055] In this embodiment, when setting the mineralization reference value, geological data of at least 1,000 mining areas are selected, the mineralization characterization value of each mining area is calculated, and a mineralization characterization value data set is constructed. The mineralization characterization value is taken as a random variable, and the random variable obeys a probability density function. Then, a normal distribution curve can be constructed according to the probability density function, and the 95% confidence interval of the normal distribution curve is determined. The mineralization characterization value corresponding to the midpoint of the 95% confidence interval is determined as the mineralization reference value.

[0056] Specifically, the determination process of whether to modify the mineralization characterization value includes:

[0057] Calculate the quantitative ratio of the number of mineralization data features in the first associated comparison data feature set to the number of mineralization data features in the first associated data feature set; if the quantitative ratio is less than or equal to a preset ratio standard, determine that the mineralization characterization value is not corrected, and determine that the mineralization grade of the mining area to be explored is a low mineralization grade; if the quantitative ratio is greater than the preset ratio standard, correct the mineralization characterization value according to the weak mineralization data features appearing in the mining area to be explored.

[0058] In this embodiment, when setting the ratio standard, geological data of at least 1,000 mining areas are selected to construct the first associated comparison data feature set and the first associated data feature set of each mining area, and the quantity ratio of the number of metallogenic data features in the first associated comparison data feature set of each mining area to the number of metallogenic data features in the first associated data feature set is calculated. By constructing a normal distribution curve, the quantity ratio corresponding to the midpoint of the 95% confidence interval is determined as the ratio standard. The above is only one implementation method, and it can also be set according to actual needs through data analysis. In this embodiment, the preferred ratio standard is greater than 0.8 and less than 1.

[0059] The present invention calculates the quantitative ratio of the number of mineralization data features in the first associated comparison data feature set to the number of mineralization data features in the first associated data feature set for a mineralization area to be explored with a low mineralization characterization value. The larger the quantitative ratio is, the more mineralization data features with associated relationships appear in the mineralization area to be explored. At this time, the mineralization characterization value is corrected according to the weak mineralization data features appearing in the mineralization area to be explored. Through the above technical scheme, the influence of the weak mineralization data features on the mineralization grade can be fully considered, the missed judgment is reduced, and the prediction accuracy of the mineralization area to be explored is thereby improved.

[0060] Specifically, the process of correcting the metallogenic characterization value according to the weak metallogenic data features appearing in the mining area to be explored includes:

[0061] Record the weak mineralization data features that appear in the mining area to be explored. If the number of the recorded weak mineralization data features is less than or equal to the preset number value, do not correct the mineralization characterization value, and judge that the mineralization grade of the mining area to be explored is a low mineralization grade. If the number of the recorded weak mineralization data features is greater than the preset number value, calculate the reference correction coefficient to correct the mineralization characterization value.

[0062] In this embodiment, the preset quantity value can be set by constructing a normal distribution curve, which will not be described in detail.

[0063] Specifically, the reference correction coefficient is calculated according to the following formula:

[0064] C=m×∑Gl

[0065] Wherein, C is the reference correction coefficient, m is the number of the weak metallogenic data features appearing in the mining area to be explored, Gl is the probability of the occurrence of the lth weak metallogenic data feature, and the value of l is 1 to m.

[0066] Specifically, the metallogenic characterization value is corrected according to the reference correction coefficient, and the corrected metallogenic characterization value F1=F×C / C0 is set. When the corrected metallogenic characterization value is greater than the metallogenic reference value, the metallogenic grade of the mining area to be explored is determined to be a high metallogenic grade.

[0067] Among them, F1 represents the corrected mineralization characterization value, C represents the reference correction coefficient, and C0 represents the reference correction coefficient comparison value.

[0068] In this embodiment, when setting the reference correction coefficient comparison value, the reference correction coefficient is used as a random variable to construct a normal distribution curve, and the reference correction coefficient corresponding to the midpoint of the 95% confidence interval of the normal distribution curve is determined as the reference correction coefficient comparison value.

[0069] The present invention records weak mineralization data features appearing in the mining area to be explored. If the number of recorded weak mineralization data features is greater than a preset value, the mineralization grade of the mining area to be explored is correspondingly improved. Even if the influence of the weak mineralization data features on the mineralization prediction of the mining area to be explored is not as great as that of the mineralization data features, such hidden mineralization data features are identified, and the mineralization probability of the mining area to be explored is characterized by calculating a reference correction coefficient, and the mineralization characterization value corresponding to the mining area to be explored is corrected based on this coefficient, which intervenes in the determination result of the mineralization grade of the mining area to be explored, thereby improving the prediction accuracy of the mining area to be explored.

[0070] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent mineralization prediction method based on geological big data, characterized in that: include: Establishing a mineralization data feature set and a weak mineralization data feature set by analyzing geological big data of the mineralization area, and establishing a mapping relationship between the mineralization data features, wherein the mapping relationship includes an association relationship between the mineralization data features and a weight value between the association relationships; Detecting geological data of the mining area to be explored, when a mineralization data feature is detected in the geological data of the mining area to be explored, marking the mineralization data feature as a first mineralization data feature, determining whether the first mineralization data feature has an associated mineralization data feature, and if so, recording the associated mineralization data feature and the corresponding weight information to form a first associated comparison data feature set; Calculating a metallogenic characterization value according to the first associated comparative data feature set, determining the metallogenic grade of the to-be-explored mineral area according to the metallogenic characterization value, and determining whether the metallogenic characterization value needs to be corrected to re-determine the metallogenic grade of the to-be-explored mineral area according to the number of metallogenic data features in the first associated comparative data feature set; When the mineralization characterization value is corrected, a reference correction coefficient is calculated according to the weak mineralization data characteristics appearing in the mining area to be explored to correct the mineralization characterization value and re-determine the mineralization grade of the mining area to be explored.

2. The intelligent mineralization prediction method based on geological big data according to claim 1 is characterized in that: When analyzing geological big data of a mineralization area, for any mineralization geological data, if the probability of its occurrence is greater than or equal to a preset probability comparison value, the mineralization geological data is marked as a mineralization data feature and divided into the mineralization data feature set, and the mineralization geological data whose probability of occurrence is less than the preset probability comparison value is marked as a weak mineralization data feature and divided into the weak mineralization data feature set.

3. The intelligent mineralization prediction method based on geological big data according to claim 2 is characterized in that: When determining whether the first mineralization data feature has associated mineralization data features, other mineralization data features associated with the first mineralization data feature and weight information are called according to the mapping relationship and recorded as a first associated data feature set, and the geological data of the mining area to be explored is continued to be detected to see whether the mineralization data features in the first associated data feature set appear, and the appearing mineralization data features and corresponding weight information are recorded as a first associated comparison data feature set.

4. The intelligent mineralization prediction method based on geological big data according to claim 3 is characterized in that: The mapping relationship is expressed as QB (Qi, Qij, Qijk), wherein Qi represents the i-th mineralization data feature, i=1, 2...n, Qij represents the j-th mineralization data feature associated with the i-th mineralization data feature, Qijk represents the weight value between the j-th mineralization data feature associated with the i-th mineralization data feature, j=1, 2...n, and the weight value represents the frequency of occurrence of the j-th mineralization data feature when the i-th mineralization data feature appears.

5. The intelligent mineralization prediction method based on geological big data according to claim 4 is characterized in that: The metallogenic characterization value is calculated according to the number of metallogenic data features in the first associated comparative data feature set and the weight data between the first metallogenic data features and the first metallogenic data features, and is set as follows: Among them, F represents the mineralization characterization value, Qi represents the weight value between the association relationship between the mineralization data features in the first associated comparison data feature set and the first mineralization data feature, Qp represents the average value of the weight value between the association relationship between the mineralization data features in the first associated comparison data feature set and the first mineralization data feature, N represents the number of mineralization data features in the first associated comparison data feature set, N>2, and S represents the number of mineralization data features in the first associated comparison data feature set.

6. The intelligent mineralization prediction method based on geological big data according to claim 5 is characterized in that: Comparing the mineralization characterization value with a preset mineralization reference value to determine the mineralization grade of the mining area to be explored; Wherein, when the metallogenic characterization value is less than or equal to the metallogenic reference value, it is determined whether to correct the metallogenic characterization value to secondarily determine the metallogenic grade of the mining area to be explored according to the ratio of the number of metallogenic data features in the first associated comparison data feature set to the number of metallogenic data features in the first associated data feature set; When the mineralization characterization value is greater than the mineralization reference value, the mineralization grade of the mining area to be explored is determined to be a high mineralization grade.

7. The intelligent mineralization prediction method based on geological big data according to claim 6 is characterized in that: The determination process of whether to modify the mineralization characterization value includes: Calculate the quantitative ratio of the number of mineralization data features in the first associated comparison data feature set to the number of mineralization data features in the first associated data feature set; if the quantitative ratio is less than or equal to a preset ratio standard, determine that the mineralization characterization value is not corrected, and determine that the mineralization grade of the mining area to be explored is a low mineralization grade; if the quantitative ratio is greater than the preset ratio standard, correct the mineralization characterization value according to the weak mineralization data features appearing in the mining area to be explored.

8. The intelligent mineralization prediction method based on geological big data according to claim 7 is characterized in that: The process of correcting the metallogenic characterization value according to the weak metallogenic data features appearing in the mining area to be explored includes: Record the weak mineralization data features that appear in the mining area to be explored. If the number of the recorded weak mineralization data features is less than or equal to the preset number value, do not correct the mineralization characterization value, and judge that the mineralization grade of the mining area to be explored is a low mineralization grade. If the number of the recorded weak mineralization data features is greater than the preset number value, calculate the reference correction coefficient to correct the mineralization characterization value.

9. The intelligent mineralization prediction method based on geological big data according to claim 8 is characterized in that: Calculate the reference correction factor according to the following formula, set C=m×∑Gl Among them, C is the reference correction coefficient, m is the number of the weak mineralization data features appearing in the mining area to be explored, Gl is the probability of the occurrence of the lth weak mineralization data feature, and the value of l is 1 to m.

10. The intelligent mineralization prediction method based on geological big data according to claim 9 is characterized in that: The mineralization characterization value is corrected according to the reference correction coefficient, and when the corrected mineralization characterization value is greater than the mineralization reference value, the mineralization grade of the mining area to be explored is determined to be a high mineralization grade.

Images