Method and device for prospecting distant scene quantitative analysis based on geoscience big data

Abstract

The invention relates to a geoscience big data-based prospecting distant scene quantitative analysis method and device, belongs to the technical field of mineral resource exploration, and solves the problem that the delineated prospecting distant scene is not accurate enough due to the fact that the level of the prospecting distant scene is judged excessively according to some subjective knowledge in an existing prospecting prediction scheme. The prospecting long-range quantitative analysis method comprises the following steps: extracting a prospecting exploration mark from a geoscience big data spatial database, dividing a training area into grid units according to a target ore deposit type, and selecting a model unit from the grid units; and training a prediction model according to the prospecting exploration mark of the model unit, calculating the metallogenic probability and prospecting prospecting areas of each grid unit in the target working area based on the prediction model, and grading the prospecting prospecting areas based on the metallogenic probability of each grid unit. Based on this, the method can provide a quantitative basis for the delineation of the prospecting area, avoids the influence of subjective factors, and improves the delineation precision of the prospecting area.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of mineral resources exploration, and particularly relates to a prospecting perspective quantitative analysis method and system based on geosciences big data. BACKGROUND

[0002] Prospecting prediction refers to a method and technology of applying geological metallogenic basic theory, combining geological, mineral, geophysical, geochemical, remote sensing and other comprehensive information, summarizing prediction elements, making analogy prediction, proposing metallogenic perspective areas, guiding exploration work, and discovering industrial deposits.

[0003] In prospecting prediction, metallogenic probability is often used to represent the metallogenic favorable degree of different grid units in the target prospecting area, and a certain threshold of the metallogenic probability is often used to determine the boundary of the prospecting perspective area, and further determine the grade of the prospecting perspective area. The traditional prospecting prediction scheme determines a certain threshold of the metallogenic probability as the boundary of the prospecting perspective area by using various calculation methods, such as 85% of the cumulative probability, mean plus two times of the standard deviation, and the grade of the prospecting perspective area is often determined by subjective methods, such as whether containing discovered deposits (points), whether containing more important prospecting exploration marks or whether the prospecting exploration marks covering the prospecting perspective area are complete.

[0004] However, the above prospecting prediction scheme either determines the boundary of the prospecting perspective area on the premise that the metallogenic probability (or its transformation) conforms to the normal distribution, or determines the grade of the prospecting perspective area too much depending on subjective understanding, resulting in that the circled prospecting perspective area is not accurate enough. SUMMARY

[0005] In view of the above analysis, the embodiments of the present application aim to provide a prospecting perspective quantitative analysis method and device based on geosciences big data, to solve the problem that the grade of the prospecting perspective area is determined too much depending on subjective understanding in the existing prospecting prediction scheme, resulting in that the circled prospecting perspective area is not accurate enough.

[0006] The embodiments of the present application provide a prospecting perspective quantitative analysis method based on geosciences big data, comprising the steps of:

[0007] Extracting prospecting exploration marks from a geosciences big data spatial database, and dividing grid units for a target deposit type in a training area, and selecting model units from the grid units; wherein the model units are divided into ore-bearing units and non-ore-bearing units, and the absence or negative anomaly of the prospecting exploration marks is the bottom line selection standard of the non-ore-bearing units;

[0008] Training a prediction model according to the prospecting exploration marks of the model units; wherein the prediction model determines the boundary of the prospecting perspective area based on a set Youden index to circulate the prospecting perspective area; the set Youden index is as follows:

[0009] P 最优= [Max(TPR / (FPR+ε))→P];

[0010] wherein, P 最优 is the optimal metallogenic probability of determining the boundary of the prospecting area, TPR represents the ratio of correctly predicting the model unit with ore to all ore units, FPR represents the ratio of incorrectly predicting the model unit with ore to all non-ore units, ε is a given value, and → represents the mapping relationship between the given Youdeng index and the metallogenic probability;

[0011] Based on the prediction model, the metallogenic probability of each grid unit in the target working area and the prospecting area are calculated, and the prospecting area is classified based on the metallogenic probability of each grid unit.

[0012] The prospecting area quantitative analysis method based on geosciences big data in the embodiment of the application extracts the prospecting exploration markers from the geosciences big data spatial database, divides the grid units for the target deposit type in the training area, and selects the model units from the grid units. The prediction model is trained according to the prospecting exploration markers of the model units, the metallogenic probability of each grid unit in the target working area and the prospecting area are calculated based on the prediction model, and the prospecting area is classified based on the metallogenic probability of each grid unit. Based on this, a quantitative basis is provided for the delineation of the prospecting area, the influence of subjective factors is avoided, and a data model basis is provided for the improvement of the delineation accuracy.

[0013] As one of the optional embodiments, the process of extracting the prospecting exploration markers from the geosciences big data spatial database includes the following steps:

[0014] The prospecting exploration markers are extracted according to the ore-controlling elements in priority;

[0015] The prospecting exploration markers are extracted according to the ore mineral assemblage in the second place;

[0016] The prospecting exploration markers are extracted according to the alteration mineral assemblage in the end.

[0017] As one of the optional embodiments, the process of selecting the model units from the grid units includes the following steps:

[0018] The grid units without discovered deposits in the selected area of the training area are selected as non-ore units in priority;

[0019] The grid units that are impossible to form the target deposit type or have formed other deposit types with opposite characteristics of the target deposit type are selected as non-ore units in the second place;

[0020] The grid units far away from the known deposits or the grid units with missing or negative anomalies of the prospecting exploration markers are selected as non-ore units in the end.

[0021] As one of the optional embodiments, the process of classifying the prospecting area based on the metallogenic probability of each grid unit is as follows:

[0022]

[0023] wherein, is the mean of the metallogenic probability in the ore-prospecting prospective area, n is the total number of grid cells in the ore-prospecting prospective area, p i is the metallogenic probability of the i-th grid cell in the ore-prospecting prospective area; the interval division based on the mean is used to grade the ore-prospecting prospective area.

[0024] As one of the optional embodiments, the process of establishing the geoscience big data spatial database includes the steps of:

[0025] obtaining geoscience big data data including geology, mineral resources, geophysical prospecting, geochemical prospecting, heavy sand and remote sensing;

[0026] transforming the geoscience big data data of different types of coordinate systems into a unified spatial coordinate system, and converting various types of data formats in the geoscience big data data into a unified data type format, to generate the geoscience big data spatial database.

[0027] As one of the optional embodiments, the process of calculating the metallogenic probability of each grid cell in the target working area based on the prediction model includes the steps of:

[0028] calculating the metallogenic probability based on the ore-prospecting exploration indicators of the grid cell; wherein the prediction model is a random forest model.

[0029] As one of the optional embodiments, it further includes the steps of:

[0030] recording the ore-prospecting prospective area and the ore-prospecting exploration result of the target working area;

[0031] establishing a mapping relationship between the ore-prospecting prospective area and the ore-prospecting exploration result based on the AI model;

[0032] adjusting the association between the mean and the grading based on the mapping relationship.

[0033] The application also provides a geoscience big data-based ore-prospecting prospective quantitative analysis device, comprising:

[0034] a data processing module for extracting ore-prospecting exploration indicators from the geoscience big data spatial database, dividing grid cells for the target ore deposit type in the training area, and selecting model cells from the grid cells; wherein the model cells are divided into ore-bearing cells and non-ore-bearing cells, and the absence or negative anomaly of the ore-prospecting exploration indicators is the bottom line selection standard for the non-ore-bearing cells;

[0035] a model training module for training a prediction model according to the model cell ore-prospecting exploration indicators; wherein the prediction model determines the ore-prospecting prospective area boundary based on a set Youden index to delineate the ore-prospecting prospective area; the set Youden index is as follows:

[0036] P 最优 = [Max (TPR / (FPR + ε)) → P] ;

[0037] Wherein, P 最优 is the optimal metallogenic probability of determining the prospecting prospect area boundary, TPR represents the ratio of correctly predicting the model unit with ore to all ore units, FPR represents the ratio of incorrectly predicting the model unit with ore to all non-ore units, ε is a given value, and → represents the mapping relationship of setting the Youden index and the metallogenic probability.

[0038] The analysis and prediction module is configured to calculate the metallogenic probability of each grid unit in the target working area and the prospecting prospect area based on the prediction model, and grade the prospecting prospect area based on the metallogenic probability of each grid unit.

[0039] The prospecting prospect quantitative analysis device based on geosciences big data according to the embodiments of the present application extracts the prospecting exploration markers from the geosciences big data spatial database, divides the grid units for the target ore deposit type in the training area, and selects the model units from the grid units. The prediction model is trained according to the prospecting exploration markers of the model units, the metallogenic probability of each grid unit in the target working area and the prospecting prospect area are calculated based on the prediction model, and the prospecting prospect area is graded based on the metallogenic probability of each grid unit. Based on this, a quantitative basis is provided for delineating the prospecting prospect area, the influence of subjective factors is avoided, and a data model basis is provided for improving the delineation accuracy.

[0040] At least one embodiment of the present application also provides a data control device, comprising:

[0041] One or more memories non-transiently storing computer executable instructions;

[0042] One or more processors configured to run the computer executable instructions, wherein the computer executable instructions, when run by the one or more processors, implement the prospecting prospect quantitative analysis method based on geosciences big data according to any embodiment of the present application.

[0043] The data control device described above performs fusion processing on multi-source national space data based on spatial data calculation rules, constructs a standardized, data-driven minimum functional statistical standard unit division model, can generate minimum functional statistical standard units with unified spatial granularity and adapt to multi-scale planning targets, significantly improves the spatial support accuracy and analysis application basis of national space planning implementation, and helps to realize the scientization, dynamicization and intelligentization of national space governance.

[0044] The at least one embodiment of the application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by a processor to implement the method for quantitative analysis of ore-prospecting prospects based on geoscience big data according to any embodiment of the application.

[0045] The non-transitory computer-readable storage medium described above performs fusion processing on multi-source land space data based on a spatial data operation rule, constructs a standardized and data-driven minimum functional statistical standard unit division model, can generate minimum functional statistical standard units with unified spatial granularity and adapt to multi-scale planning targets, significantly improves the spatial support accuracy and analysis application basis of land space planning implementation, and helps to realize the scientization, dynamicization and intelligentization of land space governance. BRIEF DESCRIPTION OF DRAWINGS

[0046] Figure 1 A flowchart of the method for quantitative analysis of ore-prospecting prospects based on geoscience big data according to an application embodiment;

[0047] Figure 2 A quantitative analysis process schematic diagram of a quantitative delineation example of a certain deposit type ore-prospecting prospect area provided by the application;

[0048] Figure 3 A module structure diagram of the device for quantitative analysis of ore-prospecting prospects based on geoscience big data according to an application embodiment;

[0049] Figure 4 A schematic block diagram of a data control device provided by the application;

[0050] Figure 5 A schematic diagram of a non-transitory computer-readable storage medium provided by the application. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical scheme and advantages of the embodiments of the application more clear, the technical scheme of the embodiments of the application will be clearly and completely described below in combination with the drawings of the embodiments of the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments. Based on the described embodiments of the application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the application.

[0052] Unless otherwise defined, technical terms or scientific terms used in the present application shall have the ordinary meanings as understood by one of ordinary skill in the art to which this application pertains. The words "first", "second", and similar words of distinction do not by themselves indicate any order, quantity, or importance, but are used to distinguish one element from another. The words "include" or "contain" and similar words of inclusion are intended to encompass the elements listed thereafter and their equivalents, without excluding other elements. The words "connect" or "connected" and similar words of connection are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The words "upper", "lower", "left", "right", and the like are used only to indicate relative positional relationships, which can change when the absolute positions of the described objects change.

[0053] In order to keep the following description of the embodiments of the present application clear and brief, detailed descriptions of some known functions and known components are omitted.

[0054] The embodiments of the present application provide a quantitative analysis method for ore prospecting vision based on geosciences big data.

[0055] Figure 1 The quantitative analysis method for ore prospecting vision based on geosciences big data of an embodiment of the present application is shown in a flowchart as Figure 1 The quantitative analysis method for ore prospecting vision based on geosciences big data of an embodiment of the present application includes steps S100 to S102:

[0056] S100, extracting ore prospecting marks from a geosciences big data spatial database, and dividing grid units for a target ore deposit type in a training area, and selecting model units from the grid units; wherein the model units are divided into ore-bearing units and non-ore-bearing units, and the absence or negative anomaly of the ore prospecting marks is the bottom line selection standard for the non-ore-bearing units;

[0057] S101, training a prediction model according to the ore prospecting marks of the model units; wherein the prediction model determines the boundary of the ore prospecting vision area based on a set improved Youden index to delineate the ore prospecting vision area; the set improved Youden index is as follows:

[0058] P 最优 = [Max(TPR / (FPR+ε))→P];

[0059] wherein P 最优 is the optimal mineralization probability for determining the boundary of the ore prospecting vision area, TPR represents the ratio of correctly predicted ore-bearing model units to all ore-bearing units, FPR represents the ratio of incorrectly predicted ore-bearing model units to all non-ore-bearing units, ε is a given value, and → represents the mapping relationship between the set improved Youden index and the mineralization probability;

[0060] S102, calculate the metallogenic probability of each grid unit in the target working area and the prospecting prospective area based on the prediction model, and grade the prospecting prospective area based on the metallogenic probability of the grid unit in each prospecting prospective area.

[0061] In the embodiments of the present application, there are a training area and a target working area. The training area is used to provide a training data set for the prediction model, and the target working area applies the prediction model to delineate and grade the prospecting prospective area. The data source of the training area is a geoscience big data spatial database.

[0062] As one of the preferred embodiments, the geoscience big data data including geology, mineral resources, geophysical prospecting, geochemical prospecting, heavy sand and remote sensing are obtained; the geoscience big data data in different types of coordinate systems are converted into a unified spatial coordinate system; various types of data formats are converted into a unified data type format, and finally a geoscience big data spatial database is generated.

[0063] Preferably, in the extraction process of the prospecting exploration mark, sequential extraction is used instead of traditional targeted extraction to establish a prospecting exploration mark system. Selecting the prospecting exploration mark from the geoscience big data spatial database is an important step for regional prospecting prediction of the target deposit type. Based on the trinity characteristics of the target deposit type, i.e. ore-controlling factors, ore mineral assemblage and altered mineral assemblage, the prospecting exploration mark is sequentially extracted from the geoscience big data spatial database, and then the prospecting exploration mark system is constructed. It should be noted that the sequential extraction is determined by the logical order of the formation of the deposit, because the ore-controlling factors are the prerequisite for the formation of the deposit, the ore mineral assemblage is the result of the formation of the deposit, and the altered mineral assemblage is a kind of associated geological phenomenon of the formation of the deposit. The construction process of the prospecting exploration mark system is illustrated by taking a porphyry-skarn-low temperature hydrothermal vein type gold-iron-copper-lead-zinc polymetallic deposit in a certain area as an example. It should be particularly noted that other target deposit types can also be constructed according to this idea, and the difference lies in the different characteristics of the target deposit type, and the constructed prospecting exploration mark system will also be different.

[0064] As one of the preferred embodiments, the process of extracting the prospecting exploration mark from the geoscience big data spatial database includes the following steps:

[0065] The prospecting exploration mark is extracted in priority according to the ore-controlling factors;

[0066] The prospecting exploration mark is extracted in the second place according to the ore mineral assemblage;

[0067] The prospecting exploration mark is finally extracted according to the altered mineral assemblage.

[0068] Firstly, the prospecting exploration markers are extracted according to the ore-controlling elements. The ore-controlling elements refer to the geological elements that control the formation of the ore deposit, mainly including strata, structures and magmatic rocks, etc. Different target ore deposit types have different ore-controlling elements. For example, the porphyry-skarn-low temperature hydrothermal vein type deposit is mainly controlled by porphyry intrusion and carbonate strata, so the porphyry and carbonate strata in the geological map are extracted from the geoscience big data spatial database as the magmatic rock and strata markers for prospecting exploration; in addition, the deposit is also controlled by intrusive contact structure, so the intrusive contact structure in the geological map is extracted from the geoscience big data spatial database as the structure marker for prospecting exploration.

[0069] Secondly, the prospecting exploration markers are extracted according to the ore mineral assemblage. The ore mineral assemblage refers to the useful mineral collection in the deposit (ore body), and different target ore deposit types have different ore mineral assemblages. For example, the ore mineral assemblage produced by the porphyry-skarn-low temperature hydrothermal vein type deposit mainly includes gold-bearing pyrite, magnetite, chalcopyrite, galena and sphalerite, etc. Among them, one important feature of magnetite is that it shows strong magnetism compared with the surrounding rock, so the magnetic anomaly in the geophysical exploration data is extracted from the geoscience big data spatial database as the geophysical exploration marker for prospecting exploration; the ore-forming elements gold, copper, lead and zinc, etc. of gold-bearing pyrite, chalcopyrite, galena and sphalerite can be relatively enriched in stream sediments under the action of surface weathering, so the element anomaly of gold, copper, lead and zinc, etc. in the geochemical exploration data in the geoscience big data spatial database can be extracted as the geochemical exploration marker for prospecting exploration; in addition, the gold element is very stable in the hypergene state, forming natural heavy sand minerals, so the gold natural heavy sand in the natural heavy sand data in the geoscience big data spatial database is extracted as the natural heavy sand marker for prospecting exploration.

[0070] Finally, the prospecting exploration markers are extracted according to the altered mineral assemblage. The altered mineral assemblage refers to the specific mineral collection formed after the original rock and fluid exchange under the action of hydrothermal fluid, and different target ore deposit types have different altered mineral assemblages. For example, the porphyry-skarn-low temperature hydrothermal vein type deposit, due to the hydrothermal alteration, will also form altered minerals (such as sericite, chlorite and epidote) rich in water and iron, so the hydroxyl or iron stain representing the altered minerals can be extracted from the remote sensing data (such as ETM or ASTER) in the geoscience big data spatial database as the remote sensing marker for prospecting exploration.

[0071] The above various prospecting exploration markers are comprehensively combined to successfully build a multi-element exploration marker system.

[0072] On the basis of dividing the grid cells of the training area for the target deposit type, a sufficient number of model cells need to be selected for training the prediction model. The model cells include ore-bearing cells and non-ore-bearing cells. The selection method of the ore-bearing cells is relatively mature, that is, those grid cells where deposits have been found through detailed exploration are selected as ore-bearing cells. In the case of insufficient ore-bearing cells, various expansion methods can be used to increase the number of ore-bearing cells, such as buffer zone analysis or conditional generative adversarial networks.

[0073] However, the selection of non-ore-bearing cells is a relatively complex task. Traditionally, only those grid cells where no deposits have been found through detailed exploration are identified as non-ore-bearing cells. However, such grid cells are very limited, often difficult to meet the number requirements of the prediction model for non-ore-bearing cells, and further channels for selecting non-ore-bearing cells need to be expanded. The present application proposes a "three-level selection method" to select non-ore-bearing cells, including the following steps:

[0074] First, grid cells where no deposits have been found in the selected area of the training area are selected as non-ore-bearing cells.

[0075] Second, grid cells where the target deposit type is impossible to form or grid cells where other deposit types that are opposite to the target deposit type have been formed are selected as non-ore-bearing cells.

[0076] Finally, grid cells far away from known deposits or grid cells where prospecting exploration markers are missing or negative anomalies are selected as non-ore-bearing cells.

[0077] First level, first of all based on the degree of work, in areas with high degree of exploration (generally reaching the level of general survey and above), grid cells where no deposits have been found are selected as non-ore-bearing cells in the model cells.

[0078] Second level, in the case where the number of non-ore-bearing cells cannot be met in the first level, grid cells where the target deposit type is impossible to form (such as granite areas where sedimentary mineral deposits cannot be formed) or grid cells where other deposit types that are opposite to the target deposit type have been formed (such as basic-ultrabasic rock type deposits which cannot form deposits related to granite) are selected as non-ore-bearing cells in the model cells.

[0079] Third level, in the case where the number of non-ore-bearing cells cannot be met in the second level, based on the basic fact that the formation of a deposit is a small probability and its output has a clustering phenomenon, grid cells far away from known deposits or all prospecting exploration markers missing (or negative anomalies) overlapping areas are selected as non-ore-bearing cells.

[0080] Based on this, mark the ore-bearing unit and non-ore unit, and train the prediction model. Among them, the optimal random forest prediction model is selected and trained. There are various model methods for calculating the probability of mineralization, such as: feature analysis method, evidence weight method, logistic regression, ensemble learning, deep learning, etc. In this embodiment, the random forest algorithm is used to train and optimize based on the prospecting exploration markers of the model unit.

[0081] At the same time, based on the improved Youden index, the threshold value (optimal mineralization probability) for delineating the prospecting area is determined. After obtaining the optimal model, a reasonable threshold value is determined as the boundary value of the prospecting area. The improved Youden index is used to determine the boundary (threshold value) of the prospecting area in this embodiment. The formula of the improved Youden index is as follows:

[0082] P 最优 =[Max(TPR / (FPR+ε))→P];

[0083] Wherein, P 最优 is the optimal mineralization probability for determining the boundary (threshold value) of the prospecting area, TPR is the true positive rate, which in this embodiment specifically represents the ratio of correctly predicted ore-bearing model units to all ore-bearing model units, FPR is the false positive rate, which is the ratio of incorrectly predicted ore-bearing model units to all non-ore units, and ε is a very small number, which is generally taken as 0.001 to prevent the denominator from being zero. Max(TPR / (FPR+ε)) is the improved Youden index, P is the mineralization probability corresponding to the improved Youden index, and → represents the mapping relationship between the improved Youden index and the mineralization probability. The mineralization probability corresponding to the maximum value of the ratio of the true positive rate to the false positive rate is the optimal threshold value for determining the boundary of the prospecting area. The range of the improved Youden index is 1 to +∞, which can further highlight the Youden index and prevent the traditional Youden index from being difficult to determine the corresponding optimal mineralization probability when the value changes slightly. It can amplify the difference between the prospecting exploration marker order extraction system and the fine acquisition of non-ore units, so as to more accurately determine the boundary of the prospecting area.

[0084] As one of the optional embodiments, the process of the prediction model calculating the probability of mineralization of each grid unit in the target working area includes the following steps:

[0085] Based on the grid unit prospecting exploration marker, the probability of mineralization is calculated; wherein the prediction model is a random forest model.

[0086] The prospecting area is delineated in the target working area. Based on the obtained random forest model, the probability of mineralization of all grid units in the target working area is calculated according to the prospecting exploration markers of the grid units, and the prospecting area is delineated based on the optimal mineralization probability.

[0087] The metallogenic probability mean value is used to classify the prospecting prospective area, and the level of the prospecting prospective area represents the advantages and disadvantages of the metallogenic conditions. However, after the prospecting prospective area is delineated, the prospecting prospective area is composed of multiple grid cells, and each grid cell is often composed of different metallogenic probability values. Therefore, the level of the prospecting prospective area lacks a unified index. Therefore, in the embodiment of the present application, the mean value of the metallogenic probability of all grid cells in the prospecting prospective area is used as a quantitative basis for classifying the level of the prospective area, and is also a quantitative index for selecting a priority exploration target. The calculation formula is as follows:

[0088]

[0089] wherein, is the mean value of the metallogenic probability in the prospecting prospective area, n is the total number of grid cells in the prospecting prospective area, and p i is the metallogenic probability of the i-th grid cell in the prospecting prospective area.

[0090] wherein, multiple intervals are divided according to the interval in which the mean value is located, and each interval corresponds to a level of the prospecting prospective area. In actual engineering, the mean value can be divided into a certain number of levels according to the granularity of prospecting, which is not limited here.

[0091] As one of the optional embodiments, it further includes the following steps:

[0092] Recording the prospecting prospective area and the prospecting exploration result of the target working area;

[0093] Establishing a mapping relationship between the prospecting prospective area and the prospecting exploration result based on the AI model;

[0094] Adjusting the association between the mean value and the classification based on the mapping relationship.

[0095] According to the recorded prospecting prospective area and the prospecting exploration result of the historical analysis, the mapping relationship between the prospecting prospective area and the prospecting exploration result is established, the association between the mean value and the classification is adjusted, the number of levels of the classification is adjusted through the AI model, the precision standard of the classification is quantified, and the quantitative analysis data model of the prospecting prospective area is integrated and automatically deployed.

[0096] Preferably, the prospecting prospective area and the prospecting exploration result are input into the AI model in the form of a graphical result, and the mapping relationship is established according to the graphical approximation.

[0097] Figure 2 A quantitative analysis process diagram of a certain area and a certain deposit type prospecting prospective area quantitative delineation example provided by the present application is shown in Figure 2 The present application is based on the quantitative delineation of the prospecting prospective area classification, and a label map (real ore and non-ore unit) of the training area is taken as an example, as shown in Figure 2Fig. 2a in the figure; the training area is predicted according to the optimal random forest model, and the mineralization probability is obtained, see Fig. 2b in the figure; Figure 2 Fig. 2c in the figure; the improved Youden index changes with the mineralization probability, and the improved Youden index is obtained according to the label map and the mineralization probability map of the training area, see Fig. 2d in the figure; Figure 2 Fig. 2e in the figure, wherein the maximum Youden index 603.3777 corresponds to the mineralization probability of 0.9686, and 0.9686 is the optimal mineralization probability, which can be used as the threshold for delineating the prospecting area; the optimal random forest model is used to predict the target prediction area, and the mineralization probability map is calculated, see Fig. 2f in the figure; Figure 2 Fig. 2g in the figure; the prospecting area is delineated according to the optimal mineralization probability in the target prediction area based on the mineralization probability map, see Fig. 2h in the figure; Figure 2 Fig. 2i in the figure; the mean value of the mineralization probability of each grid cell in each prospecting area in the target prediction area is determined by calculation, see Fig. 2j in the figure; Figure 2 Fig. 2k in the figure; on the basis of the above mean value, 0.98 is taken as the boundary, and the level is determined according to the mean value, the mineralization probability greater than or equal to 0.98 is the A-type prospecting area, and the mineralization probability less than 0.98 is the B-type prospecting area, see Fig. 2l in the figure. Figure 2

[0098] It should be particularly noted that the main purpose of setting the prospecting area category is to serve further prospecting work, and the limited manpower, financial resources and time are preferentially deployed in the most promising areas; relatively speaking, the mineralization probability of the A-type prospecting area is high, and the prospecting prospect is better than that of the B-type prospecting area, so generally the A-type prospecting area is preferentially deployed for exploration. However, this does not mean that the B-type prospecting area has no prospecting prospect, and after the A-type prospecting area is deployed, if there is still part of manpower, financial resources and time, the B-type prospecting area can be further deployed for prospecting work. Therefore, A and B are relative.

[0099] The prospecting prospect quantitative analysis method based on geosciences big data in the embodiment of the application extracts the prospecting exploration marks from the geosciences big data spatial database, divides the grid cells for the target deposit type in the training area, and selects the model cells from the grid cells. The prediction model is trained according to the prospecting exploration marks of the model cells, the mineralization probability of each grid cell in the target working area is calculated based on the prediction model, the prospecting area is calculated, and the prospecting area is graded based on the mineralization probability of each grid cell. Based on this, a quantitative basis is provided for delineating the prospecting area, the influence of subjective factors is avoided, and a data model basis is provided for improving the delineation accuracy.

[0100] The embodiment of the application further provides a prospecting prospect quantitative analysis device based on geosciences big data.

[0101] Figure 3 ​A module structure diagram of a prospecting vision quantitative analysis device based on geosciences big data is shown in FIG. 1. Figure 3 As shown in FIG. 1, the prospecting vision quantitative analysis device based on geosciences big data includes:

[0102] The data processing module 100 is configured to extract prospecting exploration marks from a geosciences big data spatial database, divide grid cells for a target deposit type in a training area, and select model cells from the grid cells; wherein the model cells are divided into ore-bearing cells and non-ore-bearing cells, and the absence or negative anomaly of the prospecting exploration marks is a bottom line selection criterion for the non-ore-bearing cells.

[0103] The model training module 101 is configured to train a prediction model according to the prospecting exploration marks of the model cells; wherein the prediction model determines the boundary of a prospecting vision area based on a set Youden index to delineate the prospecting vision area; and the improved Youden index is set as follows:

[0104] P 最优 = [Max(TPR / (FPR+ε))→P];

[0105] wherein P 最优 is the optimal metallogenic probability for determining the boundary of the prospecting vision area, TPR represents the ratio of correctly predicted ore-bearing model cells to all ore-bearing cells, FPR represents the ratio of incorrectly predicted ore-bearing model cells to all non-ore-bearing cells, ε is a given value, and → represents the mapping relationship between the set Youden index and the metallogenic probability.

[0106] The analysis and prediction module 102 is configured to calculate the metallogenic probability of each grid cell in a target working area and the prospecting vision area based on the prediction model, and grade the prospecting vision area based on the metallogenic probability of each grid cell.

[0107] The prospecting vision quantitative analysis device based on geosciences big data extracts prospecting exploration marks from a geosciences big data spatial database, divides grid cells for a target deposit type in a training area, and selects model cells from the grid cells. The prediction model is trained according to the prospecting exploration marks of the model cells, the metallogenic probability of each grid cell in a target working area and the prospecting vision area are calculated based on the prediction model, and the prospecting vision area is graded based on the metallogenic probability of each grid cell. Based on this, a quantitative basis is provided for delineating the prospecting vision area, the influence of subjective factors is avoided, and a data model basis is provided for improving the delineation accuracy.

[0108] At least one embodiment of the present application also provides a data control device. Figure 4 A schematic block diagram of a data control device provided by at least one embodiment of the present application is shown in FIG. 2. For example, as shown in FIG. 2, the data control device includes: Figure 4As shown, the data control apparatus 20 can include one or more memories 200 and one or more processors 201. The memory 200 is configured to store computer-executable instructions non-transitorily; and the processor 201 is configured to execute the computer-executable instructions, which can cause the processor 201 to perform one or more steps of the method for quantitative analysis of ore-prospecting prospect based on geosciences big data according to any of the embodiments of the present application when the computer-executable instructions are executed by the processor 201.

[0109] For the specific implementation of each step of the method for quantitative analysis of ore-prospecting prospect based on geosciences big data and related explanations, please refer to the related content in the above embodiments of the method for quantitative analysis of ore-prospecting prospect based on geosciences big data, which will not be repeated here. It should be noted that, Figure 4 The components of the data control apparatus 20 shown are exemplary only, and are not intended to be limiting. The data control apparatus 20 can also have other components according to actual application needs.

[0110] In one embodiment, the processor 201 and the memory 200 can communicate with each other directly or indirectly. For example, the processor 201 and the memory 200 can communicate with each other through a network connection. The network can include a wireless network, a wired network, and / or any combination of a wireless network and a wired network, and the type and function of the network are not limited herein. For another example, the processor 201 and the memory 200 can also communicate with each other through a bus connection. The bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. For example, the processor 201 and the memory 200 can be disposed at a remote data server end (cloud end) or a distributed energy system end (local end), or can be disposed at a client end (e.g., a mobile device such as a mobile phone, etc.). For example, the processor 201 can be a central processing unit (CPU), a tensor processing unit (TPU), or a graphics processing unit (GPU), etc. which has data processing capability and / or instruction execution capability, and can control other components in the data control apparatus 20 to perform desired functions. The central processing unit (CPU) can be of X86 or ARM architecture, etc.

[0111] In one of the embodiments, the memory 200 can include any combination of one or more computer program products. The computer program product can include various forms of computer-readable storage media, for example, volatile memory and / or non-volatile memory. The volatile memory, for example, can include random access memory (RAM), cache memory, and / or the like. The non-volatile memory, for example, can include read only memory (ROM), hard disk, erasable programmable read only memory (EPROM), compact disk read only memory (CD-ROM), USB memory, flash memory, and / or the like. One or more computer-executable instructions can be stored on the computer-readable storage medium. The processor 201 can execute the computer-executable instructions to implement various functions of the data control apparatus 20. Various application programs and various data used and / or generated by the application programs can also be stored in the memory 200.

[0112] It should be noted that the data control apparatus 20 can achieve similar technical effects as the aforementioned method for quantitatively analyzing ore prospecting prospects based on geosciences big data. The repeated parts will not be described again.

[0113] At least one embodiment of the present application also provides a non-transitory computer-readable storage medium. Figure 5 A schematic diagram of a non-transitory computer-readable storage medium provided by at least one embodiment of the present application is shown in FIG. 30. For example, as shown in FIG. 30, one or more computer-executable instructions 301 can be non-transitorily stored on the non-transitory computer-readable storage medium 30. For example, when the computer-executable instructions 301 are executed by a computer, the computer can be caused to perform one or more steps of the method for quantitatively analyzing ore prospecting prospects based on geosciences big data according to any embodiment of the present application. Figure 5

[0114] In one of the embodiments, the non-transitory computer-readable storage medium 30 can be applied in the data control apparatus 20 described above. For example, the non-transitory computer-readable storage medium 30 can be the memory 200 in the data control apparatus 20.

[0115] In one of the embodiments, the description of the non-transitory computer-readable storage medium 30 can refer to the description of the memory 200 in the embodiments of the data control apparatus 20. The repeated parts will not be described again.

[0116] It should be noted that the memory 200 stores different non-transitory computer-executable instructions. The data control apparatus 20 corresponds to a firmware upgrade apparatus. When the computer-executable instructions are executed by the processor 201, the processor 201 can be caused to perform one or more steps of the method for quantitatively analyzing ore prospecting prospects based on geosciences big data according to any embodiment of the present application.

[0117] ​For the present application, the following points need to be explained:

[0118] (1) The embodiment drawings of the present application only involve the structures involved in the embodiments of the present application, and other structures can refer to the general design.

[0119] (2) For the sake of clarity, the thickness and size of the layer or structure are exaggerated in the drawings used to describe the embodiments of the present application. It can be understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, it can be "directly" on or under the other element, or there can be an intermediate element.

[0120] (3) In the case of no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other to obtain new embodiments. The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto, and the protection scope of the present application should be subject to the protection scope of the claims.

[0121] The technical features of the above embodiments can be combined arbitrarily, and in order to make the description simple, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0122] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the patent scope of the application. It should be pointed out that for ordinary skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the protection scope of the present application. Therefore, the patent protection scope of the present application should be subject to the appended claims.

Claims

1. A quantitative analysis method for prospecting prospects based on geological big data, characterized by: Including steps: Extracting prospecting and exploration markers from a geoscience big data spatial database, dividing a training area into grid cells according to the target ore deposit type, and selecting model cells from the grid cells; wherein the model cells are divided into ore-bearing cells and non-ore-bearing cells, and the absence or negative anomaly of the prospecting and exploration markers serves as a baseline selection criterion for the non-ore-bearing cells; A prediction model is trained based on the prospecting and exploration indicators of the model unit; wherein the prediction model determines the boundary of the prospecting area based on a set Youden index to delineate the prospecting area; the set Youden index is as follows: P 最优 =[Max(TPR / (FPR+ε))→P]; Among them, P 最优 To determine the optimal mineralization probability of the prospecting area boundary; TPR represents the ratio of the model unit correctly predicted to have mineralization to all mineralization units; FPR represents the ratio of the model unit incorrectly predicted to have mineralization to all non-ore-forming units; ε is a given value; → represents the mapping relationship between the set Youden index and the mineralization probability; The metallogenic probability and prospecting areas of each grid unit in the target working area are calculated based on the prediction model, and the prospecting areas are classified based on the metallogenic probability of each grid unit.

2. The method for quantitative analysis of prospecting prospects based on geological big data according to claim 1, characterized in that: The process of extracting prospecting and exploration indicators from the geoscience big data spatial database includes the following steps: Prioritize the extraction of mineral exploration signs based on mineral-controlling factors; Secondly, the prospecting and exploration signs are extracted based on the ore mineral combination; Finally, the prospecting and exploration indicators are extracted based on the altered mineral combination.

3. The method for quantitative analysis of prospecting prospects based on geological big data according to claim 1 or 2, characterized in that: The process of selecting model units from the grid units comprises the steps of: Prioritizing grid cells in which no mineral deposits are found in the selected area of ​​the training area as non-mineralized cells; Secondly, grid cells that are unlikely to form the target deposit type or grid cells that have formed other deposit types that are opposite to the target deposit type are regarded as non-ore cells; Finally, grid cells far away from known mineral deposits or grid cells with missing or negative anomalies of prospecting and exploration signs are regarded as non-ore-bearing cells.

4. The method for quantitative analysis of prospecting prospects based on geological big data according to any one of claims 1 to 3, characterized in that: The process of grading prospecting areas based on the metallogenic probability of each grid cell is as follows: in, is the mean value of the mineralization probability in the prospecting area, n is the total number of grid cells in the prospecting area, and p i is the metallogenic probability of the i-th grid unit in the prospective mineralization area; the prospective mineralization area is classified based on the interval division of the mean value.

5. The method for quantitative analysis of prospecting prospects based on geological big data according to claim 1, characterized in that: The process of establishing the geoscience big data spatial database includes the following steps: Obtaining big geoscience data including geology, mineral resources, geophysical exploration, geochemical exploration, heavy sediments and remote sensing; Transform geoscience big data in different types of coordinate systems into a unified spatial coordinate system, and convert various types of data formats in geoscience big data into a unified data type format to generate a geoscience big data spatial database.

6. The method for quantitative analysis of prospecting prospects based on geological big data according to claim 1, characterized in that: The process of calculating the mineralization probability of each grid cell in the target working area based on the prediction model includes the steps of: The mineralization probability is calculated based on the prospecting and exploration signs of the grid cells; wherein the prediction model is a random forest model.

7. The method for quantitative analysis of prospecting prospects based on geological big data according to claim 5, characterized in that: Also includes the steps: Record prospecting areas and exploration results in the target work area; Establishing a mapping relationship between the prospecting area and the prospecting results based on the AI ​​model; The association between the mean and the grade is adjusted based on the mapping relationship.

8. A quantitative analysis device for prospecting prospects based on geological big data, characterized in that: include: A data processing module is used to extract prospecting and exploration markers from a large geological data spatial database, divide the training area into grid cells according to the target mineral deposit type, and select model units from the grid cells; wherein the model units are divided into mineralized units and non-mineralized units, and the absence or negative anomaly of the prospecting and exploration markers is used as the baseline selection criteria for the non-mineralized units; A model training module is used to train a prediction model based on the prospecting and exploration signs of the model unit; wherein the prediction model determines the boundary of the prospecting area based on a set Youden index to delineate the prospecting area; the set Youden index is as follows: P 最优 =[Max(TPR / (FPR+ε))→P]; Among them, P 最优 To determine the optimal mineralization probability of the prospecting area boundary, TPR represents the ratio of the model unit correctly predicted to have mineralization to all mineralization units, FPR represents the ratio of the model unit incorrectly predicted to have mineralization to all non-ore-forming units, ε is a given value, → represents the mapping relationship between the set Youden index and the mineralization probability; The analysis and prediction module is used to calculate the metallogenic probability and prospecting prospective areas of each grid cell in the target working area based on the prediction model, and to classify the prospecting prospective areas based on the metallogenic probability of each grid cell.

9. A non-transitory computer-readable storage medium, characterized in that The non-transitory computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, implement the quantitative analysis method for prospecting based on geological big data as described in any one of claims 1 to 7.

10. A data control device, characterized in that: include: one or more memories non-transitorily storing computer-executable instructions; One or more processors are configured to run computer-executable instructions, wherein the computer-executable instructions, when run by the one or more processors, implement the quantitative analysis method for prospecting prospects based on geological big data as described in any one of claims 1 to 7.

Images