Liafite pegmatite lithium ore prediction method and system based on li / la element ratio anomaly

Abstract

The application relates to a pegmatite type lithium ore prediction method and system based on Li / La element ratio anomaly, and belongs to the technical field of mineral quantitative prediction. The application solves the problem that key ore-forming indexes are uncertain in current pegmatite type lithium ore prediction, leading to inaccurate extraction and prediction of ore-forming information. A hierarchical lithium ore knowledge graph is constructed based on obtained mine area literature data, knowledge reasoning retrieval is performed based on the hierarchical lithium ore knowledge graph, and key features controlling ore formation are determined. The key features are extracted according to the geochemical data of the predicted mine area, and the ratios of the key features are determined. The ratios are subjected to anomaly determination to predict the ore formation of the pegmatite type lithium ore. Through knowledge reasoning retrieval of the hierarchical lithium ore knowledge graph, key ore-forming indexes are found to predict the ore formation of the pegmatite type lithium ore, so as to more accurately extract ore-forming information and predict the mine area.

Description

Technical Field

[0001] This invention relates to the field of mineral prediction technology, and in particular to a method and system for predicting pegmatite-type lithium deposits based on anomalies in the Li / La element ratio. Background Technology

[0002] With the rapid development of the new energy industry, the demand for lithium, as a key strategic metal, is experiencing explosive growth. Pegmatite-type lithium deposits, characterized by their large scale and high grade, have become a key target for lithium resource exploration and development.

[0003] However, traditional methods for predicting lithium deposits in pegmatite formations have many limitations. On the one hand, some analytical methods based on geochemical data, such as single-element content analysis, do not fully consider the intrinsic relationships between elements and cannot effectively capture key characteristic combinations and ratios reflecting mineralization information. This leads to inaccurate extraction of mineralization information and makes it difficult to achieve quantitative prediction of lithium deposits. On the other hand, in terms of knowledge integration and utilization, previous studies lacked a systematic knowledge management and reasoning mechanism. A large amount of scattered mineral area literature has not been effectively integrated, making it difficult to deeply explore the intrinsic logical relationships between mineralization knowledge, resulting in a lack of strong theoretical support and logical explanation for mineral exploration prediction.

[0004] Therefore, in the current prediction of pegmatite-type lithium deposits, the key mineralization indicators are still uncertain, resulting in insufficient accuracy in the extraction and prediction of mineralization information. Summary of the Invention

[0005] In view of the above analysis, the embodiments of the present invention aim to provide a method and system for predicting pegmatite-type lithium deposits based on the anomaly of the Li / La element ratio, in order to solve the problem that the key mineralization indicators are not yet determined in the current prediction of pegmatite-type lithium deposits in the existing technology, resulting in insufficient accuracy in the extraction and prediction of mineralization information.

[0006] This application provides a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio, including the following steps:

[0007] A hierarchical lithium ore knowledge graph was acquired and constructed based on literature data from the mining area. Knowledge reasoning and retrieval were then performed based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization. The key features are Li and La elements.

[0008] Key features are extracted from the geochemical data of the predicted mining area, and the ratios of the key features are determined.

[0009] Anomaly detection was performed on the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits, and the prediction results were obtained.

[0010] This application's embodiment of the method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies involves acquiring and constructing a hierarchical lithium deposit knowledge graph based on mining area literature data, and performing knowledge reasoning retrieval based on the hierarchical lithium deposit knowledge graph to determine key features controlling mineralization. Key features are extracted from the geochemical data of the predicted mining area, and the ratios of these key features are determined. Anomaly detection is performed on the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits, yielding the prediction results. By using knowledge reasoning retrieval from the hierarchical lithium deposit knowledge graph, key mineralization indicators are found to predict the mineralization of pegmatite-type lithium deposits, facilitating more accurate extraction of mineralization information and prediction of mining areas.

[0011] As one optional embodiment, the process of constructing a hierarchical lithium ore knowledge graph based on mining area literature data includes the following steps:

[0012] Extract information text from mining area literature data and map the information text to a predefined ontology model;

[0013] By integrating the knowledge graphs and ontology models of typical mineral deposits, a hierarchical lithium ore knowledge graph is obtained.

[0014] As one optional embodiment, the construction process of a predefined ontology model includes the following steps:

[0015] Key concept data were extracted from literature data on mining areas to construct a preliminary conceptual model;

[0016] Perform semantic adjustments and iterations on key concept data to establish a class hierarchy structure;

[0017] Define the attributes and relationships of each class, and map instance data to the ontology of the class hierarchy to form an ontology model.

[0018] As one optional embodiment, the process of determining the key features controlling mineralization by performing knowledge reasoning retrieval based on a hierarchical lithium ore knowledge graph includes the following steps:

[0019] Based on COL statements, by setting semantic query paths and structured rules, the association patterns between specific types of nodes are retrieved, and the mineralization combination features and common laws that repeatedly appear in multiple typical mineral deposits are identified, so as to determine the key features of mineralization.

[0020] As one of the alternative embodiments, the key features are the Li element and the La element.

[0021] As one optional embodiment, the process of extracting key features based on geochemical data of the predicted mining area and determining the ratios of the key features includes the following steps:

[0022] Based on the SA multifractal method, the Li / La elemental ratio anomaly in the predicted mining area was extracted, and a Li / La geochemical anomaly map was generated.

[0023] As one optional embodiment, the process of comparing values ​​to determine anomalies in order to predict the mineralization of pegmatite-type lithium deposits includes the following steps:

[0024] Key mineralization-controlling factors and mineral occurrences were extracted by interpreting remote sensing data and then displayed in combination with Li / La geochemical anomaly maps for comparison.

[0025] This application also provides a pegmatite-type lithium deposit prediction system based on Li / La elemental ratio anomalies, including:

[0026] The feature filtering module is used to acquire and construct a hierarchical lithium ore knowledge graph based on mining area literature data, and to perform knowledge reasoning retrieval based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization; wherein, the key features are Li and La elements;

[0027] The feature capture module is used to extract key features based on the geochemical data of the predicted mining area and determine the ratio of key features.

[0028] The mineralization prediction module is used to identify anomalies in the comparison values ​​in order to predict the mineralization of pegmatite-type lithium deposits.

[0029] This application's embodiment of the pegmatite-type lithium deposit prediction system based on Li / La elemental ratio anomalies acquires and constructs a hierarchical lithium deposit knowledge graph based on mining area literature data, and performs knowledge reasoning retrieval based on the hierarchical lithium deposit knowledge graph to determine key features controlling mineralization; extracts key features based on geochemical data of the predicted mining area and determines the ratios of key features; and performs anomaly judgment based on the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits. By using knowledge reasoning retrieval of the hierarchical lithium deposit knowledge graph, key mineralization indicators are found to predict the mineralization of pegmatite-type lithium deposits, thereby facilitating more accurate extraction of mineralization information and prediction of mining areas.

[0030] At least one embodiment of this application also provides a data control device, including:

[0031] One or more memories that store computer-executable instructions non-transitory;

[0032] One or more processors are configured to run computer-executable instructions, wherein the computer-executable instructions are executed by the one or more processors to implement the method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La element ratio according to any embodiment of the present application.

[0033] The aforementioned data control device acquires and constructs a hierarchical lithium ore knowledge graph based on mining area literature data, and performs knowledge reasoning retrieval based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization; it extracts key features based on the geochemical data of the predicted mining area and determines the ratios of key features; it then performs anomaly detection based on the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits. By using knowledge reasoning retrieval of the hierarchical lithium ore knowledge graph, key mineralization indicators are found to predict the mineralization of pegmatite-type lithium deposits, thereby facilitating more accurate extraction of mineralization information and prediction of mining areas.

[0034] At least one embodiment of this application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, implement a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio according to any embodiment of this application.

[0035] The aforementioned non-transitory computer-readable storage medium is used to acquire and construct a hierarchical lithium ore knowledge graph based on mining area literature data. Knowledge reasoning retrieval is then performed based on this knowledge graph to identify key features controlling mineralization. Key features are extracted from the geochemical data of the predicted mining area, and the ratios of these key features are determined. Anomaly detection is performed using these ratios to predict the mineralization of pegmatite-type lithium deposits. By using knowledge reasoning retrieval from the hierarchical lithium ore knowledge graph, key mineralization indicators are identified to predict the mineralization of pegmatite-type lithium deposits, facilitating more accurate extraction of mineralization information and prediction of mining areas. Attached Figure Description

[0036] Figure 1 A flowchart illustrating a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio, according to an embodiment of the application.

[0037] Figure 2 A flowchart of a preferred embodiment of a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio;

[0038] Figure 3 Flowchart for constructing a hierarchical lithium ore knowledge graph;

[0039] Figure 4 This is a schematic diagram of a predefined ontology model;

[0040] Figure 5 A schematic diagram illustrating the bottom-up improvement of the ontology model;

[0041] Figure 6 A schematic diagram of the expanded and improved ontology model.

[0042] Figure 7 This is a schematic diagram of the intermediate-level lithium ore knowledge graph in an embodiment of this application;

[0043] Figure 8 This is a schematic diagram of a complete hierarchical lithium ore knowledge graph according to an embodiment of this application;

[0044] Figure 9 A visual diagram illustrating knowledge reasoning;

[0045] Figure 10 Here is a flowchart of the SA multifractal method;

[0046] Figure 11 A flowchart of a mineralization prediction method for a specific application example;

[0047] Figure 12 A target area delineation and field verification map of the Koelin mining area, as a specific application example;

[0048] Figure 13 Map of Li / La geochemical anomalies in western Sichuan and the Altai region;

[0049] Figure 14 Map showing the Li / La geochemical anomalies in the Gangdise region;

[0050] Figure 15 Map showing Li / La geochemical anomalies in the Himalayas;

[0051] Figure 16 Geochemical diagram of Li / La ratios in the West Kunlun lithium-forming belt and the West Tianshan lithium-forming belt;

[0052] Figure 17 A structural diagram of a pegmatite-type lithium ore prediction system based on Li / La elemental ratio anomalies, according to an embodiment of the application;

[0053] Figure 18 A schematic block diagram of a data control device provided by the present invention;

[0054] Figure 19 This is a schematic diagram of a non-transitory computer-readable storage medium provided by the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the described embodiments of this application without creative effort are within the scope of protection of this application.

[0056] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0057] To keep the following description of the embodiments of this application clear and concise, detailed descriptions of some known functions and components have been omitted.

[0058] This application provides a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La element ratio.

[0059] Figure 1 Here is a flowchart of a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio, as described in one embodiment of the application. Figure 1 As shown, a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio, according to one embodiment of the application, includes steps S100 to S102:

[0060] S100: Acquire and construct a hierarchical lithium ore knowledge graph based on mining area literature data, and perform knowledge reasoning retrieval based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization.

[0061] S101, extract key features based on the geochemical data of the predicted mining area and determine the ratio of key features;

[0062] S102, compare the values ​​to determine anomalies in order to predict the mineralization of pegmatite-type lithium deposits.

[0063] In this embodiment of the application, the literature data of the mining area includes literature and data related to pegmatite-type lithium deposits, such as dissertations and journal articles.

[0064] The hierarchical lithium ore knowledge graph is used to characterize the relationships between concepts and data related to pegmatite-type lithium ore in mining area literature data. Through knowledge reasoning retrieval based on the relationships, the key features in the related concepts and data are identified.

[0065] As a preferred embodiment, Figure 2A flowchart of a preferred embodiment of a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio is shown below. Figure 2 As shown, the process of constructing a hierarchical lithium ore knowledge graph based on mining area literature data in step S100 includes steps S200 and S201:

[0066] S200: Extract information text from mining area literature data and map the information text to a predefined ontology model;

[0067] S201 integrates the knowledge graph and ontology model of typical mineral deposits to obtain a hierarchical lithium mineral knowledge graph.

[0068] The construction of the hierarchical lithium ore knowledge graph is a model system based on ontology models and corpus annotation results. The ontology model is the "underlying architecture" of the hierarchical lithium ore knowledge graph. By clarifying the classes (concepts), relationships, and attributes within the domain, it transforms scattered geological knowledge into a structured form that can be recognized and processed by computers, thereby achieving a standardized expression of knowledge related to "pegmatite-type lithium ore".

[0069] Specifically, by providing explicit class and relationship structures, entities are obtained through class instantiation, and the relationships between entities are obtained through relation instantiation, such as... Figure 3 As shown in the flowchart of the hierarchical lithium mine knowledge graph construction, the construction of the hierarchical lithium mine knowledge graph involves materializing the projections of these classes and relationships, transforming them into the form of triples, namely the "entity-relationship-entity" structure, and finally forming a network structure of entities and relationships.

[0070] Preferably, the construction process of the predefined ontology model in step S200 includes the following steps:

[0071] Key concept data were extracted from literature data on mining areas to construct a preliminary conceptual model;

[0072] Perform semantic adjustments and iterations on key concept data to establish a class hierarchy;

[0073] Define the attributes and relationships of each class, and map instance data to the ontology of the class hierarchy to form an ontology model.

[0074] For key concept data, classes in the ontology model are abstractions of real-world entities related to mineral deposits. For example, according to the ontology model, mineral deposit classes are subdivided into subclasses such as regional metallogenic background, mining area characteristics, mineral deposit genesis, and controlling factors. Each subclass is further subdivided; for example, mining area characteristics can be subdivided into geological features, tectonics, metamorphism, and metamorphic rocks. This hierarchical structure not only accurately describes the multidimensional information of mineral deposits but also facilitates efficient organization and management of information in the subsequent construction of a hierarchical lithium deposit knowledge graph. Relationships are the bridges connecting different mineral deposit ontology classes, describing various geological and geographical connections between these classes. Relationship types in the ontology model include attribute relationships, spatial relationships, temporal relationships, and causal relationships. For example, "located in" is a spatial relationship, indicating the specific location of one geological entity in another entity. "Formed at" is a temporal relationship, describing the time when a geological event occurred. These relationships help elucidate the interactions and historical development between different geological entities. To facilitate providing more detailed information in the ontology model, this information, which is key to understanding and classifying mineral deposit entities, is also defined as attributes. Attributes are specific descriptions of the mineral deposit ontology, including things like age, main igneous rock type, and occurrence. For example, the geological characteristics of a mining area can be described more specifically through attributes such as stratigraphic age and rock strata.

[0075] Preferably, in hierarchical lithium ore knowledge graphs, the "top-down" and "bottom-up" approaches are two classic knowledge organization and modeling methods. Combining these two approaches can achieve a unity of systematicity and practicality in domain knowledge. The "top-down" approach refers to pre-defining the ontology model of the hierarchical lithium ore knowledge graph (including core concepts, class hierarchies, relationship types, and attributes) based on existing mature theories, professional knowledge, or standard specifications within the domain, and then gradually refining the hierarchical structure. Its core is to first establish a "theoretical skeleton" to ensure the logic and domain conformity of the knowledge, such as... Figure 4 The diagram shows a predefined ontology model.

[0076] As one preferred embodiment, to ensure the systematic and rigorous nature of the construction process, it is typically combined with a seven-step ontology model construction method:

[0077] ① Define the domain and scope of the ontology model, that is, take the core objective of predicting pegmatite-type lithium deposits as the constraint condition;

[0078] ② Collect and organize relevant literature and data on mining areas, including dissertations and journal articles;

[0079] ③ Extract key concepts from it and construct a preliminary conceptual model:

[0080] ④ Based on the extracted concepts, a clear class hierarchy structure is established through subjective expert discussions and multiple rounds of iteration;

[0081] ⑤ Next, define the corresponding attributes and relationships for each type;

[0082] ⑥ Instantiate the model and map the actual data into the ontology model;

[0083] ⑦ Continuously optimize the ontology model through verification and evaluation to ensure its scientific validity and practicality.

[0084] As a preferred embodiment, a "bottom-up" approach is adopted, which involves extracting data and information from literature data of various mining areas and annotating the corpus to map the content of the text to a predefined ontology model. This achieves bottom-up supplementation, making the model more comprehensive and detailed. Figure 5 The diagram illustrates the bottom-up improvement of the ontology model. As the hierarchical lithium ore knowledge graph continues to grow, taking typical deposits within the region as the starting point, the ontology model is dynamically expanded alongside the gradual enrichment of the hierarchical lithium ore knowledge graph, such as... Figure 6 The expanded and improved ontology model is shown in the schematic diagram. Subsequently, by integrating the knowledge graphs of multiple typical deposits (such as the pegmatite-type lithium deposits in western Sichuan), a hierarchical lithium ore knowledge graph was constructed, such as... Figure 7 The schematic diagram of the intermediate-level lithium ore knowledge graph of this application embodiment is shown.

[0085] Preferably, the entity and relation-annotated text obtained through corpus annotation is transformed into triples, where each triple represents a node and an edge in the hierarchical lithium mining knowledge graph. Entities serve as nodes in the hierarchical lithium mining knowledge graph, while relations serve as edges, gradually forming a complete knowledge network. Ultimately, a complete hierarchical lithium mining knowledge graph is constructed, such as... Figure 8 The complete hierarchical lithium mine knowledge graph of this application embodiment is shown in the schematic diagram. It consists of 3345 nodes and 6141 edges. The entity types include 30 types such as rock strata, igneous rocks, and fractures; the 7 relationship types include has Attribute, is Made Of, is Caused By, etc.

[0086] Based on a complete hierarchical lithium ore knowledge graph, optional knowledge reasoning retrieval is performed to determine the key mineralization characteristics of pegmatite-type lithium deposits.

[0087] Preferably, such as Figure 2 As shown, step S100, which involves knowledge reasoning and retrieval based on a hierarchical lithium ore knowledge graph to determine the key features controlling mineralization, includes step S300:

[0088] S300, based on COL statements, retrieves the association patterns between specific types of nodes by setting semantic query paths and structured rules, identifies the recurring mineralization combination features and common patterns in multiple typical mineral deposits, and thus determines the key features of mineralization.

[0089] Specifically, CQL (Cypher Query Language) statements are used to reason and retrieve information from a hierarchical lithium ore knowledge graph. As a mainstream query language for graph databases, CQL's simple syntax allows for accurate retrieval of logical chains between entities. By setting semantic query paths and structured rules, the system systematically retrieves association patterns between specific types of nodes (such as deposit type, mineral assemblage, surrounding rock properties, and tectonic setting), identifies recurring mineralization assemblages and common patterns in multiple typical deposits, and thus reveals the key geological factors controlling mineralization as key features.

[0090] The following is a specific example of knowledge reasoning and retrieval:

[0091] Reasoning logic link 1: Songpan-Ganzi orogenic belt → (included) → Jiulong area → (developed) → Jiulong syncline → (produced) → Daqianggou lithium-beryllium deposit → (developed) → albite-spodumene pegmatite vein → (controls...formation) → Daqianggou lithium ore body.

[0092] Reasoning Logic Link 2: Songpan-Ganzi Orogenic Belt → (Developed) → Methylka Thermal Dome Structure → (Driven) → Magmatic Differentiation and Melt Replacement → (Occurrence) → Two-Mica Dimonite Granite → (Coexisting) → Contact Metamorphic Zone → (Occurrence) → Albite-Spodumene Pegmatite Dike → (Enriched) → Rare Elements such as Li, Be, Nb, and Ta → (Indication) → Methylka Lithium Ore Body.

[0093] Leveraging the visualization capabilities of a hierarchical lithium ore knowledge graph, the metallogenic logic from regional tectonics and ore-bearing intrusive bodies to mineralization anomalies is clearly displayed, providing logical support for subsequent mineral exploration prediction. This process not only enhances the systematic expression of metallogenic knowledge but also lays a solid foundation for knowledge-driven key feature identification, such as... Figure 9 The diagram illustrates the knowledge reasoning process.

[0094] Based on the aforementioned knowledge reasoning retrieval, this application preferably identifies the key mineralization characteristics of pegmatite-type lithium deposits as Li (lithium) and La (lanthanum) elements. Specifically, based on knowledge reasoning clues, the key characteristic is the ratio of Li to La elements.

[0095] Based on the identification of key features, mineralization prediction of pegmatite-type lithium deposits can be achieved by extracting geochemical data from the predicted mining area.

[0096] Preferably, such as Figure 2 As shown, step S101, which involves extracting key features based on the geochemical data of the predicted mining area and determining the ratios of these key features, includes step S400:

[0097] S400 uses the SA multifractal method to extract the Li / La elemental ratio anomaly in the predicted mining area and generates a Li / La geochemical anomaly map.

[0098] Li / La elemental ratio anomalies are identified from geochemical data as regions of Li / La ratios that deviate from the normal background value and are associated with pegmatite-type lithium mineralization. Specifically, before factor analysis, the KMO test statistic can be used to determine the suitability of factor analysis for the dataset, with a value range of [0,1]. A larger value indicates a stronger correlation between variables, and the corresponding data is more suitable for factor analysis. The Bartlett's test of sphericity tests whether the data is suitable for factor analysis by calculating the independence between variables. The rotated factor matrix in the factor analysis results can extract corresponding elemental combinations, including Li-La elemental combinations, where Li and La have a significant inverse relationship, thus identifying Li / La as an important mineral exploration indicator. The steps for accurately extracting Li / La elemental ratio anomalies using the SA multifractal method are as follows: Figure 10 The flowchart of the SA multifractal method is shown below:

[0099] ① Input geochemical data;

[0100] ② Obtain the energy spectral density by performing Fourier transform on the geochemical data;

[0101] ③ Numerical grouping of the energy spectral density and setting of distribution intervals provide a classification basis for subsequent analysis and distinguish the frequency domain characteristics of the data;

[0102] ④ Based on the energy spectral density, calculate the cumulative area A (>S) corresponding to a certain energy spectral density threshold S, and plot a double logarithmic graph (horizontal axis: lgS, vertical axis: lgA).

[0103] ⑤ Perform multi-segment fitting on the double logarithmic graph to identify the inflection points of outlier segments;

[0104] ⑥ Set up a fractal filter to extract anomalous signals related to mineralization;

[0105] ⑦ Using a fractal filter, the energy spectral density is filtered to obtain the Li / La elemental ratio anomaly map.

[0106] Preferably, such as Figure 2 As shown, step S102 involves determining anomalies in the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits, including step S500:

[0107] S500 uses remote sensing data interpretation to extract important ore-controlling factors and ore occurrences, and displays them in combination with Li / La geochemical anomaly maps for comparison of the overlay effect.

[0108] In one embodiment, to fully leverage the advantages of multi-source data, the Li / La elemental ratio data is used as one of the important indicators in mineralization prediction, as shown in Table 1:

[0109] Table 1 Other Predictive Indicators of Embodiments of the Invention

[0110]

[0111] When constructing the prediction model, 25 common machine learning algorithms were initially screened, and the four best-performing models were used as the foundation to build a Stacking ensemble learning intelligent mineral exploration prediction model, such as... Figure 11 The flowchart of a specific application example of a mineralization prediction method is shown. Based on the prediction results, target areas were delineated, including 8 Class I high-potential target areas and 13 Class II relatively high-potential target areas. One high-potential target area was discovered northeast of the Jiada mining area, where no known mineral deposits exist. Currently, field sampling verification work has been completed in the study area, with 3 samples taken in the Jiada area, 4 in the Guanyinqiao area, and 2 in the Gaorang area. The test results show that the average Li2O content in the Jiada samples is 4.37%, the average Li2O content in the Guanyinqiao samples is 2.37%, and the average Li2O content in the Gaorang samples is 2.56%, all exceeding industrial grades. Simultaneously, the associated minerals beryllium and rubidium in these areas also reach industrial grades, indicating high comprehensive utilization value. Furthermore, the three areas show significant enrichment in cesium, tungsten, tin, and cobalt, reflecting good potential for deep and peripheral mineral exploration. Figure 12 The specific application example is shown in the target area delineation and field verification map of the Keerin mining area.

[0112] In one embodiment, to verify the feasibility of the embodiments of this application, verification work was carried out in six metallogenic belts based on Li / La elemental ratio anomalies and remote sensing image interpretation. The ore-controlling factors extracted from remote sensing, mineral occurrences, and Li / La elemental ratio anomaly maps were combined and displayed, fully leveraging the advantages of multi-source data, comprehensively revealing ore-controlling factors and metallogenic regularities, and greatly improving the efficiency and accuracy of lithium exploration. Specifically, for the geochemical data of the Altai region, the cumulative frequency method and SA multifractal method were used to statistically analyze the geochemical single-element data and elemental combination data, combined with factor analysis. It was found that Li and La have a significant inverse relationship, and the known mineral occurrences in the high-value anomaly areas on the Li / La geochemical map overlap well. The Beyesamas spodumene deposit and the Koktokay beryllium-lithium deposit show a very close relationship in the distribution of geochemical anomalies, such as... Figure 13 As shown in the Li / La geochemical anomaly map of western Sichuan and the Altai region, the high Li / La anomaly area west of Gangdise corresponds well with known lithium deposits, such as... Figure 14The Li / La geochemical anomaly map of the Gangdise region is shown. The Himalayan metallogenic belt is an important rare metal mineralization area, with widespread distribution of Cenozoic leucogranite, considered a key ore-controlling factor in the region's rare metal mineralization. Geochemical data within the region show a good superposition effect between the Li / La ratio and the distribution of lithium and beryllium rare metal deposits in the southern part of the eastern segment of the Himalayan metallogenic belt, such as... Figure 15 The Li / La geochemical anomaly map of the Himalayas is shown. The West Kunlun lithium-bearing belt and the West Tianshan lithium-bearing belt are important lithium resource-rich areas. According to... Figure 16 The geochemical maps of Li / La ratios in the West Kunlun and West Tianshan lithium-bearing belts show significant high-value anomalies (Li / La ratios ranging from 1.667 to 138.134) in the central West Kunlun and western West Tianshan regions. These anomalies closely correspond to the spatial distribution of known lithium deposits (points) such as Bailongshan, Xuefengling, and Fulugou. The granitic geological bodies in the maps closely overlap with the high Li / La ratio areas, suggesting they may serve as key geological carriers for lithium enrichment and mineralization. In particular, the Li / La anomaly gradient zones in areas such as 509 Road Maintenance Station West and Kangxishan within the West Kunlun lithium-bearing belt, and the wide range of ratio variations (0.905-5.231) in areas such as Shayintubai in the West Tianshan lithium-bearing belt, vividly reveal the differentiation and enrichment patterns of lithium during magmatic-hydrothermal processes, exhibiting a good spatial correspondence and providing important geochemical indicators for regional lithium exploration.

[0113] This application's embodiment of the method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies involves acquiring and constructing a hierarchical lithium deposit knowledge graph based on mining area literature data, and performing knowledge reasoning retrieval based on the hierarchical lithium deposit knowledge graph to determine key features controlling mineralization. Key features are extracted from the geochemical data of the predicted mining area, and the ratios of these key features are determined. Anomaly detection is performed on the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits. By using knowledge reasoning retrieval from the hierarchical lithium deposit knowledge graph, key mineralization indicators are found to predict the mineralization of pegmatite-type lithium deposits, facilitating more accurate extraction of mineralization information and prediction of mining areas.

[0114] This application also provides a pegmatite-type lithium ore prediction system based on Li / La elemental ratio anomalies.

[0115] Figure 17 This is a structural diagram of a pegmatite-type lithium ore detection device module based on elemental ratio anomalies, according to one embodiment of the application. Figure 17 As shown, one embodiment of the pegmatite-type lithium ore prediction system based on Li / La elemental ratio anomalies includes:

[0116] The feature filtering module 100 is used to acquire and construct a hierarchical lithium ore knowledge graph based on mining area literature data, and to perform knowledge reasoning retrieval based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization.

[0117] The feature capture module 101 is used to extract key features based on the geochemical data of the predicted mining area and determine the ratio of the key features.

[0118] The mineralization prediction module 102 is used to determine anomalies in the comparison values ​​in order to predict the mineralization of pegmatite-type lithium deposits.

[0119] This application's embodiment of the pegmatite-type lithium deposit prediction system based on Li / La elemental ratio anomalies acquires and constructs a hierarchical lithium deposit knowledge graph based on mining area literature data, and performs knowledge reasoning retrieval based on the hierarchical lithium deposit knowledge graph to determine key features controlling mineralization; extracts the key features based on geochemical data of the predicted mining area, and determines the ratio of the key features; and performs anomaly judgment on the comparison values ​​to predict the mineralization of pegmatite-type lithium deposits. By using knowledge reasoning retrieval of the hierarchical lithium deposit knowledge graph, key mineralization indicators are found to predict the mineralization of pegmatite-type lithium deposits, thereby facilitating more accurate extraction of mineralization information and prediction of mining areas.

[0120] At least one embodiment of this application also provides a data control device. Figure 18 This is a schematic block diagram of a data control device provided for at least one embodiment of this application. For example, such as... Figure 18 As shown, the data control device 20 may include one or more memories 200 and one or more processors 201. The memories 200 are used to store computer-executable instructions non-transitory; the processors 201 are used to run the computer-executable instructions, which, when run by the processors 201, can cause the processors 201 to perform one or more steps in the pegmatite-type lithium ore prediction method based on Li / La elemental ratio anomalies according to any embodiment of this application.

[0121] The specific implementation and explanation of each step in the method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies can be found in the relevant content of the embodiments of the method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies mentioned above, and will not be repeated here. It should be noted that... Figure 18 The components of the data control device 20 shown are merely exemplary and not limiting. The data control device 20 may have other components depending on the actual application requirements.

[0122] In one embodiment, the processor 201 and the memory 200 can communicate directly or indirectly with each other. For example, the processor 201 and the memory 200 can communicate via a network connection. The network can include a wireless network, a wired network, and / or any combination of wireless and wired networks; this application does not limit the type and function of the network. Alternatively, the processor 201 and the memory 200 can also communicate via a bus connection. The bus can be a Peripheral Component Interconnect Standard (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. For example, the processor 201 and the memory 200 can be located at a remote data server (cloud) or a distributed energy system (local), or at a client (e.g., a mobile device such as a mobile phone). For example, the processor 201 can be a central processing unit (CPU), a tensor processor (TPU), or a graphics processing unit (GPU), etc., with data processing and / or instruction execution capabilities, and can control other components in the data control device 20 to perform desired functions. The central processing unit (CPU) can be an x86 or ARM architecture, etc.

[0123] In one embodiment, memory 200 may include any combination of one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, erasable programmable read-only memory (EPROM), portable compact disc read-only memory (CD-ROM), USB memory, flash memory, etc. One or more computer-executable instructions may be stored on the computer-readable storage medium, and processor 201 may execute these computer-executable instructions to implement various functions of data control device 20. Various application programs and various data, as well as various data used and / or generated by the application programs, may also be stored in memory 200.

[0124] It should be noted that the data control device 20 can achieve similar technical effects to the aforementioned pegmatite-type lithium ore prediction method based on the anomaly of the Li / La element ratio, and the repetition will not be repeated.

[0125] At least one embodiment of this application also provides a non-transitory computer-readable storage medium. Figure 19 This is a schematic diagram of a non-transitory computer-readable storage medium provided for at least one embodiment of this application. For example, such as... Figure 19As shown, one or more computer-executable instructions 301 may be stored non-transitory on the non-transitory computer-readable storage medium 30. For example, when the computer-executable instructions 301 are executed by a computer, the computer may perform one or more steps in a method for predicting pegmatite-type lithium deposits based on anomalies in the Li / La elemental ratio according to any embodiment of this application.

[0126] In one embodiment, the non-transitory computer-readable storage medium 30 can be applied to the data control device 20 described above, for example, it can be the memory 200 in the data control device 20.

[0127] In one embodiment, the description of the non-transitory computer-readable storage medium 30 can be found in the description of the memory 200 in the embodiment of the data control device 20, and will not be repeated hereafter.

[0128] It should be noted that the memory 200 stores different non-transient computer-executable instructions, and the data control device 20 corresponds to the firmware upgrade device. When the computer-executable instructions are run by the processor 201, the processor 201 can perform one or more steps in the pegmatite-type lithium ore prediction method based on the Li / La element ratio anomaly according to any embodiment of this application.

[0129] The following points should be noted regarding this application:

[0130] (1) The accompanying drawings of the embodiments of this application only involve the structures involved in the embodiments of this application. Other structures can be referred to the general design.

[0131] (2) For clarity, the thickness and dimensions of layers or structures are enlarged in the accompanying drawings used to describe embodiments of the invention. It will be understood that when an element such as a layer, film, region, or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element, or there may be intermediate elements present.

[0132] (3) Where there is no conflict, the embodiments and features in the embodiments of this application can be combined with each other to obtain new embodiments. The above are only specific implementations of this application, but the protection scope of this application is not limited thereto, and the protection scope of this application shall be determined by the protection scope of the claims.

[0133] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0134] The above embodiments merely illustrate several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies, characterized in that, Including the following steps: A hierarchical lithium ore knowledge graph was acquired and constructed based on literature data from the mining area. Knowledge reasoning and retrieval were then performed based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization. The key features are Li and La elements. The process of constructing a hierarchical lithium ore knowledge graph based on mining area literature data includes the following steps: Extract the information text from the literature data of the mining area and map the information text to a predefined ontology model; By integrating the knowledge graphs of typical mineral deposits with the ontology model, a hierarchical lithium mineral knowledge graph is obtained; wherein, the construction of the hierarchical lithium mineral knowledge graph is a model system based on the ontology model and the corpus annotation results; The process of determining the key features controlling mineralization by performing knowledge reasoning and retrieval based on a hierarchical lithium ore knowledge graph includes the following steps: Based on COL statements, by setting semantic query paths and structured rules, the association patterns between specific types of nodes are retrieved, and the metallogenic assemblages and common patterns that repeatedly appear in multiple typical mineral deposits are identified to determine the key features of mineralization; wherein, the specific types of nodes include mineral deposit type, mineral assemblage, surrounding rock properties, and tectonic setting; The key features are extracted based on the geochemical data of the predicted mining area, and the ratios of the key features are determined. The process of extracting the key features based on the geochemical data of the predicted mining area and determining the ratio of the key features includes the following steps: Based on the SA multifractal method, the Li / La element ratio anomaly of the predicted mining area was extracted, and a Li / La geochemical anomaly map was generated. The ratio is used to determine anomalies in order to predict the mineralization of pegmatite-type lithium deposits.

2. The method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies according to claim 1, characterized in that, The construction process of the predefined ontology model includes the following steps: Key concept data were extracted from the literature data of the mining area to construct a preliminary conceptual model; The key concept data is semantically adjusted and iterated to establish a class hierarchy structure; Define the attributes and relationships of each class, and map instance data to the ontology of the class hierarchy to form the ontology model.

3. The method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies according to claim 1 or 2, characterized in that, The mineralization prediction is based on a machine learning model.

4. The method for predicting pegmatite-type lithium deposits based on Li / La elemental ratio anomalies according to claim 1, characterized in that, The process of determining anomalies in the ratio for predicting the mineralization of pegmatite-type lithium deposits includes the following steps: Key mineralization-controlling factors and mineral occurrences were extracted by interpreting remote sensing data and then displayed in combination with Li / La geochemical anomaly maps for comparison.

5. A prediction system for pegmatite-type lithium deposits based on Li / La elemental ratio anomalies, characterized in that, include: The feature filtering module is used to acquire and construct a hierarchical lithium ore knowledge graph based on mining area literature data, and to perform knowledge reasoning retrieval based on the hierarchical lithium ore knowledge graph to determine the key features controlling mineralization; wherein, the key features are Li and La elements; The process of constructing a hierarchical lithium ore knowledge graph based on mining area literature data includes the following steps: Extract the information text from the literature data of the mining area and map the information text to a predefined ontology model; By integrating the knowledge graphs of typical mineral deposits with the ontology model, a hierarchical lithium mineral knowledge graph is obtained; wherein, the construction of the hierarchical lithium mineral knowledge graph is a model system based on the ontology model and the corpus annotation results; The process of determining the key features controlling mineralization by performing knowledge reasoning and retrieval based on a hierarchical lithium ore knowledge graph includes the following steps: Based on COL statements, by setting semantic query paths and structured rules, the association patterns between specific types of nodes are retrieved, and the metallogenic assemblages and common patterns that repeatedly appear in multiple typical mineral deposits are identified to determine the key features of mineralization; wherein, the specific types of nodes include mineral deposit type, mineral assemblage, surrounding rock properties, and tectonic setting; The feature capture module is used to extract the key features based on the geochemical data of the predicted mining area and determine the ratio of the key features; The process of extracting the key features based on the geochemical data of the predicted mining area and determining the ratio of the key features includes the following steps: Based on the SA multifractal method, the Li / La element ratio anomaly of the predicted mining area was extracted, and a Li / La geochemical anomaly map was generated. The mineralization prediction module is used to determine anomalies in the ratio in order to predict the mineralization of pegmatite-type lithium deposits.

6. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium stores computer-executable instructions that, when executed by a processor, implement the pegmatite-type lithium ore prediction method based on Li / La elemental ratio anomalies as described in any one of claims 1 to 4.

7. A data control device, characterized in that, include: One or more memories that store computer-executable instructions non-transitory; One or more processors configured to run computer-executable instructions, wherein the computer-executable instructions are executed by the one or more processors to implement the pegmatite-type lithium deposit prediction method based on Li / La element ratio anomalies as described in any one of claims 1 to 4.

Images