Lithium ore prospecting prediction method and system

Abstract

The invention relates to a lithium ore prospecting prediction method and system, and relates to the technical field of mineral exploration. According to the lithium ore prospecting prediction method and system, the prediction accuracy of ore body identification is high, the exploration period can be shortened, and the exploration cost can be reduced. According to the method, the multi-dimensional data fusion is combined with the intelligent algorithm, the three-dimensional mineralization probability is output through the prediction neural network model, the prediction accuracy of ore body recognition is high, the exploration period can be shortened, and the exploration cost can be reduced.

Description

Technical Field

[0001] This application relates to the field of mineral exploration technology, and in particular to a method and system for predicting lithium ore prospecting. Background Technology

[0002] Clay-type lithium deposits, also known as sedimentary lithium deposits or unconventional lithium deposits, originally referred to lithium deposits found in sedimentary rocks that did not yet have independent industrial mining value but had market potential.

[0003] Current prediction of clay-type lithium deposits faces certain technical bottlenecks: traditional geological models rely on expert experience and are difficult to quantify mineralization patterns; two-dimensional prediction methods cannot effectively express the three-dimensional distribution characteristics of ore bodies; and the integration of multi-source exploration data is low, resulting in insufficient information utilization.

[0004] These problems make it difficult to fully reflect the distribution of lithium deposits, resulting in inaccurate predictions. In addition, existing prediction methods often lack the ability to integrate multi-dimensional data, leading to low efficiency in mineral exploration prediction. Summary of the Invention

[0005] To address or partially address the problems existing in related technologies, this application provides a method and system for predicting lithium ore exploration. The prediction accuracy of ore body identification is high, which can shorten the exploration cycle and reduce exploration costs.

[0006] The first aspect of this application provides a method for predicting lithium mineral exploration, the method comprising the following steps: S1: Construct a multi-dimensional database that integrates four-dimensional geological conditions, geophysical data, and geochemical data features; S2: Construct a predictive neural network model based on multi-dimensional database and multi-scale feature fusion; S3: Output the three-dimensional mineralization probability based on the predictive neural network model.

[0007] Optionally, in some embodiments of the first aspect, the four-dimensional geological conditions include: Three-dimensional geological model of a Cenozoic sedimentary basin; Hyperspectral remote sensing data; Data on the content of adsorbed lithium phases in soil; Data on the detection of lithium-enriched bacterial biomarkers.

[0008] Optionally, in some embodiments of the first aspect, the data feature fusion specifically includes: Data cleaning is performed by filtering out abnormal data using the 3σ criterion. The seven-parameter Bursa model was used to unify the coordinate system of the three-dimensional geological model of the Cenozoic sedimentary basin, hyperspectral remote sensing data, soil adsorbed phase lithium content data, and lithium-enriched microbial community detection data, and to standardize them. Spatial registration and alignment were performed using a 50m³ base grid.

[0009] Optionally, in some embodiments of the first aspect, the data feature fusion further includes: Extracting sedimentary facies codes and tectonic parameters from geological models of sedimentary basins; Mineral identification features and texture indices are extracted from hyperspectral data; Calculate elemental phase ratios from soil data; Community structure and lithium tolerance indices were obtained from microbial data.

[0010] Optionally, in some embodiments of the first aspect, the data feature fusion further includes: Deep integration is achieved through three levels of fusion: multi-scale features are processed through spatial pyramid pooling, cross-modal associations are established through graph attention networks, and dynamic weight fusion guided by knowledge graphs is used to finally output a multi-dimensional database with uncertainty assessment.

[0011] Optionally, in some embodiments of the first aspect, the construction of the multi-scale feature fusion predictive neural network model specifically includes: Input layer: A multi-branch convolutional neural network (CNN) is used to process geological data from different sources. Feature fusion layer: Multi-scale feature fusion is achieved through a Feature Pyramid Network (FPN); Temporal processing layer: Uses a Long Short-Term Memory (LSTM) network to capture temporal geological evolution features; Attention layer: Employs a self-attention mechanism to enhance the weight allocation of key features; Output layer: Outputs the three-dimensional mineralization probability and its uncertainty assessment through a Bayesian neural network.

[0012] Optionally, in some embodiments of the first aspect, the training method for the predictive neural network model includes: The cross-entropy loss function is used to optimize the classification task; Use the Adam optimizer to update parameters; Overfitting is prevented using Dropout and Batch Normalization; A transfer learning strategy is adopted, which uses existing mining area data to pre-train the model and then fine-tunes it for the target mining area.

[0013] Optionally, in some embodiments of the first aspect, the output method of the three-dimensional mineralization probability includes: The probability of lithium ore distribution at different depths is displayed in the form of a voxel grid. Based on geological constraints, suggestions for exploitable areas are generated. Provides uncertainty heatmaps to aid decision analysis.

[0014] A second aspect of this application provides a lithium ore prospecting prediction system, comprising: The data acquisition module is used to acquire four-dimensional geological conditions, geophysical and geochemical data; The data processing module is used to execute the data feature fusion method; The neural network prediction module is used to build and run prediction models that fuse multiple scale features. The 3D visualization module is used to display the results of mineralization probability and uncertainty assessment.

[0015] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the lithium ore prospecting prediction method described in the first aspect above. The technical solution provided in this application may include the following beneficial effects: This application achieves high prediction accuracy for ore body identification by combining multi-dimensional data fusion with intelligent algorithms and outputting three-dimensional mineralization probability through a predictive neural network model. This can shorten the exploration cycle and reduce exploration costs.

[0016] This application integrates Cenozoic sedimentary basin model data to reflect the dynamic process of lithium enrichment, provides spectral characteristics through hyperspectral data to identify surface lithium mineral anomalies, reflects the degree of secondary lithium enrichment through soil adsorption state data, and provides indirect evidence from a biological perspective through the detection of lithium-enriched microbial biomarkers. This multi-dimensional data integration can improve the completeness of mineral exploration marker identification and enhance the accuracy of mineral exploration.

[0017] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0018] The above and other objects, features and advantages of this application will become more apparent from the more detailed description of exemplary embodiments thereof in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments thereof.

[0019] Figure 1 This is a schematic diagram of a lithium ore prospecting prediction method in an embodiment of this application. Detailed Implementation

[0020] Embodiments of this application will now be described in more detail with reference to the accompanying drawings. While embodiments of this application are shown in the drawings, it should be understood that this application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to make this application more thorough and complete, and to fully convey the scope of this application to those skilled in the art.

[0021] It should be understood that although the terms "first," "second," "third," etc., may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0022] In the description of this application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0023] Unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0024] Current prediction of clay-type lithium deposits faces certain technical bottlenecks: traditional geological models rely on expert experience and are difficult to quantify mineralization patterns; two-dimensional prediction methods cannot effectively express the three-dimensional distribution characteristics of ore bodies; and the integration of multi-source exploration data is low, resulting in insufficient information utilization.

[0025] These problems make it difficult to fully reflect the distribution of lithium deposits, resulting in inaccurate predictions. In addition, existing prediction methods often lack the ability to integrate multi-dimensional data, leading to low efficiency in mineral exploration prediction.

[0026] To address the aforementioned issues, this application provides a method for predicting lithium ore deposits, which offers high accuracy in identifying ore bodies, shortens the exploration cycle, and reduces exploration costs.

[0027] The technical solutions of the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0028] Figure 1 This is a schematic diagram of a lithium ore prospecting prediction method in an embodiment of this application.

[0029] See Figure 1 A method for predicting lithium mineral exploration, comprising the following steps: S1: Construct a multi-dimensional database that integrates four-dimensional geological conditions, geophysical data, and geochemical data features.

[0030] Specifically, the four-dimensional geological conditions include: a three-dimensional geological model of the Cenozoic sedimentary basin; hyperspectral remote sensing data; soil adsorbed phase lithium content data; and detection data of lithium-enriched bacterial community biomarkers.

[0031] In this embodiment, by integrating Cenozoic sedimentary basin model data, the dynamic process of lithium enrichment can be reflected. Hyperspectral data provides spectral characteristics that can identify surface lithium mineral anomalies. Soil adsorption state data reflects the degree of secondary lithium enrichment. Detection of lithium-enriched microbial biomarkers can provide indirect evidence from a biological perspective. This multi-dimensional data integration can improve the completeness of mineral exploration marker identification and enhance the accuracy of mineral exploration.

[0032] Specifically, this data feature fusion includes: Data cleaning is performed by filtering outlier data using the 3σ criterion; the 3σ criterion can automatically adjust the filtering threshold according to different data types.

[0033] The seven-parameter Bursa model unifies the coordinate system of the three-dimensional geological model of the Cenozoic sedimentary basin, hyperspectral remote sensing data, soil adsorbed phase lithium content data, and lithium-enriched microbial community detection data, and standardizes them. In particular, by transforming parameters such as translation, rotation, and scaling, the planar positioning error of data from different sources is reduced, thereby improving the accuracy of subsequent feature extraction.

[0034] A 50m³ base grid was used for spatial registration and alignment. Spatial alignment further improved the accuracy of feature extraction.

[0035] In this embodiment, by performing data cleaning, standardization, and spatial alignment, abnormal interference data can be effectively eliminated, the accuracy of subsequent feature extraction can be improved, and the stability of subsequent modeling can be ensured.

[0036] Specifically, this data feature fusion also includes: Sedimentary facies codes and tectonic parameters were extracted from geological models of sedimentary basins; mineral identification features and texture indicators were extracted from hyperspectral data; elemental phase ratios were calculated from soil data; and community structure and lithium tolerance indicators were obtained from microbial data.

[0037] In this embodiment, by extracting features from different types of data, it is possible to achieve in-depth mining and collaborative analysis of lithium mineralization information in multiple dimensions and at multiple scales, thereby improving the prediction accuracy of subsequent modeling.

[0038] Specifically, this data feature fusion also includes: Deep integration is achieved through three levels of fusion: multi-scale features are processed through spatial pyramid pooling, cross-modal associations are established through graph attention networks, and dynamic weight fusion guided by knowledge graphs is used to finally output a multi-dimensional database with uncertainty assessment.

[0039] In this embodiment, a three-level fusion technique is used to achieve deep intelligent integration and uncertainty quantification assessment of multi-source geological data: First, spatial pyramid pooling is used to process multi-scale features to improve the ore body identification rate; second, cross-modal association is established through graph attention network to improve feature extraction efficiency; finally, dynamic weight fusion is achieved based on lithium ore knowledge graph to improve the spatial resolution and accuracy of prediction model.

[0040] S2: Based on a multi-dimensional database, construct a multi-scale feature fusion predictive neural network model. Specifically, constructing this multi-scale feature fusion predictive neural network model includes: Input layer: A multi-branch convolutional neural network (CNN) is used to process geological data from different sources. Feature fusion layer: Multi-scale feature fusion is achieved through a Feature Pyramid Network (FPN); Temporal processing layer: Uses a Long Short-Term Memory (LSTM) network to capture temporal geological evolution features; Attention layer: Employs a self-attention mechanism to enhance the weight allocation of key features; Output layer: Outputs the three-dimensional mineralization probability and its uncertainty assessment through a Bayesian neural network.

[0041] In this embodiment, by constructing a predictive neural network model that integrates multi-scale features, it is possible to achieve multi-dimensional intelligent prediction of lithium mineralization patterns, shorten the exploration cycle, reduce exploration costs, and effectively form intelligent prediction technical specifications.

[0042] Specifically, the training method for this predictive neural network model includes: The cross-entropy loss function is used to optimize the classification task; Use the Adam optimizer to update parameters; Overfitting is prevented using Dropout and Batch Normalization; A transfer learning strategy is adopted, which uses existing mining area data to pre-train the model and then fine-tunes it for the target mining area.

[0043] In this embodiment, by employing a weighted cross-entropy loss function, the data imbalance problem is effectively solved, the hyperparameter configuration of the Adam optimizer is optimized, the convergence speed of the loss function is accelerated during training, and overfitting is prevented through Dropout and Batch Normalization, which can effectively improve the prediction accuracy of ore body identification.

[0044] S3: Output the three-dimensional mineralization probability based on the predictive neural network model.

[0045] Specifically, the output methods for this three-dimensional mineralization probability include: It displays the probability of lithium ore distribution at different depths in the form of voxel grids; it can intuitively show the spatial distribution pattern of lithium ore, support three-dimensional exploration planning, and avoid the limitations of traditional two-dimensional maps.

[0046] Based on geological constraints, suggestions for exploitable areas are generated. Provides uncertainty heatmaps to aid decision analysis.

[0047] In this embodiment, by combining multi-dimensional data fusion with intelligent algorithms, a three-dimensional mineralization probability is output through a predictive neural network model. The prediction accuracy of ore body identification is high, which can shorten the exploration cycle and reduce exploration costs.

[0048] Corresponding to the aforementioned application function implementation device embodiments, this application also provides a lithium ore prospecting prediction system and corresponding embodiments.

[0049] A lithium ore prospecting prediction system, comprising: The data acquisition module is used to acquire four-dimensional geological conditions, geophysical and geochemical data; The data processing module is used to execute the data feature fusion method; The neural network prediction module is used to build and run prediction models that fuse multiple scale features. The 3D visualization module is used to display the results of mineralization probability and uncertainty assessment.

[0050] Corresponding to the aforementioned application function implementation device embodiments, this application also provides a computer-readable storage medium and corresponding embodiments.

[0051] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the lithium ore prospecting prediction method described in the above embodiments.

[0052] The technical solution provided in this application may include the following beneficial effects: This application achieves high prediction accuracy for ore body identification by combining multi-dimensional data fusion with intelligent algorithms and outputting three-dimensional mineralization probability through a predictive neural network model. This can shorten the exploration cycle and reduce exploration costs.

[0053] This application integrates Cenozoic sedimentary basin model data to reflect the dynamic process of lithium enrichment, provides spectral characteristics through hyperspectral data to identify surface lithium mineral anomalies, reflects the degree of secondary lithium enrichment through soil adsorption state data, and provides indirect evidence from a biological perspective through the detection of lithium-enriched microbial biomarkers. This multi-dimensional data integration can improve the completeness of mineral exploration marker identification and enhance the accuracy of mineral exploration.

[0054] Finally, it should be noted that in this document, relationships such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "include," "contain," or any other variations are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus.

[0055] The unit described as a separate component may or may not be physically separate. The component shown as a unit may or may not be a physical unit; that is, it may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0056] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0057] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0058] The various embodiments of this application have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or improvement of the technology in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A method for predicting lithium deposits, characterized in that, The method includes the following steps: S1: Construct a multi-dimensional database that integrates four-dimensional geological conditions, geophysical data, and geochemical data features; S2: Construct a predictive neural network model based on multi-dimensional database and multi-scale feature fusion; S3: Output the three-dimensional mineralization probability based on the predictive neural network model.

2. The method for lithium ore prospecting and prediction according to claim 1, characterized in that: The four-dimensional geological conditions include: Three-dimensional geological model of a Cenozoic sedimentary basin; Hyperspectral remote sensing data; Data on the content of adsorbed lithium phases in soil; Data on the detection of lithium-enriched bacterial biomarkers.

3. The method for lithium ore prospecting and prediction according to claim 2, characterized in that: The data feature fusion specifically includes: Data cleaning is performed by filtering out abnormal data using the 3σ criterion. The seven-parameter Bursa model was used to unify the coordinate system of the three-dimensional geological model of the Cenozoic sedimentary basin, hyperspectral remote sensing data, soil adsorbed phase lithium content data, and lithium-enriched microbial community detection data, and to standardize them. Spatial registration and alignment were performed using a 50m³ base grid.

4. The method for lithium ore prospecting and prediction according to claim 3, characterized in that: The data feature fusion specifically also includes: Extracting sedimentary facies codes and tectonic parameters from geological models of sedimentary basins; Mineral identification features and texture indices are extracted from hyperspectral data; Calculate elemental phase ratios from soil data; Community structure and lithium tolerance indices were obtained from microbial data.

5. The method for lithium ore prospecting and prediction according to claim 4, characterized in that: The data feature fusion specifically also includes: Deep integration is achieved through three levels of fusion: multi-scale features are processed through spatial pyramid pooling, cross-modal associations are established through graph attention networks, and dynamic weight fusion guided by knowledge graphs is used to finally output a multi-dimensional database with uncertainty assessment.

6. The method for lithium ore prospecting prediction according to claim 1 or 5, characterized in that: The construction of the multi-scale feature fusion predictive neural network model specifically includes: Input layer: A multi-branch convolutional neural network (CNN) is used to process geological data from different sources. Feature fusion layer: Multi-scale feature fusion is achieved through a Feature Pyramid Network (FPN); Temporal processing layer: Uses a Long Short-Term Memory (LSTM) network to capture temporal geological evolution features; Attention layer: Employs a self-attention mechanism to enhance the weight allocation of key features; Output layer: Outputs the three-dimensional mineralization probability and its uncertainty assessment through a Bayesian neural network.

7. The method for lithium ore prospecting and prediction according to claim 6, characterized in that: The training method for the predictive neural network model includes: The cross-entropy loss function is used to optimize the classification task; Use the Adam optimizer to update parameters; Overfitting is prevented using Dropout and Batch Normalization; A transfer learning strategy is adopted, which uses existing mining area data to pre-train the model and then fine-tunes it for the target mining area.

8. The method for lithium ore prospecting and prediction according to claim 7, characterized in that: The output methods for the three-dimensional mineralization probability include: The probability of lithium ore distribution at different depths is displayed in the form of a voxel grid. Based on geological constraints, suggestions for exploitable areas are generated. Provides uncertainty heatmaps to aid decision analysis.

9. A lithium ore prospecting prediction system, characterized in that, include: The data acquisition module is used to acquire four-dimensional geological conditions, geophysical and geochemical data; The data processing module is used to execute the data feature fusion method according to any one of claims 1-8; The neural network prediction module is used to build and run prediction models that fuse multiple scale features. The 3D visualization module is used to display the results of mineralization probability and uncertainty assessment.

10. A computer-readable storage medium, characterized in that, The device contains a computer program that, when executed by a processor, implements the lithium ore prospecting prediction method according to any one of claims 1-8.

Images