Quantitative Prediction Method for Metallogenic Prospects Based on Multi-Source Geological Data Fusion

By integrating multi-source geological data to establish cross-source geological evidence relationships and consistency judgments, and by combining historical borehole data to verify the prediction results of favorable mineralization areas, the source of abnormal data is traced in reverse and supplemented by collection. This solves the problem of tracing abnormal data in the prediction of prospective mineralization areas, and improves the reliability of prediction and exploration efficiency.

CN123113031APending Publication Date: 2026-11-17四川省第二地质大队
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611364516.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-04
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing methods for predicting prospective mineral deposits have failed to effectively verify the spatial correspondence and mutual support of geological data from different sources. This makes it difficult to trace the source of anomalous data, and the supplementary data collection is not targeted enough, which affects the reliability of quantitative prediction of prospective mineral deposits and the efficiency of exploration.

Method used

By acquiring gravity exploration data, geochemical sampling data, and seismic exploration data, multi-source geological features are formed, cross-source geological evidence relationships are established, consistency is determined, a mineralization prediction model is constructed, the prediction results are reviewed in conjunction with historical borehole data, the source of abnormal data is traced in reverse, and targeted supplementary collection is performed.

Benefits of technology

It improves the reliability of quantitative prediction of mineralized prospective areas, reduces the impact of local anomalies on prediction results, reduces redundant exploration work, and improves exploration efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113031A_ABST
    Figure CN123113031A_ABST
Patent Text Reader

Abstract

This invention relates to the field of mineralization prediction technology, and particularly to a quantitative prediction method for prospective mineralization areas based on multi-source geological data fusion. The method includes: acquiring gravity exploration data, geochemical sampling data, seismic exploration data, and historical borehole data within the exploration area to form corresponding multi-source geological features; establishing cross-source geological evidence relationships and determining the consistency of the multi-source geological features to identify the source of initial anomaly data when consistency requirements are not met; constructing a mineralization prediction model to predict the mineralization potential of the exploration area, obtaining prediction results for mineralization-favorable zones; verifying the prediction results, tracing back the multi-source geological features involved in forming the corresponding prediction results to determine the source of target anomaly data and target supplementary collection locations, and performing corresponding supplementary data collection; and generating a quantitative prediction map of prospective mineralization areas. This invention improves the reliability of quantitative prediction of prospective mineralization areas.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of mineralization prediction technology, and in particular to a quantitative prediction method for mineralization prospective areas based on the fusion of multi-source geological data. Background Technology

[0002] In existing technologies, the prediction of potential mineralization areas typically involves comprehensive analysis of multi-source geological data, such as gravity, geochemical, seismic, and borehole data, combined with statistical or machine learning models to delineate favorable mineralization areas. However, geological data from different sources vary in spatial scale, anomaly patterns, and acquisition quality. Existing methods often focus on directly overlaying, feature fusion, or unified modeling of multi-source data, lacking further verification of the spatial correspondence and mutual support between geological anomalies from different sources. When local gravity, geochemical, or seismic data exhibits anomalies, gaps, or does not correspond to actual geological conditions, these anomalous data can be directly transmitted to the prediction model, leading to distortion of local prediction results. Meanwhile, existing methods, after obtaining mineralization prediction results, mostly rely on model output or existing borehole data for result verification, making it difficult to further determine the source of abnormal prediction results by combining the continuous state of geological characteristics in adjacent areas. When abnormal prediction results are found, data are often obtained again by uniform supplementary mining or expanding the exploration scope, making it difficult to accurately locate the specific data source and corresponding collection location that caused the prediction anomaly. This results in insufficient targeted supplementary collection and a large amount of repetitive exploration work, which in turn affects the reliability and exploration efficiency of quantitative prediction of mineralization prospective areas.

[0003] Chinese Patent Publication No. CN116611697A discloses a method for delineating mineral exploration target areas based on multi-source heterogeneous information using machine learning, comprising: S1, collecting multi-source geoscientific information data from geological, geochemical, and remote sensing sources; S2, establishing a conceptual model of a metallogenic system; S3, extracting multi-source spatial proxy mineralization indicator information based on the metallogenic system conceptual model; S4, completing data integration and training based on a neural network to obtain a multi-dimensional spatial proxy layer dataset and training points; S5, inputting the multi-dimensional spatial proxy layer dataset and training points, applying machine learning algorithms to optimize hyperparameters, and obtaining an optimized machine learning model; S6, applying the optimized machine learning model to complete the evaluation of machine learning results and target area delineation.

[0004] Therefore, it can be seen that the method of delineating mineral exploration target areas based on multi-source heterogeneous information using machine learning has the following problems: because mineral exploration prediction is carried out by fusing multi-source geological information and machine learning, the consistency of geological features from different sources is not verified, making it difficult to trace the source of abnormal data and resulting in insufficient targeted supplementary data collection. Summary of the Invention

[0005] To address this, the present invention provides a quantitative prediction method for prospective mineral deposits based on the fusion of multi-source geological data. This method overcomes the shortcomings of existing technologies that rely on the fusion of multi-source geoscientific information and machine learning for mineral exploration prediction. These technologies fail to verify the consistency of geological features from different sources, making it difficult to trace the source of abnormal data and resulting in insufficient targeted supplementary data collection.

[0006] To achieve the above objectives, this invention provides a quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion, comprising: Acquire gravity exploration data, geochemical sampling data, seismic exploration data, and historical borehole data within the exploration area, and correlate them with the corresponding spatial prediction units to form corresponding multi-source geological features; Establish cross-source geological evidence relationships based on the spatial correspondence between the multi-source geological features, and determine the consistency of multi-source geological features according to the corresponding state of various geological features along the cross-source geological evidence relationships, so as to determine the source of initial abnormal data when the consistency does not meet the requirements. Based on the consistent multi-source geological features and corresponding historical borehole data, a mineralization prediction model is constructed to predict the mineralization favorability of the exploration area and obtain the prediction results of the mineralization favorable area. The prediction results of the favorable metallogenic areas are reviewed based on the consistency of spatial variation of multi-source geological characteristics. When historical borehole data exists, the prediction results of the favorable metallogenic areas are further reviewed in conjunction with the historical borehole data. If the verification results do not meet the requirements, the multi-source geological features formed by the prediction results are traced back to determine the source of the target anomaly data and the target supplementary collection location, and the corresponding supplementary data collection is performed. The multi-source geological features and metallogenic favorable area prediction results are updated based on the supplemented data, and a quantitative prediction map of metallogenic prospective areas is generated based on the metallogenic favorable area prediction results that meet the requirements of the verification results.

[0007] Furthermore, the formation of the corresponding multi-source geological features includes: Based on the changes in gravity acceleration and spatial distribution of gravity at each gravity measuring point in the gravity survey data, the gravity anomaly areas and their extension states corresponding to the changes in underground medium density are determined to form gravity anomaly characteristics. Based on the occurrence state and spatial distribution relationship of the elements corresponding to each sampling point in the geochemical sampling data, the geochemical anomaly areas and their extension states corresponding to element enrichment, depletion or combination anomalies are determined to form geochemical anomaly characteristics. Based on the propagation, reflection, and continuous changes of seismic waves in seismic exploration data, the seismic anomaly areas and their extensions corresponding to changes in subsurface media interfaces or geological structures are identified to form seismic response characteristics. Furthermore, based on the spatial correspondence between the aforementioned multi-source geological features, cross-source geological evidence relationships are established, including: Based on the spatial location and anomalous extension state corresponding to the gravity anomaly characteristics, geochemical anomaly characteristics, and seismic response characteristics, geological features from different sources with spatial correspondence are associated to form the cross-source geological evidence relationship.

[0008] Furthermore, the determination of the consistency of multi-source geological features includes: When geological features form mutually supporting spatial correspondences along the aforementioned cross-source geological evidence relationships, the consistency is determined to meet the requirements; When at least one type of geological feature cannot form a mutually supportive spatial correspondence with the other geological features, the consistency is determined to be non-compliant, and the source of the initial abnormal data is determined based on the correspondence and spatial continuity between various types of geological features.

[0009] Furthermore, the construction of the mineralization prediction model includes: Correlate gravity anomaly features, geochemical anomaly features, and seismic response features that meet the consistency requirements to form a joint geological feature; Based on historical borehole data, the mineralization state corresponding to the combined geological features is determined, and the correspondence between the combined geological features and the mineralization state is established to construct the mineralization prediction model.

[0010] Furthermore, the predicted results of the favorable mineralization areas are reviewed, including: Based on the continuous state of multi-source geological features between adjacent spatial prediction units, determine whether the spatial variation of the prediction results of the corresponding metallogenic favorable area corresponds to the variation of the multi-source geological features; For spatial prediction units with historical borehole data, the verification results are further determined based on the correspondence between the predicted results of the favorable mineralization areas and the corresponding historical borehole data.

[0011] Furthermore, determining the source of the abnormal target data and the location for supplementary target data collection includes: Based on the correlation between spatial prediction units that do not meet the requirements of the review results and their corresponding multi-source geological features, the various geological features formed by the prediction results are traced back in reverse. Based on the corresponding status of various geological features along the cross-source geological evidence relationship during the reverse tracing process, the source of the target anomaly data and the target supplementary collection location are determined.

[0012] Furthermore, the corresponding data supplementary collection includes: Depending on the data type of the source of the target anomaly data, supplementary acquisition of gravity exploration data, geochemical sampling data, or seismic exploration data is performed at the target supplementary acquisition location.

[0013] Furthermore, the prediction results of the multi-source geological features and favorable metallogenic areas are updated, including: The corresponding geological features are updated based on the supplementary data collected, and the mineralization favorableness prediction and prediction results are re-executed. If the verification results still do not meet the requirements, the source of the target abnormal data is re-identified and the corresponding supplementary data collection is performed.

[0014] Furthermore, the generation of the quantitative prediction map for the mineralization prospect area includes: The prediction results of the mineralization favorable areas corresponding to each spatial prediction unit that meets the verification requirements are mapped according to their spatial location, and the spatial distribution range of the mineralization favorable areas is determined based on the prediction results of adjacent spatial prediction units, so as to generate the quantitative prediction map of the mineralization prospect area.

[0015] Compared with existing technologies, the beneficial effects of this invention are as follows: the method of this invention acquires gravity exploration data, geochemical sampling data, seismic exploration data, and historical borehole data, and associates geological data from different sources with corresponding spatial prediction units to form gravity anomaly features, geochemical anomaly features, and seismic response features; it establishes cross-source geological evidence relationships through the spatial correspondence between various geological features, and performs consistency judgment based on the correspondence status between geological features from different sources, enabling geological anomaly information obtained from different physical mechanisms to mutually verify each other; furthermore, it constructs a mineralization prediction model using multi-source geological features that meet consistency requirements, and reviews the prediction results by combining the spatial variation of multi-source geological features and historical borehole data; when the review results are abnormal, it traces back to the source of the target abnormal data and performs targeted supplementary acquisition, so that data acquisition, feature fusion, mineralization prediction, and anomaly correction form a feedback closed loop, thereby reducing the impact of local abnormal data on mineralization prediction results and improving the reliability of quantitative prediction of mineralization prospective areas.

[0016] Furthermore, the method described in this invention establishes cross-source geological evidence relationships and determines the consistency of multi-source geological features. Since gravity exploration, geochemical sampling, and seismic exploration respectively reflect changes in underground medium density, elemental occurrence anomalies, and underground tectonic responses, the observation objects and spatial manifestations of data from different sources differ. This makes it difficult to determine whether the corresponding anomaly can truly reflect the metallogenic geological conditions based solely on the anomaly information from any one source, and local anomalies are easily mistaken for actual geological anomalies. By associating gravity anomaly features, geochemical anomaly features, and seismic response features according to their spatial location and anomaly extension state, and determining their consistency based on the mutual support of various geological features along the cross-source geological evidence relationships, the spatial correspondence between different geological evidences can be used to identify anomaly features lacking support from other sources of geological evidence. This reduces the interference caused by single-source anomaly data directly participating in metallogenic prediction, further improving the reliability of quantitative prediction of prospective metallogenic areas.

[0017] Furthermore, the method described in this invention verifies the prediction results of favorable mineralization areas. Since the mineralization prediction model determines the favorable mineralization state based on the combination relationship between multiple geological features, local geological data anomalies, changes in feature spatial relationships, or insufficient training data coverage may cause the prediction results of some spatial prediction units to be inconsistent with the surrounding geological evolution trend. It is difficult to identify the corresponding abnormal prediction results in a timely manner based solely on the model's own output. By analyzing the correspondence between the prediction results of the favorable mineralization areas of the target spatial prediction unit and the continuous state of the multi-source geological features of adjacent spatial prediction units, and further verifying the prediction results in conjunction with the mineralization state or mineralization occurrence state revealed by the boreholes at locations with historical borehole data, the prediction results can be verified from two levels: spatial continuity and actual borehole evidence. This allows for the identification of abnormal prediction results that do not correspond to changes in actual geological features, providing a basis for subsequent abnormal data tracing and further improving the reliability of quantitative prediction of prospective mineralization areas.

[0018] Furthermore, the method described in this invention reverse-tracks and performs targeted data supplementation collection on prediction results that do not meet the requirements of the verification results. Since geological data from different sources participate in mineralization prediction, it is difficult to directly determine the source of the anomaly when the prediction result is abnormal, whether it is gravity exploration data, geochemical sampling data, or seismic exploration data. If a unified supplementary collection is performed on the exploration area directly, it is easy to generate untargeted duplicate exploration and increase the workload of data collection. By reverse-tracking based on the correlation between the abnormal prediction result and the corresponding multi-source geological features, and combining the cross-source geological evidence relationship to determine the source of the target abnormal data and the corresponding target supplementary collection location, corresponding type of data supplementation collection is performed on the target location, and the geological features and mineralization favorable area prediction results are updated using the supplementary collection data. This allows the data supplementation collection to be implemented on the data source that caused the prediction anomaly, thereby reducing invalid duplicate collection and reducing the possibility that abnormal data will continue to affect the prediction results, further improving the reliability of quantitative prediction of mineralization prospective areas. Attached Figure Description

[0019] Figure 1 This is an overall flowchart of the quantitative prediction method for mineralized prospective areas based on multi-source geological data fusion, as described in this embodiment of the invention. Figure 2 This is a flowchart illustrating the process of determining the consistency of multi-source geological features in a quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion, as described in an embodiment of the present invention. Figure 3 This is a flowchart illustrating the process of verifying the prediction results in the quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion, as described in an embodiment of the present invention. Figure 4 This is a flowchart illustrating the abnormal data tracing and targeted supplementary data collection process of the quantitative prediction method for mineralized prospective areas based on multi-source geological data fusion, as described in an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0021] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0022] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0023] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; 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; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0024] Please see Figure 1 The diagram shown is an overall flowchart of a quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to an embodiment of the present invention. The method described in this embodiment includes: Step S1: Obtain gravity exploration data, geochemical sampling data, seismic exploration data and historical borehole data within the exploration area, and associate them with the corresponding spatial prediction units to form corresponding multi-source geological features; Step S2: Establish cross-source geological evidence relationships based on the spatial correspondence between the multi-source geological features, and determine the consistency of the multi-source geological features according to the corresponding state of various geological features along the cross-source geological evidence relationships, so as to determine the source of the initial abnormal data when the consistency does not meet the requirements. Step S3: Based on the consistent multi-source geological features and corresponding historical borehole data, construct a mineralization prediction model, predict the mineralization favorability of the exploration area, and obtain the prediction results of the mineralization favorable area. Step S4: Verify the prediction results of the favorable mineralization area based on the consistency of the spatial variation of the multi-source geological characteristics, and further verify the prediction results of the favorable mineralization area in combination with the historical borehole data when historical borehole data exists. If the verification results do not meet the requirements, the multi-source geological features formed by the prediction results are traced back to determine the source of the target anomaly data and the target supplementary collection location, and the corresponding supplementary data collection is performed. Step S5: Update the multi-source geological features and mineralization favorable area prediction results based on the supplemented data, and generate a quantitative prediction map of mineralization prospective areas based on the mineralization favorable area prediction results that meet the requirements of the verification results.

[0025] Specifically, a spatial prediction unit is a basic prediction area formed by dividing the exploration area according to a unified spatial coordinate relationship. It is used to associate gravity exploration data, geochemical sampling data, seismic exploration data and historical borehole data within the same area, and serves as the spatial counterpart of the mineralization favorability prediction results.

[0026] Specifically, the consistent state refers to the spatial variation of the predicted results of favorable mineralization areas between adjacent spatial prediction units, which corresponds to the abnormal distribution and continuous variation of the corresponding multi-source geological features in the same area.

[0027] In this embodiment, the method of the present invention acquires gravity exploration data, geochemical sampling data, seismic exploration data, and historical borehole data, and associates geological data from different sources with corresponding spatial prediction units to form gravity anomaly features, geochemical anomaly features, and seismic response features. It establishes cross-source geological evidence relationships through the spatial correspondence between various geological features and performs consistency judgment based on the correspondence states between geological features from different sources, enabling geological anomaly information obtained from different physical mechanisms to mutually verify each other. Furthermore, it constructs a mineralization prediction model using multi-source geological features that meet consistency requirements, and verifies the prediction results by combining the spatial variations of multi-source geological features and historical borehole data. When the verification results are abnormal, it traces back to the source of the target anomaly data and performs targeted supplementary acquisition, forming a feedback loop of data acquisition, feature fusion, mineralization prediction, and anomaly correction. This reduces the impact of local anomaly data on the mineralization prediction results and improves the reliability of quantitative prediction of prospective mineralization areas.

[0028] Specifically, the corresponding multi-source geological features include: Based on the changes in gravity acceleration and spatial distribution of gravity at each gravity measuring point in the gravity survey data, the gravity anomaly areas and their extension states corresponding to the changes in underground medium density are determined to form gravity anomaly characteristics. Based on the occurrence state and spatial distribution relationship of the elements corresponding to each sampling point in the geochemical sampling data, the geochemical anomaly areas and their extension states corresponding to element enrichment, depletion or combination anomalies are determined to form geochemical anomaly characteristics. Based on the propagation, reflection, and continuous changes of seismic waves in seismic exploration data, the seismic anomaly areas and their extension states corresponding to changes in underground medium interfaces or geological structures are determined to form seismic response characteristics.

[0029] Specifically, establishing cross-source geological evidence relationships based on the spatial correspondence between the aforementioned multi-source geological features includes: Based on the spatial location and anomalous extension state corresponding to the gravity anomaly characteristics, geochemical anomaly characteristics, and seismic response characteristics, geological features from different sources with spatial correspondence are associated to form the cross-source geological evidence relationship.

[0030] Specifically, cross-source geological evidence relationships are the correlations formed by geological features from different sources based on spatial location, anomalous extension state, and continuity of adjacent areas. They are used to characterize whether gravity anomaly features, geochemical anomaly features, and seismic response features can jointly indicate the same or related geological anomaly state.

[0031] In this embodiment, establishing cross-source geological evidence relationships includes: The exploration area is divided into several spatial prediction units, and the gravity anomaly characteristics, geochemical anomaly characteristics, and seismic response characteristics corresponding to each spatial prediction unit are correlated according to a unified spatial coordinate relationship. Based on the spatial location, anomalous extension state, and continuity relationship between adjacent spatial prediction units corresponding to geological features from different sources, determine whether there is a spatial succession relationship between various geological features; For geological features from different sources that can form continuous correspondences in spatial location and anomalous extension relationships, establish a correlation between them to form cross-source geological evidence relationships, and provide a basis for subsequent determination of the consistency of multi-source geological features.

[0032] Please see Figure 2 As shown, this is a logical flowchart of the process for determining the consistency of multi-source geological features according to an embodiment of the present invention. The consistency of multi-source geological features is determined based on the corresponding states of various geological features along the cross-source geological evidence relationships. When geological features form mutually supporting spatial correspondences along the aforementioned cross-source geological evidence relationships, the consistency is determined to meet the requirements; When at least one type of geological feature cannot form a mutually supportive spatial correspondence with the other geological features, the consistency is determined to be non-compliant, and the source of the initial abnormal data is determined based on the correspondence and spatial continuity between various types of geological features.

[0033] Specifically, the consistency of multi-source geological features refers to the spatial correspondence and mutual support between gravity anomaly features, geochemical anomaly features, and seismic response features along the cross-source geological evidence relationship; the consistency refers to the ability of different geological features to form a reasonable correspondence in terms of anomaly location, extension trend, and continuous change relationship.

[0034] Specifically, the initial source of anomalous data is the data source that may contain anomalies, determined before mineralization prediction based on the cross-source geological evidence relationship and spatial continuity between geological features from different sources. This includes gravity exploration data, geochemical sampling data, or seismic exploration data, in order to avoid anomalous geological data from directly participating in subsequent mineralization prediction.

[0035] In this embodiment, determining the consistency of multi-source geological features and identifying the source of initial anomaly data includes: Gravity anomaly features, geochemical anomaly features, and seismic response features that have established cross-source geological evidence relationships within the same spatial prediction unit are obtained, and the spatial correspondence between various geological features is determined along the cross-source geological evidence relationships. When the underground anomaly region reflected by gravity anomaly features can form a spatial connection with the elemental anomaly region corresponding to geochemical anomaly features, and at the same time correspond to the underground structural changes reflected by seismic response features, and the three types of geological features maintain continuous changes between adjacent spatial prediction units, it is determined that the multi-source geological features can support each other, and their consistency meets the requirements. When at least one type of geological feature cannot form a continuous spatial correspondence with the other geological features along the cross-source geological evidence relationship, it is determined that the consistency of the multi-source geological features does not meet the requirements; further compare the correspondence between the inconsistent geological features and the other geological features, as well as the continuity of various geological features in adjacent spatial prediction units. If one type of geological feature cannot form a continuous correspondence with the other geological features, but the other geological features can support each other and maintain continuity between adjacent spatial prediction units, then the geological feature that cannot form a continuous correspondence is identified as an abnormal geological feature, and its corresponding exploration data is identified as the source of initial abnormal data. If inconsistencies exist among multiple types of geological features, further judgment should be made by combining the continuous state of geological features in adjacent spatial prediction units to avoid directly determining the source of anomalies based solely on local anomalies in a single spatial prediction unit.

[0036] In this embodiment, the method of the present invention establishes cross-source geological evidence relationships and determines the consistency of multi-source geological features. Since gravity exploration, geochemical sampling, and seismic exploration respectively reflect changes in underground medium density, elemental occurrence anomalies, and underground tectonic responses, the observation objects and spatial manifestations of data from different sources differ. This makes it difficult to determine whether the corresponding anomaly can truly reflect the metallogenic geological conditions based solely on the anomaly information from any one source, and local anomalies are easily mistaken for actual geological anomalies. By associating gravity anomaly features, geochemical anomaly features, and seismic response features according to their spatial location and anomaly extension state, and determining their consistency based on the mutual support of various geological features along the cross-source geological evidence relationships, the spatial correspondence between different geological evidences can be used to identify anomaly features lacking support from other sources of geological evidence. This reduces the interference caused by single-source anomaly data directly participating in metallogenic prediction, and further improves the reliability of quantitative prediction of prospective metallogenic areas.

[0037] Specifically, the construction of the mineralization prediction model includes: Correlate gravity anomaly features, geochemical anomaly features, and seismic response features that meet the consistency requirements to form a joint geological feature; Based on historical borehole data, the mineralization state corresponding to the combined geological features is determined, and the correspondence between the combined geological features and the mineralization state is established to construct the mineralization prediction model.

[0038] Please see Figure 3 The diagram shown is a logical flowchart illustrating the process of verifying the prediction results according to an embodiment of the present invention. The prediction results are verified based on the correspondence between the predicted favorable mineralization areas and the corresponding multi-source geological features and historical borehole data. The review of the predicted favorable mineralization areas includes: Based on the continuous state of multi-source geological features between adjacent spatial prediction units, determine whether the spatial variation of the prediction results of the corresponding metallogenic favorable area corresponds to the variation of the multi-source geological features; For spatial prediction units with historical borehole data, the verification results are further determined based on the correspondence between the predicted results of the favorable mineralization areas and the corresponding historical borehole data.

[0039] In this embodiment, the prediction results of favorable mineralization areas are reviewed, including: Obtain the prediction results of multi-source geological features and metallogenic favorable areas corresponding to the target spatial prediction unit and its adjacent spatial prediction units. Based on the continuous change state of multi-source geological features between adjacent spatial prediction units, determine whether the spatial change of the corresponding prediction results corresponds to the change of geological features. For spatial prediction units with historical borehole data, the prediction results of the favorable mineralization area are further correlated with the mineralization state or mineralization occurrence state revealed by the borehole. If the prediction results correspond to the spatial variations of multi-source geological features, and if historical borehole data is available, they correspond to the borehole reveal status, then the verification results are deemed to meet the requirements; otherwise, the verification results are deemed to not meet the requirements, and the corresponding spatial prediction unit is used as the object for subsequent anomaly tracing.

[0040] In this embodiment, the method of the present invention verifies the prediction results of favorable mineralization areas. Since the mineralization prediction model determines the favorable mineralization state based on the combination relationship between multiple geological features, local geological data anomalies, changes in feature spatial relationships, or insufficient training data coverage may cause the prediction results of some spatial prediction units to be inconsistent with the surrounding geological evolution trend. It is difficult to identify the corresponding abnormal prediction results in a timely manner based solely on the model's own output. By analyzing the correspondence between the prediction results of the favorable mineralization areas of the target spatial prediction unit and the continuous state of the multi-source geological features of adjacent spatial prediction units, and further verifying the prediction results in conjunction with the mineralization state or mineralization occurrence state revealed by the boreholes at locations with historical borehole data, the prediction results can be verified from two levels: spatial continuity and actual borehole evidence. This allows for the identification of abnormal prediction results that do not correspond to actual geological feature changes, providing a basis for subsequent abnormal data tracing and further improving the reliability of quantitative prediction of prospective mineralization areas.

[0041] Please see Figure 4 As shown, it is a logical flowchart of the abnormal data tracing and targeted supplementary collection process according to an embodiment of the present invention. The step of determining the source of the target abnormal data and the target supplementary collection location includes: Based on the correlation between spatial prediction units that do not meet the requirements of the review results and their corresponding multi-source geological features, the various geological features formed by the prediction results are traced back in reverse. Based on the corresponding status of various geological features along the cross-source geological evidence relationship during the reverse tracing process, the source of the target anomaly data and the target supplementary collection location are determined.

[0042] The execution of corresponding supplementary data collection includes: Depending on the data type of the source of the target anomaly data, supplementary acquisition of gravity exploration data, geochemical sampling data, or seismic exploration data is performed at the target supplementary acquisition location.

[0043] Specifically, reverse tracing involves starting from the predicted results of mineralized favorable areas where the verification results do not meet the requirements, and then tracing step by step along the directions of the predicted results, multi-source geological features, cross-source geological evidence relationships, and corresponding exploration data based on the correlation between the predicted results and the multi-source geological features involved in the prediction, in order to determine the data sources that caused the abnormal prediction results.

[0044] Specifically, the source of the target anomaly data is the data source corresponding to the anomaly prediction results determined by reverse tracing of the multi-source geological features involved in the prediction after the prediction results in the favorable mineralization area do not meet the requirements, and by combining the cross-source geological evidence relationship. This includes target gravity exploration data, target geochemical sampling data, or target seismic exploration data.

[0045] Specifically, the target supplementary acquisition location is the location to be supplemented for exploration, determined based on the location where the geological features corresponding to the source of the target anomaly data have spatial gaps, anomalous changes, or lose correspondence with other source geological features in the cross-source geological evidence relationship. It is used to limit the scope of data supplementation for subsequent gravity exploration, geochemical sampling, or seismic exploration.

[0046] In this embodiment, the abnormal prediction results are traced back to determine the source of the target abnormal data, including: When the verification results of the target spatial prediction unit do not meet the requirements, the gravity anomaly characteristics, geochemical anomaly characteristics and seismic response characteristics involved in the formation of the prediction results are determined based on the correlation between the prediction results of the metallogenic favorable area corresponding to the spatial prediction unit and the input data of the metallogenic prediction model. By tracing back along the cross-source geological evidence relationships corresponding to various geological features, the continuous state between various geological features and the corresponding geological features of adjacent spatial prediction units, as well as the spatial correspondence between them and geological features from other sources, is determined. If a certain type of geological feature is abnormally interrupted at the target spatial prediction unit, while the other geological features can still maintain a continuous correspondence along the original cross-source geological evidence relationship, then the geological feature that is abnormally interrupted is identified as the target abnormal geological feature, and its corresponding exploration data is identified as the source of the target abnormal data. For spatial prediction units with historical borehole data, the correspondence between various geological features and borehole reveal status is further verified. If a certain type of geological feature cannot form a mutually supportive relationship with the other geological features, nor can it correspond to the borehole reveal status, then the data corresponding to that type of geological feature is preferentially identified as the source of target anomaly data.

[0047] In this embodiment, determining the target supplementary data acquisition location and performing directional supplementary data acquisition includes: After determining the source of the target anomaly data, based on the location where the target anomaly geological features have spatial gaps, abnormal changes, or lose correspondence with other source geological features along the cross-source geological evidence relationship, the target supplementary collection area is determined, and the actual exploration location in this area corresponding to the source of the target anomaly data is determined as the target supplementary collection location. When the source of abnormal data for the target is gravity survey data, gravity survey data is re-collected at the gravity measurement points corresponding to the target's supplementary collection location or at related gravity measurement points that have a spatial continuity with it. When the source of the target anomaly data is geochemical sampling data, soil or rock samples are collected in the geochemical anomaly gap area or anomaly extension area corresponding to the target supplementary sampling location; when the source of the target anomaly data is seismic exploration data, seismic exploration data is reacquired in the seismic survey line section, shot point or signal receiving location corresponding to the target supplementary sampling location. Based on the supplemented data, the corresponding gravity anomaly features, geochemical anomaly features, or seismic response features are reconstructed, and the updated geological features are re-incorporated into the multi-source geological features of the corresponding spatial prediction unit to update the cross-source geological evidence relationship and the input data of the metallogenic prediction model. Based on the updated multi-source geological features, the mineralization aptitude prediction was re-executed, and the prediction results were reviewed again according to the aforementioned review process. If the updated prediction results correspond to the spatial variation of multi-source geological features and historical borehole data, the verification results are deemed to meet the requirements, and the updated prediction results are maintained; if the verification results still do not meet the requirements, the process of tracing the source of abnormal data and determining the target supplementary collection location is re-executed.

[0048] In this embodiment, the method of the present invention reverse-tracks the prediction results that do not meet the requirements and performs targeted data supplementation. Since geological data from different sources participate in mineralization prediction, when anomalies occur in the prediction results, it is difficult to directly determine whether the anomaly originates from gravity exploration data, geochemical sampling data, or seismic exploration data. If unified supplementation is performed on the exploration area directly, it is easy to generate untargeted duplicate exploration and increase the workload of data collection. By reverse-tracking based on the correlation between the anomaly prediction results and the corresponding multi-source geological features, and combining cross-source geological evidence relationships to determine the source of the target anomaly data and the corresponding target supplementation location, corresponding type of data supplementation is performed for the target location. Then, the geological features and mineralization favorable area prediction results are updated using the supplemented data. This allows the data supplementation to be implemented for the data source that caused the prediction anomaly, thereby reducing invalid duplicate collection and reducing the possibility that anomaly data will continue to affect the prediction results, further improving the reliability of quantitative prediction of mineralization prospective areas.

[0049] Specifically, updating the prediction results of the multi-source geological features and favorable metallogenic areas includes: The corresponding geological features are updated based on the supplementary data collected, and the mineralization favorableness prediction and prediction results are re-executed. If the verification results still do not meet the requirements, the source of the target abnormal data is re-identified and the corresponding supplementary data collection is performed.

[0050] Specifically, the generation of the quantitative prediction map of the mineralization prospect area includes: The prediction results of the mineralization favorable areas corresponding to each spatial prediction unit that meets the verification requirements are mapped according to their spatial location, and the spatial distribution range of the mineralization favorable areas is determined based on the prediction results of adjacent spatial prediction units, so as to generate the quantitative prediction map of the mineralization prospect area.

[0051] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion, characterized in that, include: Acquire gravity exploration data, geochemical sampling data, seismic exploration data, and historical borehole data within the exploration area, and correlate them with the corresponding spatial prediction units to form corresponding multi-source geological features; Establish cross-source geological evidence relationships based on the spatial correspondence between the multi-source geological features, and determine the consistency of multi-source geological features according to the corresponding state of various geological features along the cross-source geological evidence relationships, so as to determine the source of initial abnormal data when the consistency does not meet the requirements. Based on the consistent multi-source geological features and corresponding historical borehole data, a mineralization prediction model is constructed to predict the mineralization favorability of the exploration area and obtain the prediction results of the mineralization favorable area. The prediction results of the favorable metallogenic areas are reviewed based on the consistency of spatial variation of multi-source geological characteristics. When historical borehole data exists, the prediction results of the favorable metallogenic areas are further reviewed in conjunction with the historical borehole data. If the verification results do not meet the requirements, the multi-source geological features formed by the prediction results are traced back to determine the source of the target anomaly data and the target supplementary collection location, and the corresponding supplementary data collection is performed. The multi-source geological features and metallogenic favorable area prediction results are updated based on the supplemented data, and a quantitative prediction map of metallogenic prospective areas is generated based on the metallogenic favorable area prediction results that meet the requirements of the verification results.

2. The method according to claim 1, characterized in that, The corresponding multi-source geological features include: Based on the changes in gravity acceleration and spatial distribution of gravity at each gravity measuring point in the gravity survey data, the gravity anomaly areas and their extension states corresponding to the changes in underground medium density are determined to form gravity anomaly characteristics. Based on the occurrence state and spatial distribution relationship of the elements corresponding to each sampling point in the geochemical sampling data, the geochemical anomaly areas and their extension states corresponding to element enrichment, depletion or combination anomalies are determined to form geochemical anomaly characteristics. Based on the propagation, reflection, and continuous changes of seismic waves in seismic exploration data, the seismic anomaly areas and their extension states corresponding to changes in underground medium interfaces or geological structures are determined to form seismic response characteristics.

3. The method according to claim 2, characterized in that, Establishing cross-source geological evidence relationships based on the spatial correspondence between the aforementioned multi-source geological features includes: Based on the spatial location and anomalous extension state corresponding to the gravity anomaly characteristics, geochemical anomaly characteristics, and seismic response characteristics, geological features from different sources with spatial correspondence are associated to form the cross-source geological evidence relationship.

4. The method according to claim 3, characterized in that, The determination of the consistency of multi-source geological features includes: When geological features form mutually supporting spatial correspondences along the aforementioned cross-source geological evidence relationships, the consistency is determined to meet the requirements; When at least one type of geological feature cannot form a mutually supportive spatial correspondence with the other geological features, the consistency is determined to be non-compliant, and the source of the initial abnormal data is determined based on the correspondence and spatial continuity between various types of geological features.

5. The quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to claim 4, characterized in that, The constructed mineralization prediction model includes: Correlate gravity anomaly features, geochemical anomaly features, and seismic response features that meet the consistency requirements to form a joint geological feature; Based on historical borehole data, the mineralization state corresponding to the combined geological features is determined, and the correspondence between the combined geological features and the mineralization state is established to construct the mineralization prediction model.

6. The quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to claim 5, characterized in that, The review of the predicted favorable mineralization areas includes: Based on the continuous state of multi-source geological features between adjacent spatial prediction units, determine whether the spatial variation of the prediction results of the corresponding favorable mineralization area corresponds to the variation of the multi-source geological features; For spatial prediction units with historical borehole data, the verification results are further determined based on the correspondence between the predicted results of the favorable mineralization areas and the corresponding historical borehole data.

7. The quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to claim 6, characterized in that, The determination of the source of the abnormal data and the location for supplementary data collection includes: Based on the correlation between spatial prediction units that do not meet the requirements of the review results and their corresponding multi-source geological features, the various geological features formed by the prediction results are traced back in reverse. Based on the corresponding status of various geological features along the cross-source geological evidence relationship during the reverse tracing process, the source of the target anomaly data and the target supplementary collection location are determined.

8. The quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to claim 7, characterized in that, The execution of corresponding supplementary data collection includes: Depending on the data type of the source of the target anomaly data, supplementary acquisition of gravity exploration data, geochemical sampling data, or seismic exploration data is performed at the target supplementary acquisition location.

9. The quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to claim 8, characterized in that, Update the prediction results of the multi-source geological features and favorable metallogenic areas, including: The corresponding geological features are updated based on the supplemented data, and the mineralization favorableness prediction and prediction results are re-executed. If the verification results still do not meet the requirements, the source of the target abnormal data is re-identified and the corresponding supplementary data collection is performed.

10. The quantitative prediction method for metallogenic prospective areas based on multi-source geological data fusion according to claim 9, characterized in that, The generated quantitative prediction map of mineralization prospective areas includes: The prediction results of the mineralization favorable areas corresponding to each spatial prediction unit that meets the verification requirements are mapped according to their spatial location, and the spatial distribution range of the mineralization favorable areas is determined based on the prediction results of adjacent spatial prediction units, so as to generate the quantitative prediction map of the mineralization prospect area.

Citation Information

Patent Citations

  • Method for delineating prospecting target area through multi-source heterogeneous information based on machine learning

    CN116611697A