CN-121978772-A - Method for rapidly delineating gold mine prospecting target area

Abstract

The invention relates to the technical field of geology science, in particular to a method for rapidly delineating a gold mine prospecting target area, which comprises the following steps of firstly constructing a mineral deposit geological feature library, secondly integrating multi-source heterogeneous data and dynamically compiling basic drawing pieces, thirdly reconstructing a structure-mineralization event sequence and dynamically dividing a mineralization system, fourthly constructing a regional mineral deposit pedigree and extracting a representative mineral deposit combination in a multi-method collaborative manner, fifthly analyzing multi-period abnormal superposition and extracting a dynamic mineralization geological necessary condition, sixthly analyzing dynamic element combination and modeling non-linear integrated prediction, and seventhly classifying and dynamically verifying target areas.

Inventors

• ZHANG ZHIYUAN
• WANG BINSHAN
• CHANG CHAOHONG

Assignees

• 河北地质大学

Dates

Publication Date

20260505

Application Date

20260121

Claims (8)

1. 1. The method for rapidly delineating the gold mine prospecting target area is characterized by comprising the following steps: firstly, constructing a deposit geological feature library, namely acquiring and structuring deposit geological features of all known gold deposits and mine points in a research area to form a feature database; Integrating a geological map of a research area, multi-period isotope annual survey data, structural deformation measurement data and geophysical, geochemical, remote sensing and heavy sand original data, constructing a multi-dimensional data base, and compiling a dynamic geological mineral map and a structural-thermal event sequence map to form a gold ore prediction base map; constructing a multi-stage construction-thermal event sequence after a mountain making period of a research area based on a multi-dimensional data base, dividing a dynamic ore forming system main class according to the multi-stage construction-thermal event sequence, and analyzing the overlapping and transformation relations of the ore forming systems at different stages in space; Dividing ore deposit lineages according to the structure-ore formation event sequence and ore deposit geological characteristics by adopting a clustering algorithm, selecting representative ore deposit combinations of the ore deposit lineages, carrying out cross setting by utilizing a multi-isotope setting technology, determining a main ore formation period and a later transformation period, and constructing a dynamic ore formation mode of multi-period ore formation element superposition; Identifying and distinguishing remote sensing alteration anomalies of different periods by utilizing a machine learning image segmentation technology based on the ore deposit pedigree and the dynamic ore formation mode, establishing a staged anomaly identification mark by combining a geochemical element combination and geophysical anomaly characteristics, and extracting the dynamic ore formation geological requirements; Under the constraint of the structure-thermal event sequence, dividing a geological-structure domain, identifying the exclusive element combinations of different periods of ore deposit in each geological-structure domain, training a nonlinear gold mineralization strength prediction model by adopting a machine learning algorithm, and generating a continuous mineralization benefit index; And step seven, grading dynamic verification of the target area, namely delineating and grading and predicting the target area by utilizing a three-dimensional geologic modeling technology based on the diagenetic vantage index and the dynamic diagenetic geological requirement.
2. 2. The method for rapidly delineating a gold mine prospecting target area according to claim 1, wherein the gold mine prediction basis map is formed by the steps of: Preprocessing the multi-source heterogeneous data, namely preprocessing coordinate system unification, format standardization and quality inspection on the multi-source heterogeneous data; The construction of a multidimensional data base, namely importing the preprocessed multi-source heterogeneous data into three-dimensional geological modeling software, carrying out spatial registration and time attribute association according to geographic coordinates and stratum age, and establishing the multidimensional data base comprising spatial geometry, stratum age, rock attribute, construction element, geophysical field, geochemical field and thermal event age label; dynamically extracting and rendering stratum boundary lines, magma rock masses, structure trails and known deposit distribution information in a corresponding space-time range based on time slices or structure-thermal event periods set by a user under the support of the multidimensional data base, and generating a dynamic geological mineral map reflecting geological mineral faces in a specific geological history period; Interpreting and generating a construction-thermal event sequence diagram, namely identifying a concentrated interval of age data by adopting a density clustering algorithm based on multi-period isotope annual measurement data and construction deformation data in the multi-dimensional data base to calibrate main thermal event periods, judging a sequence of the main thermal event periods by combining a space intersection relation of the construction deformation patterns, and drawing the construction-thermal event sequence diagram taking time as a vertical axis; and (3) carrying out association analysis on the dynamic geological mineral map and the structure-thermal event sequence map, and marking the cause association of a specific mineral deposit and a specific structure-thermal event period by linking with attributes through spatial superposition in a multidimensional data base to synthesize a gold ore prediction base map.
3. 3. The method for rapidly delineating a gold mine prospecting target region according to claim 1, wherein the reconstruction flow of the structure-thermal event sequence is: The ancient construction grid recovery, namely placing a research area in a background of a belonging geoconstruction unit, recovering the ancient geographic position, the plate movement direction and the speed change track of the research area after the mountain making period by using a plate reconstruction model, and establishing a regional construction power background evolution frame; Event space-time visualization and clustering, namely visualizing isotope age data of a research area in a three-dimensional geological modeling environment according to space positions, rock units and measured mineral types, identifying concentrated distribution intervals of the age data on a time domain by using a space-time sequence matching algorithm and a cluster analysis method, associating the concentrated distribution intervals with specific structural motions by combining the time division of structural deformation data, defining structural-thermal events, and arranging according to the occurrence sequence of the structural-thermal events to form a primary structural-thermal event sequence.
4. 4. The method for rapidly delineating a gold mine prospecting target area according to claim 1, wherein in the fourth step, the classification of the mine lineages is specifically performed by classifying all known mine points in the research area by adopting a hierarchical clustering algorithm according to the cause type, the surrounding rock of the mine, the changed mineral combination and the time-space correlation with a specific structure-thermal event of the mine, so as to form a cause-period-coupling classification system.
5. 5. The method for rapidly delineating a target area for gold mine exploration as claimed in claim 4, wherein the construction of the dynamic ore formation pattern of multi-stage ore formation element superposition comprises the steps of dividing ore deposit lineages Within each of the deposit lineages, selecting a number of deposits as representative deposit combinations for that lineage; The multi-mineral-multi-method isotope cross dating and age determination comprises the steps of combining selected representative mineral deposits, collecting mineral samples of different generations for dating, carrying out cross verification by comparing and analyzing annual measurement results of different minerals in the same mineral deposit, determining a main mineral formation period, and identifying a later hydrothermal modification period or a constructional heat disturbance event; And (3) constructing a dynamic ore forming mode of multi-stage ore forming element superposition, namely integrating a mineral deposit pedigree, spatial distribution characteristics and ore forming and transformation time sequences, and constructing a multi-stage ore forming element superposition flow chart by taking a time axis as an order for each mineral deposit pedigree.
6. 6. The method for rapidly delineating a gold mine prospecting target area according to claim 1, wherein in the fifth step, the machine learning image segmentation technology is utilized to identify and distinguish different periods of remote sensing alteration anomalies, the training of multispectral or hyperspectral remote sensing images is performed by using a U-Net convolutional neural network model, and siliconization and sericite rock anomalies related to early deep hydrothermal activities and argillization anomalies related to late shallow activities are extracted.
7. 7. The method for rapidly delineating a gold mine prospecting target area according to claim 1, wherein in the sixth step, a machine learning algorithm is adopted to train a nonlinear gold mine intensity prediction model, specifically: Using known deposit points and mineralization points as label samples, using a stage abnormality identification mark as an input characteristic, training a model by using XGBoost or a support vector regression algorithm, and outputting an ore formation favorability index, wherein the ore formation favorability index is a continuous variable.
8. 8. The method for rapidly delineating a gold mine prospecting target area according to claim 1, wherein in the seventh step, the target area is classified according to the complexity of the mine forming system, the intensity and scale of multi-stage anomaly superposition, and the three-dimensional space resource potential estimation result, and the target area is classified into at least three classes of high-confidence deep target area, mid-shallow validation target area, and long-range exploration area.

Description

Method for rapidly delineating gold mine prospecting target area Technical Field The invention relates to the technical field of geological science, in particular to a method for rapidly delineating a gold mine prospecting target area. Background The traditional single method (such as geological map filling, chemical exploration and geophysical exploration) of gold mine serving as a global key strategic resource in resource exploration has the challenges of low efficiency, high cost, long period, difficult recognition of complex geological conditions and the like, and the rapid screening of a target area, the accurate positioning of an abnormal area, the quantitative evaluation of the ore forming potential and the intelligent optimization of an exploration path are realized by integrating multi-source heterogeneous data and combining an artificial intelligent algorithm to perform feature extraction, association analysis and pattern recognition on multi-dimensional information, so that the ore searching efficiency is improved, the exploration cost is reduced, invalid drilling is reduced, the gold mine discovery probability is improved, the traditional single method limitation is broken through multi-technology fusion, the hidden ore body discovery is accelerated, the exploration resource configuration is optimized to support the national resource safety and the local economic development, and the geological exploration mode is further promoted to be driven from experience to intelligent driving of data, and the deep cross fusion of geology and computer science and big data technology is promoted. In the prior art, the abnormal elements of the gold deposit in the research area are counted, the correlation coefficient of the abnormal elements and Au is calculated, then the comprehensive predicted value is calculated by the correlation coefficient, and then the lower limit of the Au prediction abnormality is calculated according to the geochemical element data processing method to carry out geochemical anomaly delineation, wherein the correlation coefficient is calculated by an R-type cluster analysis method, the R-type cluster analysis method assumes that the element correlation is stable in time-space dimension, but the element combination can be changed along with the construction period aiming at the multi-period construction activity, for example, the Au and the As are strongly correlated in the early ancient mining period, the Au and the Hg correlation in the medium-period transformation period are enhanced, and the traditional method can not dynamically capture the change. Therefore, a method for rapidly delineating a gold mine prospecting target area is provided. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides a method for rapidly delineating a gold mine prospecting target area. In order to achieve the above purpose, the present invention adopts the following technical scheme: a method for rapidly delineating a gold mine prospecting target area comprises the following steps: the method comprises the steps of constructing a deposit geological feature library, namely obtaining and structuring deposit geological features of each object through field geological investigation, core cataloging, system sampling and rock-ore identification and geochemistry and isotope analysis of all known gold deposits and mine points in a research area to form a feature database containing information of an ore-forming age, ore-forming lithology, alteration combination, mineral symbiotic sequences and element combination; Integrating high-precision geological map, multi-period isotope annual survey data, structural deformation measurement data and geophysical, geochemical, remote sensing and heavy sand initial data of a research area, constructing a space-time unified multi-dimensional data base by utilizing three-dimensional geological modeling software, and compiling a dynamic geological mineral map and a structure-thermal event sequence map to form a gold ore prediction base map supporting multi-period structure-ore formation event analysis; constructing a multi-stage construction-thermal event sequence after a mountain making period of a research area based on a multi-dimensional data base, dividing a dynamic ore forming system main class according to the multi-stage construction-thermal event sequence, and analyzing the overlapping and transformation relations of the ore forming systems at different stages in space; Dividing ore deposit lineages according to the structure-ore formation event sequence and ore deposit geological characteristics by adopting a clustering algorithm, selecting representative ore deposit combinations of the ore deposit lineages, carrying out cross setting by utilizing a multi-isotope setting technology, determining a main ore formation period and a later transformation period, and constructing a dynamic ore formation mode of multi-period ore format