CN117576335B - Three-dimensional space model data processing method and system for mineral area investigation - Google Patents
Three-dimensional space model data processing method and system for mineral area investigation Download PDFInfo
- Publication number
- CN117576335B CN117576335B CN202410079706.5A CN202410079706A CN117576335B CN 117576335 B CN117576335 B CN 117576335B CN 202410079706 A CN202410079706 A CN 202410079706A CN 117576335 B CN117576335 B CN 117576335B
- Authority
- CN
- China
- Prior art keywords
- geological
- data
- mineral
- spatial
- survey data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 229910052500 inorganic mineral Inorganic materials 0.000 title claims abstract description 293
- 239000011707 mineral Substances 0.000 title claims abstract description 293
- 238000011835 investigation Methods 0.000 title claims abstract description 49
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 239000013598 vector Substances 0.000 claims abstract description 60
- 238000012545 processing Methods 0.000 claims abstract description 57
- 238000000034 method Methods 0.000 claims abstract description 52
- 230000000877 morphologic effect Effects 0.000 claims description 125
- 238000005065 mining Methods 0.000 claims description 111
- 238000000605 extraction Methods 0.000 claims description 30
- 238000013507 mapping Methods 0.000 claims description 15
- 238000013135 deep learning Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 3
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 17
- 229910052737 gold Inorganic materials 0.000 description 17
- 239000010931 gold Substances 0.000 description 17
- 230000008569 process Effects 0.000 description 13
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 11
- 229910052802 copper Inorganic materials 0.000 description 11
- 239000010949 copper Substances 0.000 description 11
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 10
- 230000015572 biosynthetic process Effects 0.000 description 8
- 230000033558 biomineral tissue development Effects 0.000 description 7
- 238000010276 construction Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 229910052742 iron Inorganic materials 0.000 description 6
- 239000011435 rock Substances 0.000 description 5
- 230000037237 body shape Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 229910003460 diamond Inorganic materials 0.000 description 2
- 239000010432 diamond Substances 0.000 description 2
- 150000002505 iron Chemical class 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 238000012876 topography Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 229910052770 Uranium Inorganic materials 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013499 data model Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011946 reduction process Methods 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000002689 soil Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- JFALSRSLKYAFGM-UHFFFAOYSA-N uranium(0) Chemical compound [U] JFALSRSLKYAFGM-UHFFFAOYSA-N 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Remote Sensing (AREA)
- Computer Graphics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of data processing, and provides a three-dimensional space model data processing method and system for mineral area investigation. The method not only can improve the investigation precision, but also can improve the investigation efficiency, and saves time and resources. Based on the three-dimensional mineral geologic model data and the data clustering results at the descriptive level, a prediction space vector for constructing a mineral distribution point cloud can be determined, which helps predict areas where mineral may exist in the future. By constructing mineral distribution point clouds for target mineral areas, richer and more accurate information support can be provided, helping to make more scientific and reasonable decisions. The method is not limited to a certain type of mineral products or a certain investigation method, but can be flexibly applied to various mineral products and various investigation methods, and has strong universality and adaptability.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to a three-dimensional space model data processing method and system for mineral area investigation.
Background
In the field of mineral exploration, acquisition, processing and analysis of geological data has been a central task. These data not only relate to the location, quantification and development of mineral resources, but also directly affect the strategic planning and investment decisions of the relevant enterprises. However, conventional geological survey data processing methods often provide only relatively coarse and one-sided information due to technical limitations and limitations of theoretical frameworks.
In particular, conventional geological survey methods rely primarily on manual measurements, sample collection, and laboratory analysis. These processes are not only time consuming and labor intensive, but also the accuracy and reliability of the data is often questioned due to human error, equipment accuracy limitations, sample representativeness, and the like. In addition, the traditional method is hard to extract truly valuable geological features and mineral distribution rules from massive information without going through the mind when processing a large amount of data.
More importantly, the need for accurate investigation and effective decision-making places higher demands on the comprehensiveness, real-time and visualization of the data. For example, in complex mineral areas, factors such as geological structures, lithology changes, mineral occurrence states and the like can have a great influence on the exploitation and utilization of mineral resources. Conventional methods often have difficulty in fully describing and analyzing these factors, thereby limiting the accuracy of the survey and the effectiveness of decision support.
Therefore, it is important to develop a method that can comprehensively, accurately and intuitively describe and understand the geological conditions and mineral distribution characteristics of a target mineral area.
Disclosure of Invention
In order to solve the problems, the invention provides a three-dimensional space model data processing method and system for mineral area investigation.
In a first aspect of the embodiment of the present invention, a three-dimensional space model data processing method for mineral area exploration is provided, and the method is applied to a data processing system, and includes: acquiring a plurality of geological survey data about a target mineral area, and respectively carrying out three-dimensional modeling processing on the plurality of geological survey data based on at least two description layers to obtain three-dimensional mineral geological model data of the plurality of geological survey data in each description layer, wherein each three-dimensional mineral geological model data comprises local geological model data corresponding to at least two deposit state labels; for the three-dimensional mineral geological model data corresponding to each description layer, performing data clustering on geological investigation data included in at least one local geological model data in the three-dimensional mineral geological model data according to geological commonality statistical weights to obtain corresponding data clustering results; and determining a prediction space vector for constructing a mineral distribution point cloud set through three-dimensional mineral geological model data and data clustering results corresponding to the description layers, and constructing the mineral distribution point cloud set aiming at the target mineral area according to the prediction space vector.
Optionally, the three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description layers, so as to obtain three-dimensional mineral geological model data of the plurality of geological survey data on each description layer, including: when the minimum two description layers comprise deposit grade description layers, acquiring minimum two deposit grade keywords comprising the deposit grade description layers, wherein each deposit grade keyword corresponds to one deposit state label; and for each geological survey data, pairing the geological survey data with geological survey features corresponding to each mineral deposit grade keyword respectively, and determining a mineral deposit state label to which the geological survey data belongs based on pairing results.
Optionally, the three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description layers, so as to obtain three-dimensional mineral geological model data of the plurality of geological survey data on each description layer, including: when the at least two description layers comprise mineral deposit morphological description layers, spatial morphological feature mining is carried out on each geological survey data respectively to obtain spatial morphological features of each geological survey data; and carrying out three-dimensional modeling processing on the plurality of geological survey data based on the spatial morphological characteristics of the geological survey data to obtain three-dimensional mineral geological model data of the plurality of geological survey data on the mineral deposit morphological description level.
Optionally, the three-dimensional modeling processing is performed on the plurality of geological survey data based on the spatial morphological characteristics of each geological survey data to obtain three-dimensional mineral geological model data of the plurality of geological survey data on the mineral deposit morphological description layer, including: for each of the geological survey data, the following operations are performed: when the spatial morphological characteristics of the geological survey data are target spatial morphological characteristics, performing mineral deposit morphological modeling on the geological survey data by using a mineral deposit morphological modeling network corresponding to the target spatial morphological characteristics to obtain three-dimensional mineral geological model data of the geological survey data on the mineral deposit morphological description layer; and when the spatial morphological characteristics of the geological survey data are not target spatial morphological characteristics, performing feature mapping on the geological survey data so that the spatial morphological characteristics of the geological survey data with the completed feature mapping are target spatial morphological characteristics, and performing deposit morphological modeling on the geological survey data with the completed feature mapping by using the deposit morphological modeling network to obtain three-dimensional mineral geological model data of the geological survey data in the deposit morphological description layer.
Optionally, the performing spatial morphological feature mining on each geological survey data to obtain spatial morphological features of each geological survey data includes: for each of the geological survey data, the following operations are performed: pairing the geological survey data with geological survey data contained in a pre-configured data pool to obtain a pairing result; and when the pairing result represents that the geological survey data does not belong to the pre-configured data pool, spatial morphological feature mining is carried out on the geological survey data by utilizing a spatial morphological feature mining network, so that the spatial morphological features of the geological survey data are obtained.
Optionally, the spatial morphology feature mining network includes: a target spatial morphology feature mining network, a residual spatial morphology feature mining network, and a attention spatial morphology feature mining network; the step of performing spatial morphological feature mining on the geological survey data by using a spatial morphological feature mining network to obtain spatial morphological features of the geological survey data comprises the following steps: performing spatial morphological feature mining on the geological survey data by utilizing the target spatial morphological feature mining network to obtain a first spatial morphological feature mining result of the geological survey data; when the first spatial morphology feature mining result represents that the geological survey data is not the target spatial morphology feature, spatial morphology feature mining is carried out on the geological survey data by utilizing the residual spatial morphology feature mining network, and a second spatial morphology feature mining result of the geological survey data is obtained; when the first spatial morphology feature mining result characterizes the geological survey data as the target spatial morphology feature, judging that the first spatial morphology feature mining result meets expectations; when the second spatial morphology feature mining result represents that the geological survey data is not the residual spatial morphology feature, performing spatial morphology feature mining on the geological survey data by using the attention spatial morphology feature mining network to obtain a third spatial morphology feature mining result of the geological survey data; when the second spatial morphology feature mining result characterizes that the geological survey data is a residual spatial morphology feature, judging that the second spatial morphology feature mining result meets expectations; and determining the spatial morphological characteristics of the geological survey data according to the first spatial morphological characteristic mining result, the second spatial morphological characteristic mining result or the third spatial morphological characteristic mining result.
Optionally, the three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description layers, so as to obtain three-dimensional mineral geological model data of the plurality of geological survey data on each description layer, including: when the at least two description levels include mineral resource attribute description levels, the following operations are performed for each of the geological survey data: extracting a feature chain from the geological survey data to obtain a feature chain corresponding to the geological survey data, wherein three feature vectors in the feature chain comprise mineral scale, mineral resource attribute and mineral deposit morphology; and determining the feature chain corresponding to the geological survey data as three-dimensional mineral geological model data of the geological survey data on the mineral resource attribute description level.
Optionally, the number of the target mineral areas is at least two, and constructing a mineral distribution point cloud set for the target mineral areas according to the prediction space vector includes: acquiring a reference mineral distribution point cloud set, wherein the reference mineral distribution point cloud set comprises a point cloud generation strategy for each target mineral area; and processing the prediction space vector corresponding to each target mineral area by utilizing a corresponding point cloud generation strategy in the reference mineral distribution point cloud set to obtain the mineral distribution point cloud set aiming at the minimum two target mineral areas.
Optionally, for the three-dimensional mineral geological model data corresponding to each description layer, performing data clustering on geological survey data included in at least one local geological model data in the three-dimensional mineral geological model data according to a geological commonality statistical weight to obtain a corresponding data clustering result, where the data clustering result includes: for the three-dimensional mineral geological model data corresponding to each description layer, determining local geological model data with the number of geological survey data reaching a number threshold in the three-dimensional mineral geological model data as target local geological model data; obtaining geological commonality statistical weights between any two geological investigation data in the target local geological model data; and performing data clustering on geological investigation data in the target local geological model data according to the geological commonality statistical weight to obtain a data clustering result corresponding to the target local geological model data.
Optionally, the determining, according to the three-dimensional mineral geological model data and the data clustering result corresponding to each description layer, a prediction space vector for constructing a mineral distribution point cloud set includes: for each of the description layers, the following operations are performed: determining local geologic model data, which is subjected to grouping processing, in the three-dimensional mineral geologic model data as first local geologic model data, and determining local geologic model data, except the first local geologic model data, in the three-dimensional mineral geologic model data as second local geologic model data; performing knowledge refinement on geological investigation data in the first local geological model data by using a deep learning network to obtain knowledge refinement results corresponding to the first local geological model data; knowledge extraction is carried out on geological investigation data in the second local geological model data by utilizing the deep learning network, so that knowledge extraction results corresponding to the second local geological model data are obtained; and determining a prediction space vector for constructing a mineral distribution point cloud set based on the knowledge extraction result corresponding to each first local geological model data and the knowledge extraction result corresponding to each first local geological model data.
Optionally, the determining a prediction space vector for constructing a mineral distribution point cloud set based on the knowledge refinement result corresponding to each second local geological model data and the knowledge refinement result corresponding to each deposit state label includes: respectively determining influence weights of knowledge extraction results corresponding to the second local geological model data and knowledge extraction results corresponding to the deposit state labels to obtain influence weights of the knowledge extraction results; based on the influence weight of each knowledge extraction result, sampling from a plurality of knowledge extraction results to obtain a target knowledge extraction result with the influence weight reaching an influence weight threshold, wherein the target knowledge extraction result is used as a prediction space vector for constructing a mineral distribution point cloud set.
In a second aspect of an embodiment of the present invention, there is provided a data processing system, including: a processor, a memory and a bus connected to the processor; the processor and the memory complete communication with each other through the bus; the processor is used for calling the computer program in the memory to execute the three-dimensional space model data processing method facing the mineral area investigation.
In a third aspect of the embodiments of the present invention, a computer readable storage medium is provided, on which a program is stored, which when executed by a processor, implements the above-mentioned three-dimensional spatial model data processing method for mineral area investigation.
According to the three-dimensional space model data processing method and system for mineral area investigation, provided by the embodiment of the invention, geological conditions and mineral distribution characteristics of a target mineral area can be more intuitively and accurately understood through three-dimensional modeling processing on geological investigation data. The method not only can improve the investigation precision, but also can improve the investigation efficiency, and saves time and resources. Based on the three-dimensional mineral geologic model data and the data clustering results at the descriptive level, a prediction space vector for constructing a mineral distribution point cloud can be determined, which helps predict areas where mineral may exist in the future. By constructing mineral distribution point clouds aiming at target mineral areas, richer and more accurate information support can be provided for decision makers, and more scientific and reasonable decisions can be made. In addition, the technical scheme is not limited to a certain type of mineral products or a certain investigation method, but can be flexibly applied to various mineral products and various investigation methods, and has strong universality and adaptability. Through the technical scheme, knowledge and experience of geological investigation can be converted into specific data and models, and the data and models are convenient to store and analyze.
In addition, after the plurality of geological survey data of the target mineral area are obtained, three-dimensional mineral geological model data of the plurality of geological survey data on each description layer are obtained through three-dimensional modeling processing of the geological survey data, so that the geological survey data included in each description layer cannot be interfered by the geological survey data included in the rest description layers when the mineral distribution point cloud set is constructed, and therefore the pertinence and the focusing performance of the constructed mineral distribution point cloud set are stronger. And then, carrying out data clustering on the local geologic model data in each three-dimensional mineral geologic model data, so that when the mineral distribution point cloud set is constructed, the local geologic model data with fewer geologic survey data cannot be interfered by the local geologic model data with more geologic survey data, and the rationality and the interpretability of the constructed mineral distribution point cloud set are improved.
In conclusion, the technical scheme can comprehensively, accurately and intuitively describe and understand the geological conditions and mineral distribution characteristics of the target mineral area.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a three-dimensional space model data processing method for mineral area investigation according to an embodiment of the present invention.
FIG. 2 is a schematic diagram of a product module of a data processing system according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
In order to better understand the above technical solutions, the following detailed description of the technical solutions of the present invention is made by using the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and the embodiments of the present invention are detailed descriptions of the technical solutions of the present invention, and not limiting the technical solutions of the present invention, and the technical features of the embodiments and the embodiments of the present invention may be combined with each other without conflict.
Referring to fig. 1, a flowchart of a three-dimensional space model data processing method for mineral area investigation according to an embodiment of the present invention is provided, and the method is applied to a data processing system, and the specific content description included in the method includes S110-S130.
S110, acquiring a plurality of geological survey data about a target mineral area, and respectively carrying out three-dimensional modeling processing on the geological survey data based on at least two description layers to obtain three-dimensional mineral geological model data of the geological survey data on the description layers, wherein each three-dimensional mineral geological model data comprises local geological model data corresponding to at least two mineral deposit state labels.
S120, for the three-dimensional mineral geological model data corresponding to each description layer, performing data clustering on geological investigation data included in at least one piece of local geological model data in the three-dimensional mineral geological model data according to the geological commonality statistical weight to obtain a corresponding data clustering result.
S130, determining a prediction space vector for constructing a mineral distribution point cloud set through three-dimensional mineral geological model data and data clustering results corresponding to the description layers, and constructing the mineral distribution point cloud set aiming at the target mineral area according to the prediction space vector.
In the above solutions, the target mineral area refers to the geological area to be surveyed, in which it is desired to find or further understand the mineral resources. Geological survey data is information about geologic structures, deposit distribution, ore grade, etc., collected by various geological survey methods (e.g., geophysics, geochemistry, remote sensing, etc.). Description planes refer to different geologic structure layers, such as earth formation layers, geologic unit layers, deposit layers, and the like. The three-dimensional modeling process is to build a three-dimensional numerical model of mineral geology by using methods such as computer technology, geostatistics and the like according to the collected geological survey data. The deposit status label indicates an identification of the deposit status, such as deposit type, ore grade, mineralization level, etc. Local geologic model data: the geologic model data corresponding to each deposit state label describes the geologic model under a particular geologic state. The geological commonality statistical weight is a weight for weighting each local geological model data according to the statistical commonality of geological features. Data clustering is the grouping of geologic survey data having similar geologic features together to form one or more groups of "clusters". The prediction space vector is a space vector which is determined according to the three-dimensional geological model and the data clustering result and is used for constructing mineral distribution prediction. The mineral distribution point cloud set is a three-dimensional point cloud set which is constructed by using a prediction space vector and represents the possible distribution of mineral.
In one exemplary application scenario, the goal is to survey a gold mine located in a region. To accomplish this task, first, geological survey data, including geological, geophysical and geochemical data, is acquired regarding the gold region. Then, two description levels are selected: a formation structure level and a deposit distribution level. And carrying out three-dimensional modeling processing on the exploration data of each layer to obtain three-dimensional mineral geological model data of each layer. In each three-dimensional model, at least two deposit status labels, such as "high gold content" and "low gold content", are also labeled, and corresponding local geologic model data is generated.
And then, data clustering is carried out on the three-dimensional geologic model data of each description layer according to the geologic commonality statistical weight. For example, it may be found that at the formation structure level, rock types in some places have a strong correlation with gold distribution, so that data in these places can be clustered. Also, at the deposit distribution level, it may be found that gold content is particularly high in some places, and data in these places may be clustered. Thus, data clustering results at each level were obtained.
Finally, a prediction space vector for constructing mineral distribution point clouds is determined through three-dimensional geological model data and data clustering results of each description layer. For example, the center position, shape, and size of each data blob may be part of the prediction space vector. Then, based on these prediction space vectors, a three-dimensional point cloud set representing the possible distribution of gold ore is constructed.
In this example, a three-dimensional geologic model was successfully created by processing and analyzing multiple types of geologic survey data, and a mineral distribution point cloud was generated. The method not only can help understand the geological conditions and distribution characteristics of the gold ore more accurately, but also can provide important basis for further investigation and exploitation.
In other examples, predictive spatial vectors are obtained by processing and analyzing geologic model data and data clustering results, which are commonly used to characterize the likely distribution characteristics of mineral products. In order to more intuitively interpret the prediction space vector, it can be thought of as a multi-dimensional vector, where each dimension represents a geologic property.
For example, in the case of a gold mine investigation, it is assumed that the geological properties of interest are rock type, crust structure, mineralization degree, gold content, and the like. Then, the prediction space vector V can be expressed as: v= [ V1, V2, V3, V4];
Wherein:
v1 represents a rock type, possibly represented by a number or code of the rock type;
v2 represents the crust construction, possibly represented by the direction and magnitude of crust movement;
v3 represents the degree of mineralization, possibly represented by a numerical value that measures the degree of mineralization;
v4 represents the gold content, possibly represented by an average value of the gold content or other statistical indicators.
It should be noted that the specific form of the predicted spatial vector will vary depending on the actual geological conditions and the survey target. Meanwhile, for a high-dimensional prediction space vector, a dimension reduction process such as Principal Component Analysis (PCA) or the like may be required in order to better understand and utilize such information.
In still other examples, taking copper ore as an example, a detailed explanation of how three-dimensional mineral geologic model data is acquired and processed is provided.
First, geological survey data concerning a target copper mine area needs to be acquired. This may include various types of data such as topography, formation, lithology, architecture, geophysics (e.g., gravity, magnetic field), geochemistry (e.g., soil, copper content in water samples), etc. Such data is typically obtained by ground or aeronautical geological survey methods, and has spatially accurate coordinates.
Next, two description levels are selected for modeling, for example: a ground formation level and a deposit level. At the layer of the earth structure, a three-dimensional model showing information such as earth movement, structural band and the like is generated by using data such as topography, stratum, lithology, structure and the like through a geostatistical method and a computer graphics method. In the mineral deposit layer, a three-dimensional model showing the mineral form, mineralization degree, copper content distribution and other information is generated by using geophysical and geochemical data and also by using a geostatistical and computer graphics method.
In each three-dimensional model, key geological features are also marked, such as where copper ores may be present, which are typically assigned a deposit status label, such as "possibly copper-containing mineralized zone", "high copper-containing mine zone", etc. Each tag corresponds to a local geological model data, including geological information and copper content of the place.
By the steps, three-dimensional mineral geological model data about the target copper mine area are obtained. The data not only can help intuitively understand the geological condition, but also can provide important basis for subsequent prediction and decision.
By applying the steps S110-S130, geological conditions and mineral distribution characteristics of the target mineral area can be more intuitively and accurately understood through three-dimensional modeling processing on geological survey data. The method not only can improve the investigation precision, but also can improve the investigation efficiency, and saves time and resources. Based on the three-dimensional mineral geologic model data and the data clustering results at the descriptive level, a prediction space vector for constructing a mineral distribution point cloud can be determined, which helps predict areas where mineral may exist in the future. By constructing mineral distribution point clouds aiming at target mineral areas, richer and more accurate information support can be provided for decision makers, and more scientific and reasonable decisions can be made. In addition, the technical scheme is not limited to a certain type of mineral products or a certain investigation method, but can be flexibly applied to various mineral products and various investigation methods, and has strong universality and adaptability. Through the technical scheme, knowledge and experience of geological investigation can be converted into specific data and models, and the data and models are convenient to store and analyze.
The following provides a pseudo code example to describe the above technical solution, and when in actual application, the specific programming language may be used and the corresponding library or framework is called to implement the flow. Meanwhile, the following may be appropriately adjusted and optimized according to specific needs and conditions.
# Definition geology _data is acquired geological survey data, layers are layers to be described, models are three-dimensional modeling processing functions, weight_ calculator is a function for calculating geological commonality statistical weights, data_ clustering is a function for data clustering, vector_generator is a function for determining prediction space vectors, and point_closed_ constructor is a function for constructing mineral distribution point clouds.
geology_data=get_geology_data(target_mining_area)
three_dim_models={}
for layer in layers:
three_dim_model=model(geology_data，layer)
three_dim_models[layer]=three_dim_model
clustered_results={}
for layer，model_data in three_dim_models.items():
weights=weight_calculator(model_data)
clustered_result=data_clustering(model_data，weights)
clustered_results[layer]=clustered_result
prediction_vectors={}
for layer，clustered_result in clustered_results.items():
prediction_vector=vector_generator(clustered_result)
prediction_vectors[layer]=prediction_vector
point_clouds={}
for area，prediction_vector in prediction_vectors.items():
point_cloud=point_cloud_constructor(prediction_vector)
point_clouds[area]=point_cloud
return point_clouds.
In some possible embodiments, the three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description levels, to obtain three-dimensional mineral geological model data of the plurality of geological survey data at each description level, including: when the minimum two description layers comprise deposit grade description layers, acquiring minimum two deposit grade keywords comprising the deposit grade description layers, wherein each deposit grade keyword corresponds to one deposit state label; and for each geological survey data, pairing the geological survey data with geological survey features corresponding to each mineral deposit grade keyword respectively, and determining a mineral deposit state label to which the geological survey data belongs based on pairing results.
In this solution, a deposit grade description level will be introduced and geological survey data will be tagged with deposit grade keywords. Taking a survey of iron ore as an example: geological survey data, including lithology, crust construction, geophysical and geochemical information, for example, has been acquired for this iron ore region. First, at least two description levels are selected for modeling, one of which is a deposit grade description level. At this level, two deposit grade keywords are defined: each keyword corresponds to a deposit state label.
Next, processing of the geological survey data begins. For each piece of data, it is paired with a geologic survey feature corresponding to each deposit grade keyword. For example, if a piece of data shows that the iron content of the site exceeds 30%, it will be paired to a high-grade keyword, and vice versa. In this way, the deposit status label to which each piece of geological survey data belongs can be determined.
Based on the above steps, three-dimensional mineral geological model data at each description level (including the deposit grade description level) can be obtained.
By the design, the ore deposit grade description level and the corresponding keywords are introduced, so that geological survey data can be labeled and classified more accurately, and the accuracy of three-dimensional modeling is improved. By matching the geological survey data with the ore deposit grade keywords, the ore deposit grade conditions of all places can be intuitively displayed, and the interpretation capability of geological models is enhanced. Based on a more accurate and visual three-dimensional geological model, more powerful support can be provided for mineral exploration and exploitation decisions.
In some possible embodiments, the three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description levels, to obtain three-dimensional mineral geological model data of the plurality of geological survey data at each description level, including: when the at least two description layers comprise mineral deposit morphological description layers, spatial morphological feature mining is carried out on each geological survey data respectively to obtain spatial morphological features of each geological survey data; and carrying out three-dimensional modeling processing on the plurality of geological survey data based on the spatial morphological characteristics of the geological survey data to obtain three-dimensional mineral geological model data of the plurality of geological survey data on the mineral deposit morphological description level.
Suppose that a certain diamond mine in south africa is being surveyed. First, several geological survey data are acquired about the diamond mine region, which may include information about formation, lithology, architecture, geophysics, and geochemistry.
In processing these data, two description planes are selected for modeling, one of which is the mineral deposit morphology description plane. The main concern at this level is the morphological characteristics of the deposit, such as size, shape, orientation and continuity.
Then, spatial morphological feature mining is started for each geological survey data. For example, by analyzing information such as crust movement and structural deformation, it can be presumed that rock at a certain location may have cracks or faults; by analyzing geophysical and geochemical data, it can be inferred that mineralization zones may exist at a site. This information is extracted as spatial morphological features.
And finally, carrying out three-dimensional modeling processing on geological survey data based on the spatial morphological characteristics to obtain three-dimensional mineral geological model data on the mineral deposit morphological description level. The model not only displays the geological conditions of various places, but also displays morphological characteristics of mineral deposits, such as cracks, faults, mineralization zones and the like which may exist.
Thus, through the space morphological feature mining, more useful information can be extracted from geological survey data, and the precision of three-dimensional modeling is improved. At the mineral deposit morphological description level, morphological characteristics of mineral deposits can be intuitively seen, which has important value for understanding mineral distribution and predicting mineral occurrence.
In some possible embodiments, the three-dimensional modeling processing is performed on the plurality of geological survey data based on the spatial morphological characteristics of each geological survey data to obtain three-dimensional mineral geological model data of the plurality of geological survey data at the mineral deposit morphological description level, including: for each of the geological survey data, the following operations are performed: when the spatial morphological characteristics of the geological survey data are target spatial morphological characteristics, performing mineral deposit morphological modeling on the geological survey data by using a mineral deposit morphological modeling network corresponding to the target spatial morphological characteristics to obtain three-dimensional mineral geological model data of the geological survey data on the mineral deposit morphological description layer; and when the spatial morphological characteristics of the geological survey data are not target spatial morphological characteristics, performing feature mapping on the geological survey data so that the spatial morphological characteristics of the geological survey data with the completed feature mapping are target spatial morphological characteristics, and performing deposit morphological modeling on the geological survey data with the completed feature mapping by using the deposit morphological modeling network to obtain three-dimensional mineral geological model data of the geological survey data in the deposit morphological description layer.
In the implementation process, the technical scheme utilizes the methods of model, feature mapping and the like in machine learning to model the mineral deposit morphology. The method can be described in detail by taking a gold mine investigation as an example: first, several geological survey data about this gold region are acquired and spatial morphological features are extracted therefrom.
Then, it is assumed that a deposit morphology modeling network has been trained that is capable of processing target spatial morphology features (e.g., ore body shape, size, and distribution, etc.). For each piece of geological survey data, it is checked whether its spatial morphology is the target spatial morphology.
If so, the network is directly used for modeling the mineral deposit morphology, and three-dimensional mineral geological model data of the mineral deposit morphology at the mineral deposit morphology description level is obtained.
If not, it is necessary to first perform feature mapping. Feature mapping may include operations such as dimension reduction, normalization, encoding, etc., in order to transform the original spatial morphology features into target spatial morphology features. After the feature mapping is completed, the mineral deposit morphology modeling network is used for carrying out mineral deposit morphology modeling on the mineral deposit, and three-dimensional mineral geological model data of the mineral deposit morphology modeling network in the mineral deposit morphology description layer are obtained.
Therefore, geological survey data which do not accord with the target space morphological characteristics can be converted into data which accord with the target space morphological characteristics through feature mapping, so that a deposit morphological modeling network can process more kinds of data, and the adaptability of the model is improved. Deposit morphology modeling networks are typically based on deep learning techniques that automatically learn and extract features from large amounts of data, and thus may have higher accuracy when processing complex geological survey data than traditional geostatistical methods. The whole processing flow comprises the steps of feature inspection, feature mapping, deposit morphology modeling and the like, and can realize automatic processing through programming, so that a great deal of manpower and time are saved. The geological conditions and mineral deposit morphological characteristics of each place can be intuitively seen through the three-dimensional mineral geological model, and the three-dimensional mineral geological model has important value for understanding mineral distribution and predicting mineral occurrence.
In some possible embodiments, the performing spatial morphological feature mining on each geological survey data to obtain spatial morphological features of each geological survey data includes: for each of the geological survey data, the following operations are performed: pairing the geological survey data with geological survey data contained in a pre-configured data pool to obtain a pairing result; and when the pairing result represents that the geological survey data does not belong to the pre-configured data pool, spatial morphological feature mining is carried out on the geological survey data by utilizing a spatial morphological feature mining network, so that the spatial morphological features of the geological survey data are obtained.
For example, a certain iron ore is being explored, which is located in brazil. First, several geological survey data are acquired for this iron ore region.
When the space morphological feature is mined, a pre-configured data pool is needed, and the data pool contains geological investigation data which is subjected to the space morphological feature mining.
Each newly acquired geological survey is then paired with data in the pre-configured data pool. The pairing may be by comparing their geographical location, lithology, formation and crust construction information. If the pairing result shows that the data already exists in the pre-configured data pool, feature mining is not performed any more, and the feature result in the pre-configured data pool is directly used.
If the pairing result shows that the data does not belong to the pre-configured data pool, then the spatial morphology feature mining network is required to be utilized for feature mining. This network may be based on deep learning techniques, which enable automatic extraction of spatial morphological features from the data, such as ore body shape, size and distribution.
Therefore, repeated feature mining on processed data can be avoided through pairing check, and a large amount of computing resources are saved. The characteristics which are already mined can be directly obtained by utilizing the pre-configured data pool, so that the data processing speed is improved. Because the data in the pre-configured data pool is subjected to unified feature mining, the consistency of results obtained by multiple surveys at the same place can be ensured. When the new geological survey data does not belong to the pre-configured data pool, the key features can be effectively extracted by adopting the spatial morphological feature mining network, and the accuracy and the robustness of the model are further improved.
In some possible embodiments, the spatial morphology feature mining network comprises: a target spatial morphology feature mining network, a residual spatial morphology feature mining network, and a attention spatial morphology feature mining network; the step of performing spatial morphological feature mining on the geological survey data by using a spatial morphological feature mining network to obtain spatial morphological features of the geological survey data comprises the following steps: performing spatial morphological feature mining on the geological survey data by utilizing the target spatial morphological feature mining network to obtain a first spatial morphological feature mining result of the geological survey data; when the first spatial morphology feature mining result represents that the geological survey data is not the target spatial morphology feature, spatial morphology feature mining is carried out on the geological survey data by utilizing the residual spatial morphology feature mining network, and a second spatial morphology feature mining result of the geological survey data is obtained; when the second spatial morphology feature mining result represents that the geological survey data is not the residual spatial morphology feature, performing spatial morphology feature mining on the geological survey data by using the attention spatial morphology feature mining network to obtain a third spatial morphology feature mining result of the geological survey data; and determining the spatial morphological characteristics of the geological survey data according to the first spatial morphological characteristic mining result, the second spatial morphological characteristic mining result or the third spatial morphological characteristic mining result.
In the above technical solution, three different spatial morphological feature mining networks are adopted: a target spatial morphology feature mining network, a residual spatial morphology feature mining network, and an attention spatial morphology feature mining network. Each network has its specific functions and application scenarios. Taking a copper mine investigation as an example for illustration: first, for newly acquired geological survey data, a target spatial morphology feature mining network is used for processing. This network may be specifically designed to extract target features such as ore body shape, size and distribution.
If the first spatial morphology mining result obtained by the network shows that the data does not exhibit the desired target spatial morphology, the residual spatial morphology mining network is further used for processing. This network may be designed to mine other important features that differ significantly from the target feature (i.e., residuals).
If the second spatial morphology feature mining result obtained by the residual spatial morphology feature mining network still does not meet the requirement, the attention spatial morphology feature mining network is used again for processing. This network may be designed to focus on certain specific parts or features in the data to extract more useful information.
In addition, when the first spatial morphology feature mining result characterizes the geological survey data as the target spatial morphology feature, the first spatial morphology feature mining result is determined to satisfy a desire. And when the second spatial morphology feature mining result represents that the geological survey data is a residual spatial morphology feature, judging that the second spatial morphology feature mining result meets expectations.
Finally, the spatial morphological characteristics of the geological survey data are determined according to the results obtained by the three networks.
Therefore, by using different networks to process different types of data, effective feature mining can be ensured for various conditions, and the adaptability of the model is improved. Each network is specifically designed to handle a specific class of problems, so they may have higher accuracy than general feature mining methods in the respective domain. When a certain network does not get satisfactory results, other networks may also be tried. This way the robustness of the system is improved, enabling it to better cope with various complex and varying situations. The attention space morphological feature mining network can focus on important parts in data, and feature resolution is improved, so that finer space morphological features can be obtained.
In some possible embodiments, the three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description levels, to obtain three-dimensional mineral geological model data of the plurality of geological survey data at each description level, including: when the at least two description levels include mineral resource attribute description levels, the following operations are performed for each of the geological survey data: extracting a feature chain from the geological survey data to obtain a feature chain corresponding to the geological survey data, wherein three feature vectors in the feature chain comprise mineral scale, mineral resource attribute and mineral deposit morphology; and determining the feature chain corresponding to the geological survey data as three-dimensional mineral geological model data of the geological survey data on the mineral resource attribute description level.
In this solution, mineral resource attribute description layers are added to the three-dimensional modeling process. Suppose that a uranium ore located in australia is being investigated.
For each piece of geological survey data, feature chain extraction is performed. This feature chain includes three main feature vectors: mineral scale, mineral resource attributes, and deposit morphology.
Mineral scale may include information such as length, width, depth, and reserves of the deposit;
Mineral resource attributes may include information such as the grade, type, physical and chemical properties of the ore;
the deposit morphology may then include information about the shape, size, orientation, and continuity of the ore body.
By means of these three eigenvectors, a comprehensive and specific description of mineral resources can be obtained. This feature chain is determined as three-dimensional mineral geologic model data of geologic survey data at the mineral resource attribute description level.
Thus, by modeling at the mineral resource attribute description level, more detailed information about mineral resources, such as mineral deposit size, ore grade, and physicochemical attributes, can be obtained. Such comprehensive information facilitates more accurate decisions by mining companies and construction segments, such as assessing mineral value, planning mining activities, and making environmental policies. Accurate mineral resource attributes can help optimize mining plans, improve resource utilization and reduce waste. Taking into account factors such as deposit morphology, mineral scale, mineral resource attributes, etc., a more accurate prediction of future mineral development potential may be provided.
In some possible embodiments, the number of the target mineral regions is at least two, and the constructing a mineral distribution point cloud set for the target mineral regions according to the prediction space vector includes: acquiring a reference mineral distribution point cloud set, wherein the reference mineral distribution point cloud set comprises a point cloud generation strategy for each target mineral area; and processing the prediction space vector corresponding to each target mineral area by utilizing a corresponding point cloud generation strategy in the reference mineral distribution point cloud set to obtain the mineral distribution point cloud set aiming at the minimum two target mineral areas.
This solution mainly focuses on how to construct mineral distribution point clouds for at least two target mineral areas using predictive spatial vectors. Suppose that two gold ores located in congo gold are being explored.
First, a reference mineral distribution point cloud is acquired. This point cloud includes point cloud generation strategies for each target mineral area, such as determining the location, color, size, etc. of each point based on geological data and mineral resource attributes.
Then, according to the strategy in the reference point cloud set, the prediction space vector corresponding to each target mineral area is processed. The prediction space vector may be a prediction result obtained by a deep learning model based on geological survey data, and may reflect a possible distribution situation of mineral products.
In this way, a mineral distribution point cloud for at least two target mineral areas can be obtained. This point cloud provides an intuitive, three-dimensional way to understand and demonstrate the distribution of mineral products.
By the design, the mineral distribution point cloud set is constructed, so that the distribution condition of the mineral can be intuitively seen in a three-dimensional space, and the visual effect of data is improved. The point cloud set can clearly reflect the concrete condition of mineral distribution, and help mining companies and construction parts to make faster and accurate decisions. The technical scheme allows a plurality of target mineral areas to be processed simultaneously, and further improves the processing efficiency and range. By comparing point clouds of different mineral areas, the mineral resource conditions of each area can be better understood and evaluated, and the optimal allocation of resources is facilitated.
In some possible embodiments, for the three-dimensional mineral geological model data corresponding to each of the description layers, performing data clustering on geological survey data included in at least one local geological model data in the three-dimensional mineral geological model data according to a geological commonality statistical weight to obtain corresponding data clustering results, where the data clustering results include: for the three-dimensional mineral geological model data corresponding to each description layer, determining local geological model data with the number of geological survey data reaching a number threshold in the three-dimensional mineral geological model data as target local geological model data; obtaining geological commonality statistical weights between any two geological investigation data in the target local geological model data; and performing data clustering on geological investigation data in the target local geological model data according to the geological commonality statistical weight to obtain a data clustering result corresponding to the target local geological model data.
The main aim of the technical scheme is to perform data clustering on at least one local geologic model data in the three-dimensional mineral geologic model data according to the geologic commonality statistical weight. Suppose that three-dimensional mineral geologic model data is being processed for a piece of iron ore area in brazil.
Firstly, local geological model data, of which the number of geological survey data contained in the model data reaches a certain threshold, are found out, and the local geological model data are determined to be target local geological model data. For example, it is possible to set the threshold to 100, i.e. a local model is considered as a target local model only if it contains at least 100 geological survey data.
Then, a statistical weight of the geological commonalities between any two geological survey data in the target local model is obtained. This weight may be calculated based on information such as geographic location, lithology, formation and crust construction.
Finally, clustering the data in the target local model according to the weights. For example, it is possible to categorize data with a high degree of geological commonality into one cluster and data with a lower degree of commonality into another cluster. Thus, the data clustering result corresponding to each target local model is obtained.
In this way, by grouping data, data having a high degree of commonality can be grouped into one group, making management and processing of data more efficient. Data clustering based on geological commonality statistical weight can ensure that data in the same group have higher internal similarity, which is beneficial to subsequent data analysis and model training. By clustering the data, the relation between different geological survey data can be better understood, and the interpretation of the result is enhanced. When processing data groups with high commonality, prediction errors can be reduced, thereby improving the prediction accuracy of the model.
In some possible embodiments, the determining the prediction space vector for constructing the mineral distribution point cloud set through the three-dimensional mineral geological model data and the data clustering result corresponding to each description layer includes: for each of the description layers, the following operations are performed: determining local geologic model data, which is subjected to grouping processing, in the three-dimensional mineral geologic model data as first local geologic model data, and determining local geologic model data, except the first local geologic model data, in the three-dimensional mineral geologic model data as second local geologic model data; performing knowledge refinement on geological investigation data in the first local geological model data by using a deep learning network to obtain knowledge refinement results corresponding to the first local geological model data; knowledge extraction is carried out on geological investigation data in the second local geological model data by utilizing the deep learning network, so that knowledge extraction results corresponding to the second local geological model data are obtained; and determining a prediction space vector for constructing a mineral distribution point cloud set based on the knowledge extraction result corresponding to each first local geological model data and the knowledge extraction result corresponding to each first local geological model data.
The technical scheme mainly comprises the step of determining prediction space vectors for constructing mineral distribution point clouds by using a deep learning network to conduct knowledge refinement on three-dimensional mineral geological model data and data clustering results. The description will be given taking copper ore as an example.
First, local geologic model data on which the clustering process is performed is defined as first local geologic model data, and the remaining local geologic model data is defined as second local geologic model data.
And then, knowledge refinement is carried out on geological investigation data in the two types of local geological model data by utilizing a deep learning network. This process may include identifying and extracting key features, learning correlations between data, classifying or regressing, and the like.
Finally, a predictive spatial vector for constructing a mineral distribution point cloud set is determined based on knowledge refinement results corresponding to the first and second local geologic model data. This predictive spatial vector reflects the results of understanding and predicting geological data through a deep learning network, which can be used to create a comprehensive and accurate mineral distribution point cloud.
In this way, the deep learning network can extract deep features of the geological data and utilize the features to predict, which helps to improve the prediction accuracy of the model. By knowledge refinement of different types of local geologic model data, accurate prediction results can be ensured when the model faces various conditions, so that the robustness of the model is enhanced. The predictive spatial vector reflects the possible mineral distribution, which can help optimize resource allocation and mining strategies. The prediction result based on the deep learning network can provide more comprehensive and accurate information, and is helpful for accelerating the decision making process.
In some possible embodiments, the determining a prediction space vector for constructing a mineral distribution point cloud set based on the knowledge refinement result corresponding to each of the second local geologic model data and the knowledge refinement result corresponding to each of the deposit state labels includes: respectively determining influence weights of knowledge extraction results corresponding to the second local geological model data and knowledge extraction results corresponding to the deposit state labels to obtain influence weights of the knowledge extraction results; based on the influence weight of each knowledge extraction result, sampling from a plurality of knowledge extraction results to obtain a target knowledge extraction result with the influence weight reaching an influence weight threshold, wherein the target knowledge extraction result is used as a prediction space vector for constructing a mineral distribution point cloud set.
In this solution, the prediction space vector for constructing the mineral distribution point cloud set is determined based on the knowledge-refined result corresponding to the second local geologic model data and the knowledge-refined result corresponding to the deposit status label. Taking a gold investigation of a drilling area somewhere as an example.
First, it is necessary to determine the impact weight of each knowledge refinement result. This process may involve a number of factors such as the quality of the data, the quantity, the update frequency, etc. For example, if a particular piece of geological survey data covers a large number of deposits and the update frequency is high, the impact of that piece of data may be significant.
An impact weight threshold is then set and only knowledge refinements that reach this threshold are selected into the prediction space vector. For example, a threshold of 0.5 may be set, i.e., only knowledge refinements with impact weights exceeding 0.5 may be selected.
Finally, according to the extracted results of the selected target knowledge, a prediction space vector is constructed and used for generating mineral distribution point clouds.
Therefore, through weight distribution and screening, knowledge extraction results with larger influence can be prioritized, so that prediction accuracy is improved. By setting the weight threshold, the scheme avoids excessive calculation and processing of the data with smaller influence and optimizes the use of resources. The determination of the impact weight is not fixed but can be adjusted according to the actual situation. This allows the model to maintain good performance in different environments and requirements. By converting the knowledge extraction result with the most influence into the prediction space vector, key information can be acquired more quickly, so that the decision efficiency is improved.
Referring to FIG. 2 in combination, the embodiment of the invention further provides a data processing system 100, which includes a processor 111, and a memory 112 and a bus 113 connected to the processor 111. Wherein the processor 111 and the memory 112 perform communication with each other via a bus 113. The processor 111 is adapted to call program instructions in the memory 112 to perform the three-dimensional spatial model data processing method for mineral area investigation described above.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or cloud server that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or cloud server. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude that an additional identical element is present in a process, method, article, or cloud server comprising the element.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The foregoing is merely exemplary of the present invention and is not intended to limit the present invention. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the invention are to be included in the scope of the claims of the present invention.
Claims (5)
1. A three-dimensional spatial model data processing method for mineral area exploration, characterized by being applied to a data processing system, the method comprising:
acquiring a plurality of geological survey data about a target mineral area, and respectively carrying out three-dimensional modeling processing on the plurality of geological survey data based on at least two description layers to obtain three-dimensional mineral geological model data of the plurality of geological survey data in each description layer, wherein each three-dimensional mineral geological model data comprises local geological model data corresponding to at least two deposit state labels;
For the three-dimensional mineral geological model data corresponding to each description layer, performing data clustering on geological investigation data included in at least one local geological model data in the three-dimensional mineral geological model data according to geological commonality statistical weights to obtain corresponding data clustering results;
Determining a prediction space vector for constructing a mineral distribution point cloud set through three-dimensional mineral geological model data and data clustering results corresponding to each description layer, and constructing the mineral distribution point cloud set aiming at the target mineral area according to the prediction space vector;
Wherein, the description level refers to different geological structure levels, including a ground structure level, a geological unit level and a mineral deposit level; the local geologic model data is geologic model data corresponding to each deposit state label; the geological commonality statistical weight is a weight for carrying out weighting treatment on each local geological model data according to the statistical commonality of geological features; data clustering is to aggregate geological survey data with similar geological features together to form one or more clusters;
The three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description layers to obtain three-dimensional mineral geological model data of the plurality of geological survey data on each description layer, and the three-dimensional mineral geological model data comprises:
When the minimum two description layers comprise deposit grade description layers, acquiring minimum two deposit grade keywords comprising the deposit grade description layers, wherein each deposit grade keyword corresponds to one deposit state label;
For each geological survey data, pairing the geological survey data with geological survey features corresponding to each mineral deposit grade keyword respectively, and determining mineral deposit state labels to which the geological survey data belong based on pairing results;
The three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description layers to obtain three-dimensional mineral geological model data of the plurality of geological survey data on each description layer, and the three-dimensional mineral geological model data comprises:
When the at least two description layers comprise mineral deposit morphological description layers, spatial morphological feature mining is carried out on each geological survey data respectively to obtain spatial morphological features of each geological survey data;
Based on the spatial morphological characteristics of each geological survey data, carrying out three-dimensional modeling processing on the plurality of geological survey data to obtain three-dimensional mineral geological model data of the plurality of geological survey data on the mineral deposit morphological description layer;
The three-dimensional modeling processing is performed on the plurality of geological survey data based on at least two description layers to obtain three-dimensional mineral geological model data of the plurality of geological survey data on each description layer, and the three-dimensional mineral geological model data comprises:
when the at least two description levels include mineral resource attribute description levels, the following operations are performed for each of the geological survey data:
Extracting a feature chain from the geological survey data to obtain a feature chain corresponding to the geological survey data, wherein three feature vectors in the feature chain comprise mineral scale, mineral resource attribute and mineral deposit morphology;
Determining a feature chain corresponding to the geological survey data as three-dimensional mineral geological model data of the geological survey data on the mineral resource attribute description level;
The method for performing data clustering on geological survey data included in at least one local geological model data in the three-dimensional mineral geological model data according to geological commonality statistical weights to obtain corresponding data clustering results, and the method comprises the following steps:
For the three-dimensional mineral geological model data corresponding to each description layer, determining local geological model data with the number of geological survey data reaching a number threshold in the three-dimensional mineral geological model data as target local geological model data;
obtaining geological commonality statistical weights between any two geological investigation data in the target local geological model data;
Performing data clustering on geological investigation data in the target local geological model data according to the geological commonality statistical weight to obtain a data clustering result corresponding to the target local geological model data;
The determining a prediction space vector for constructing a mineral distribution point cloud set through the three-dimensional mineral geological model data and the data clustering result corresponding to each description layer comprises the following steps:
for each of the description layers, the following operations are performed:
Determining local geologic model data, which is subjected to grouping processing, in the three-dimensional mineral geologic model data as first local geologic model data, and determining local geologic model data, except the first local geologic model data, in the three-dimensional mineral geologic model data as second local geologic model data;
Performing knowledge refinement on geological investigation data in the first local geological model data by using a deep learning network to obtain knowledge refinement results corresponding to the first local geological model data;
Knowledge extraction is carried out on geological investigation data in the second local geological model data by utilizing the deep learning network, so that knowledge extraction results corresponding to the second local geological model data are obtained;
And determining a prediction space vector for constructing a mineral distribution point cloud set based on the knowledge extraction result corresponding to each first local geological model data and the knowledge extraction result corresponding to each second local geological model data.
2. The method of claim 1, wherein the performing three-dimensional modeling processing on the plurality of geological survey data based on the spatial morphological characteristics of each geological survey data to obtain three-dimensional mineral geological model data of the plurality of geological survey data at the mineral deposit morphological descriptive level comprises:
For each of the geological survey data, the following operations are performed:
when the spatial morphological characteristics of the geological survey data are target spatial morphological characteristics, performing mineral deposit morphological modeling on the geological survey data by using a mineral deposit morphological modeling network corresponding to the target spatial morphological characteristics to obtain three-dimensional mineral geological model data of the geological survey data on the mineral deposit morphological description layer;
And when the spatial morphological characteristics of the geological survey data are not target spatial morphological characteristics, performing feature mapping on the geological survey data so that the spatial morphological characteristics of the geological survey data with the completed feature mapping are target spatial morphological characteristics, and performing deposit morphological modeling on the geological survey data with the completed feature mapping by using the deposit morphological modeling network to obtain three-dimensional mineral geological model data of the geological survey data in the deposit morphological description layer.
3. The method of claim 1, wherein the performing spatial morphology feature mining on each of the geological survey data to obtain spatial morphology features of each of the geological survey data comprises:
For each of the geological survey data, the following operations are performed:
Pairing the geological survey data with geological survey data contained in a pre-configured data pool to obtain a pairing result;
when the pairing result represents that the geological survey data does not belong to the pre-configured data pool, spatial morphological feature mining is carried out on the geological survey data by utilizing a spatial morphological feature mining network, so that spatial morphological features of the geological survey data are obtained;
Wherein, the spatial morphology feature mining network comprises: a target spatial morphology feature mining network, a residual spatial morphology feature mining network, and a attention spatial morphology feature mining network; the step of performing spatial morphological feature mining on the geological survey data by using a spatial morphological feature mining network to obtain spatial morphological features of the geological survey data comprises the following steps:
performing spatial morphological feature mining on the geological survey data by utilizing the target spatial morphological feature mining network to obtain a first spatial morphological feature mining result of the geological survey data;
When the first spatial morphology feature mining result represents that the geological survey data is not the target spatial morphology feature, spatial morphology feature mining is carried out on the geological survey data by utilizing the residual spatial morphology feature mining network, and a second spatial morphology feature mining result of the geological survey data is obtained; when the first spatial morphology feature mining result characterizes the geological survey data as the target spatial morphology feature, judging that the first spatial morphology feature mining result meets expectations;
When the second spatial morphology feature mining result represents that the geological survey data is not the residual spatial morphology feature, performing spatial morphology feature mining on the geological survey data by using the attention spatial morphology feature mining network to obtain a third spatial morphology feature mining result of the geological survey data; when the second spatial morphology feature mining result characterizes that the geological survey data is a residual spatial morphology feature, judging that the second spatial morphology feature mining result meets expectations;
And determining the spatial morphological characteristics of the geological survey data according to the first spatial morphological characteristic mining result, the second spatial morphological characteristic mining result or the third spatial morphological characteristic mining result.
4. The method of claim 1, wherein the number of target mineral regions is a minimum of two, and the constructing a mineral distribution point cloud for the target mineral regions based on the predictive spatial vector comprises:
acquiring a reference mineral distribution point cloud set, wherein the reference mineral distribution point cloud set comprises a point cloud generation strategy for each target mineral area;
And processing the prediction space vector corresponding to each target mineral area by utilizing a corresponding point cloud generation strategy in the reference mineral distribution point cloud set to obtain the mineral distribution point cloud set aiming at the minimum two target mineral areas.
5. A data processing system comprising a processor, and a memory and bus coupled to the processor; the processor and the memory complete communication with each other through the bus; the processor is configured to invoke a computer program in the memory to perform the three-dimensional spatial model data processing method for mineral area-oriented exploration of any of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410079706.5A CN117576335B (en) | 2024-01-19 | 2024-01-19 | Three-dimensional space model data processing method and system for mineral area investigation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410079706.5A CN117576335B (en) | 2024-01-19 | 2024-01-19 | Three-dimensional space model data processing method and system for mineral area investigation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117576335A CN117576335A (en) | 2024-02-20 |
| CN117576335B true CN117576335B (en) | 2024-04-26 |
Family
ID=89892332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410079706.5A Active CN117576335B (en) | 2024-01-19 | 2024-01-19 | Three-dimensional space model data processing method and system for mineral area investigation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117576335B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117576335B (en) * | 2024-01-19 | 2024-04-26 | 长安大学 | Three-dimensional space model data processing method and system for mineral area investigation |
| CN117876623B (en) * | 2024-03-11 | 2024-06-04 | 云南省有色地质局三0六队 | Mineral resource investigation digital terrain model generation method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179415A (en) * | 2019-12-30 | 2020-05-19 | 核工业北京地质研究院 | Three-dimensional geological model construction method for calcium-bonded rock type uranium ore |
| CN113361112A (en) * | 2021-06-10 | 2021-09-07 | 河南发恩德矿业有限公司 | Method for realizing gold mine resource detection by using existing gold mine exploration and mining data in auxiliary manner |
| CN115294290A (en) * | 2022-08-18 | 2022-11-04 | 四川水利职业技术学院 | Method for extracting data in geological mineral exploration |
| CN115983505A (en) * | 2023-03-20 | 2023-04-18 | 山东黄金地质矿产勘查有限公司 | Solid mineral three-dimensional ore formation prediction method and device |
| WO2023185734A1 (en) * | 2022-03-28 | 2023-10-05 | 中国矿业大学 | Mine digital model construction method based on three-dimensional point cloud database |
| CN117576335A (en) * | 2024-01-19 | 2024-02-20 | 长安大学 | Three-dimensional space model data processing method and system for mineral area investigation |
-
2024
- 2024-01-19 CN CN202410079706.5A patent/CN117576335B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179415A (en) * | 2019-12-30 | 2020-05-19 | 核工业北京地质研究院 | Three-dimensional geological model construction method for calcium-bonded rock type uranium ore |
| CN113361112A (en) * | 2021-06-10 | 2021-09-07 | 河南发恩德矿业有限公司 | Method for realizing gold mine resource detection by using existing gold mine exploration and mining data in auxiliary manner |
| WO2023185734A1 (en) * | 2022-03-28 | 2023-10-05 | 中国矿业大学 | Mine digital model construction method based on three-dimensional point cloud database |
| CN115294290A (en) * | 2022-08-18 | 2022-11-04 | 四川水利职业技术学院 | Method for extracting data in geological mineral exploration |
| CN115983505A (en) * | 2023-03-20 | 2023-04-18 | 山东黄金地质矿产勘查有限公司 | Solid mineral three-dimensional ore formation prediction method and device |
| CN117576335A (en) * | 2024-01-19 | 2024-02-20 | 长安大学 | Three-dimensional space model data processing method and system for mineral area investigation |
Non-Patent Citations (2)
| Title |
|---|
| "基于3DMine软件三维地质模型在陕南汉阴长沟金矿中的应用";褚娜娜;《中国优秀硕士学位论文全文数据库（基础科学辑）》;20130715(第07期);第A011-64页 * |
| 三维地质建模技术在矿产勘查中的应用;刘琳琳;贾德龙;李永玲;;矿床地质;20141215(第S1期);第742-743页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117576335A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117576335B (en) | Three-dimensional space model data processing method and system for mineral area investigation | |
| US11693140B2 (en) | Identifying hydrocarbon reserves of a subterranean region using a reservoir earth model that models characteristics of the region | |
| Pazand et al. | Using TOPSIS approaches for predictive porphyry Cu potential mapping: A case study in Ahar-Arasbaran area (NW, Iran) | |
| US11486230B2 (en) | Allocating resources for implementing a well-planning process | |
| US11815650B2 (en) | Optimization of well-planning process for identifying hydrocarbon reserves using an integrated multi-dimensional geological model | |
| CN106066493A (en) | Bayes's petrofacies method of discrimination and device | |
| CN113344050B (en) | Lithology intelligent recognition method and system based on deep learning | |
| CN103345566A (en) | Geochemical exploration anomaly recognizing and evaluating method based on contents contained in geology | |
| Esmaeiloghli et al. | Spatio-geologically informed fuzzy classification: an innovative method for recognition of mineralization-related patterns by integration of elemental, 3D spatial, and geological information | |
| CA3109021A1 (en) | Facilitating hydrocarbon exploration by applying a machine-learning model to basin data | |
| CN114997501A (en) | Deep learning mineral resource classification prediction method and system based on sample unbalance | |
| Iqbal et al. | Boreholes data analysis architecture based on clustering and prediction models for enhancing underground safety verification | |
| Hall et al. | Sequential value of information for subsurface exploration drilling | |
| Wu et al. | Coarse-graining research of the thermal infrared anomalies before earthquakes in the Sichuan area on Google Earth engine | |
| US20240127039A1 (en) | Geologic learning framework | |
| Chen et al. | A geospatial case‐based reasoning model for oil–gas reservoir evaluation | |
| Zhang et al. | Development and future prospects of quantitative mineral assessment in China | |
| CN114239655A (en) | Seismic facies recognition model training method and device and seismic facies prediction method and device | |
| Msaddek et al. | A fuzzy mathematical model for evaluation of rock-fracture and structural complexity: application for Southern Atlas in Tunisia | |
| San Roman Iturbide et al. | Identification of geotechnical units in soil exploration through principal component analysis and clustering | |
| CN116090662B (en) | Knowledge-graph-based method and system for predicting potential of copper mine outside environment and electronic equipment | |
| Mbah et al. | Implementation of Statistical Analysis to Determine Optimum Ultimate Pit | |
| Vantassel et al. | AutoHVSR: A machine-learning-supported algorithm for the fully-automated processing of horizontal-to-vertical spectral ratio measurements | |
| Dagasan | Development of a grade control technique optimizing dilution and ore loss trade-off in lateritic bauxite deposits | |
| Granek et al. | Resource Management through Machine Learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |