CN-122020429-A - Ore type classification and ore body delineation method, system, medium and product

CN122020429ACN 122020429 ACN122020429 ACN 122020429ACN-122020429-A

Abstract

The invention relates to the technical field of ore type division, in particular to an ore type division and ore body delineating method, an ore type division and ore body delineating system, a medium and a product. The ore type dividing and ore body delineating method constructs a sample set by using a plurality of groups of exploration auxiliary samples with determined ore types in ore deposits, acquires target element content data in each sample set by using p-XRF, and inputs the target element content data into a machine learning classification model for training after standardized processing. And automatically classifying the exploration subsamples of unknown types in the ore deposit based on the trained machine learning classification model. Further, sequentially drawing a drilling histogram, a geological section and a ore body distribution diagram according to the classification result. The method realizes intelligent and efficient division of ore types, effectively reduces subjective influence of manual division, and remarkably improves division efficiency and accuracy.

Inventors

ZHANG YIFAN
WANG BIAO
FAN YU
Zhao Shouheng
WU SHUO
CHANG SEN
Zhou Taofa
Bao Qinyue
WANG SHIWEI

Assignees

合肥工业大学
铜陵有色金属集团股份有限公司矿产资源中心

Dates

Publication Date: 20260512
Application Date: 20260410

Claims (10)

1. A method of ore type classification and ore body delineation, comprising: acquiring a plurality of groups of exploration auxiliary samples in a mineral deposit, determining the ore types of the exploration auxiliary samples, and taking the exploration auxiliary samples as a sample set; After standardized processing is carried out on the content data, the content data of each target element in each sample is taken as a characteristic variable, the known ore type in the ore deposit is taken as a real label, a machine learning classification model is built and trained, and thus, the trained machine learning classification model is obtained; obtaining content data of each target element of an unknown ore type exploration auxiliary sample in any borehole through p-XRF, and inputting the content data into a trained machine learning classification model after standardized processing, so as to output the ore type corresponding to the exploration auxiliary sample; The method comprises the steps of mapping ore types corresponding to exploration subsamples in all drilled holes into corresponding drilled hole histograms according to space positions of the ore types, forming the drilled hole histograms with the ore types being marked in a layered mode, delineating and connecting ore body boundaries based on the drilled hole histograms with the ore types being marked in a layered mode and combining geological constraint conditions, so that geological section views of ore beds are drawn, and compiling ore body distribution diagrams according to the geological section views to obtain space distribution characteristics of different ore types in the ore beds.
2. The ore type classification and ore body delineation method of claim 1, wherein the known ore types in the ore deposit are classified and defined based on ore deposit survey reports and ore characteristics, and the classification principle of ore types in the ore deposit is that the ore types of the ore deposit are classified based on the characteristics of industry index, ore characteristics, mineral composition, copper sulfur grade and content of harmful elemental magnesium on the basis of geological studies.
3. The method for classifying and circumscribing ore types as set forth in claim 1, wherein the ore types in the ore deposit are classified into (1) copper-containing pyrrhotite serpentine ore: cu >0.5%, S >12%, mg >5%; (2) high grade copper-containing pyrrhotite ore: cu >1%, S >12%, mg <1%; (3) copper-containing pyrrhotite ore: 0.5% < Cu <1%, S >10%, mg <1%; (4) copper-containing pyrrhotite type ore: cu >0.5%, S <6%, si >14%, al >5%; (5) copper-containing pyrrhotite serpentine type ore: cu >0.5%, S <6%, 1%; mg <5%, mn >0.1%; (6) copper-containing siltstone type ore: cu >0.5%, S <6%, si >28%, ca <2%, mg <1%; (7) pyrrhotite ore: cu > 0.3%, S < 20%, mg <1%; S <8%, si >14%, al >5%, cu >0.5%, cu >5%, S <5%, S > 5%.
4. The ore type classification and ore body delineation method of claim 1, wherein the target elements include Cu, S, mg, al, si, fe, ca, mn, mo, W, zn, ag, sr, ti, cr, co, ni, as, se, rb, Y, zr, nb, cd, sn, sb, hg, pb, bi, th, U, wherein the seven elements of definition Cu, S, mg, al, si and Fe are primary defining elements of ore type classification, and definition Mo, W, zn, ag, sr, ti, cr, co, ni and As are secondary defining elements of ore type classification.
5. The ore type classification and ore body circumscribing method as claimed in claim 1, wherein each of the subsidiary explorations is subjected to pretreatment before p-XRF measurement, the pretreatment is carried out by crushing the subsidiary explorations to 200 meshes, sequentially carrying out shrinkage and drying treatment, and then placing the dried subsidiary explorations in a powder press for compression molding, thereby preparing the pretreated subsidiary explorations.
6. The ore type classification and ore body delineation method of claim 1, wherein the target element content data in each exploration subsamples obtained by p-XRF is corrected before being input into the machine learning classification model, and the corrected target element content data is input into the machine learning classification model as a characteristic variable.
7. The ore type classification and ore body delineation method as set forth in claim 6, wherein the correction process includes selecting an exploration subsamples with representative element contents, measuring the exploration subsamples by a sample dissolution method to obtain data of each target element content as a reference value, establishing a linear regression model between the element contents measured by p-XRF and the reference values measured by the sample dissolution method, and correcting the original data measured by p-XRF to enable correlation coefficients between the corrected data and the reference values to reach more than 0.95.
8. An ore type classification and ore body delineation system employing the ore type classification and ore body delineation method of any one of claims 1-7, the ore type classification and ore body delineation system comprising: the sample acquisition module is used for acquiring a plurality of groups of exploration auxiliary samples in the ore deposit, determining the corresponding ore types of the exploration auxiliary samples and further constructing a sample set; the data collection processing module is used for obtaining the content data of the target elements of each sample in the sample set through p-XRF and carrying out standardized processing on the content data; The model construction and training module is used for taking the content data of the target elements in each sample obtained by the data collection and processing module as a characteristic variable, taking the known ore types in the ore deposit as real labels, constructing a machine learning classification model and completing training to obtain a trained machine learning classification model; The prediction module predicts an exploration auxiliary sample of an unknown ore type in any borehole based on the trained machine learning classification model and outputs the ore type corresponding to the exploration auxiliary sample; The drawing module is used for firstly mapping ore types corresponding to the exploration auxiliary samples in all the drill holes into corresponding drill hole histograms according to the spatial positions to form drill hole histograms with ore type layered labels, completing delineation and connection of ore body boundaries according to the drill hole histograms with the ore type layered labels and combining geological constraint conditions, drawing geological section views of ore beds, and compiling ore body distribution diagrams according to the geological section views to obtain spatial distribution characteristics of different ore types in the ore beds.
9. A storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the ore type classification and ore body delineation method of any one of claims 1 to 7.
10. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the ore type classification and ore body delineation method of any one of claims 1 to 7.

Description

Ore type classification and ore body delineation method, system, medium and product Technical Field The invention relates to the technical field of ore type classification, in particular to an ore type classification and ore body delineation method, an ore type classification and ore body delineation system, a storage medium and a computer program product. Background Mineral deposit exploration is an important link before mineral deposit development and utilization. By adopting the system to carry out geological survey, ore body delineation, ore type division and determination of valuable and harmful element distribution characteristics can be realized, so that scientific basis is provided for subsequent exploitation and production. However, the classification of ore types in current borehole cataloging processes mainly depends on visual observation of geological personnel, lacks quantitative data support, results in relatively coarse cataloging results, is easily interfered by subjective factors, and is difficult to accurately define boundaries of different ore types. In addition, the existing geological exploration report is used for defining ore bodies according to ore grades, and spatial distribution characteristics of the internal structures of the ore bodies and the ore types are difficult to systematically reveal. If the traditional method is adopted to finely divide the ore types, a large amount of chemical analysis work is required to be carried out, so that the method is time-consuming and labor-consuming, the problem of data acquisition lag exists, and the requirement of high-efficiency production of modern mines is difficult to meet. Meanwhile, the existing ore classification technology depends on single element or single index, and the comprehensive utilization of multi-element information is lacking, so that the accuracy and reliability of classification results are insufficient, and the related method often involves a complex sample pretreatment process, so that the operation difficulty and the time cost are further increased. In actual production, due to complex geological conditions of mining areas, mining is usually required to face mixed mining and dressing and smelting treatment of various ore types, and the traditional ore classification method is difficult to adapt to the requirements of fine classification and quick identification of the various ore types, so that popularization and application of the method in large-scale ore deposit exploration and production practice are restricted. In recent years, data-driven techniques have been increasingly used in geologic research. As a high-efficiency data processing means, machine learning significantly improves the processing and analysis capacity of mass geochemical data, and becomes an important research hotspot in the field of quantitative geography gradually, and the application of various machine learning algorithms in the field of geography is continuously expanded and deepened. During mine production, large amounts of ore geochemical data have been accumulated, mainly from routine laboratory chemical analysis and p-XRF testing. The conventional laboratory analysis method still takes the dominant role in the aspects of core sample splitting element analysis, ore geochemistry anomaly identification and the like due to higher test precision. However, the method generally needs to go through a plurality of links such as sample collection, transportation, processing, chemical analysis and the like to obtain a test result, and the whole flow period is long, the cost is high, and the actual requirements of quickly acquiring data and making a decision with high efficiency are difficult to meet. Disclosure of Invention The invention provides an ore type classification and ore body delineation method, an ore type classification and ore body delineation system, a storage medium and a computer program product, which are used for solving the technical problems that the subjective judgment of researchers is highly dependent and the accuracy of classification results is insufficient in the existing ore type classification process. The ore type classification and ore body delineation method comprises the following steps of obtaining a plurality of groups of exploration subsamples in a mineral deposit, determining the ore types of the exploration subsamples, and taking the exploration subsamples as a sample set. After the content data is standardized, the content data of each target element in each sample is used as a characteristic variable, the known ore type in the ore deposit is used as a real label, a machine learning classification model is built and trained, and therefore the trained machine learning classification model is obtained. And obtaining content data of each target element of the unknown ore type exploration auxiliary sample in any borehole through p-XRF, and inputting the content data into a machine learning classification model wh