CN110907379A - Method for measuring components and content of Anshan type iron ore based on random forest algorithm - Google Patents
Method for measuring components and content of Anshan type iron ore based on random forest algorithm Download PDFInfo
- Publication number
- CN110907379A CN110907379A CN201911334487.6A CN201911334487A CN110907379A CN 110907379 A CN110907379 A CN 110907379A CN 201911334487 A CN201911334487 A CN 201911334487A CN 110907379 A CN110907379 A CN 110907379A
- Authority
- CN
- China
- Prior art keywords
- sample
- content
- spectrum
- random forest
- forest algorithm
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
- G01N21/3103—Atomic absorption analysis
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention aims to determine Fe in blast hole powder in situ, in real time, rapidly and efficiently in a blast hole site of an open pit mine3O4、Fe2O3、SiO2And the content of each component of TFe, based on which, a method for measuring the components and the content of the Anshan-type iron ore based on a random forest algorithm is provided, which belongs to the technical field of mine geological analysis. The method comprises the steps of firstly carrying out visible light-near infrared spectrum test on an open stope blast hole powder sample site, then applying a random forest algorithm to construct a corresponding relation between a sample spectrum and components based on the spectral characteristics of an experimental sample, and realizing Fe in the blast hole powder sample according to the established corresponding relation3O4、Fe2O3、SiO2And the prediction and estimation of the TFe component content lay a foundation for the real-time rapid delineation and accurate zoning of the mine stope ore body boundary.
Description
Technical Field
The invention belongs to the technical field of mine geological analysis, and particularly relates to a method for measuring the components and content of Anshan-type iron ore based on a random forest algorithm.
Background
The Anshan-type iron ore is the most important deposit modified iron ore deposit in China, accounts for about 50 percent of the total iron ore reserves in China, and is the first place in China. Since this ore is mainly composed of alternating thin layers of siliceous (flint, jade, quartz) and ferrous (hematite, magnetite), it is also known internationally as Bandcast Iron Formations (BIF).
In the conventional method for measuring the content of the iron ore components, the content of each component in an experimental sample needs to be determined by different methods. The traditional determination method has the defects of high workload, high cost, complex operation, long period and the like although the accuracy is high, and can not perform instant determination on the open pit powdery experimental sample. Although the research at home and abroad also has the advantage of a method for detecting ores by using a spectrum test, such as the application of an I-7000 portable mine grade component analyzer for grade analysis, the instrument tests X-ray spectrum, also needs to carry out a necessary pretreatment process on an experimental sample, and can only determine the element composition of the sample, but cannot test the contents of mineral composition and other substance components in the experimental sample. Therefore, how to determine the content of all the material components in the iron ore in situ, in real time, rapidly and efficiently by using a spectral analysis method for the Anshan iron ore is a problem to be studied intensively.
Disclosure of Invention
The invention aims to determine Fe in blast hole powder in situ, in real time, rapidly and efficiently in a blast hole site of an open pit mine3O4、Fe2O3、SiO2And TFe, therefore, a method for measuring the components and the content of the Anshan iron ore based on a random forest algorithm is provided. The method comprises the steps of firstly carrying out visible light-near infrared spectrum test and chemical component test on blast hole powder samples of an open stope, and then analyzing the visible light of the samples by using a random forest algorithm based on a Matlab platformThe relation between the near infrared spectrum and the content of the components thereof, and then realizing the Fe in the blast hole powder sample according to the relation3O4、Fe2O3、SiO2And the prediction and estimation of the TFe component content lay a foundation for the real-time rapid delineation and accurate zoning of the mine stope ore body boundary.
The invention has the technical scheme that a method for measuring the components and the content of Anshan type iron ore based on a random forest algorithm comprises the following steps:
1) performing visible light-near infrared spectrum test on a known sample with known component content, and analyzing the reflection spectrum and Fe of the known sample by using a random forest algorithm3O4、Fe2O3、SiO2And the corresponding relation among the contents of the 4 components TFe, and performing adjustment and training on the corresponding relation to serve as a database;
preferably, the database is overlaid with Fe3O4、Fe2O3、SiO2Spectra of known samples having different TFe component contents;
2) sampling a powdery sample at a blast hole after blasting at a certain place on an iron ore stope site, performing spectrum test on the sample by using a visible light-near infrared spectrometer, and taking an obtained spectrum curve as a spectrum curve of an experimental sample at the blast hole;
furthermore, in order to ensure the representativeness, the accuracy and the systematicness of the sample, a ditch method is adopted for sampling at the position of a shot hole with better rock mine exposure, namely, after the powder pile of the shot hole is selected, a tool is used for shoveling a groove which runs through the radius of the powder pile to ensure that the section of the groove is completely exposed, then a middle section of the groove is cut on the wall of the groove and penetrates through the whole layer to collect the sample, and then the sample is uniformly mixed and loaded into a sample box to be used as an experimental sample;
3) applying a random forest algorithm based on the visible-near infrared spectral characteristics and Fe of the known samples in the existing database of step 1)3O4、Fe2O3、SiO2And the content of TFe component, for Fe in the experimental sample of the step 2)3O4、Fe2O3、SiO2Predicting and estimating the TFe content;
3.1) taking the visible-near infrared spectrum characteristics of the known samples in the existing database and corresponding Fe3O4The content data is used as a training sample in a model training library, and a random forest algorithm is applied to establish the reflection spectrum and Fe of a known sample3O4The corresponding relation of the contents; then, the spectrum of the known sample in the existing database is used as the prior knowledge, and the relationship is used for the Fe of the experimental sample in the step 2)3O4Predicting and estimating the content;
3.2) taking visible-near infrared spectrum characteristics of known samples in the existing database and corresponding Fe2O3The content data is used as a training sample in a model training library, and a random forest algorithm is applied to establish the reflection spectrum and Fe of a known sample2O3The corresponding relation of the contents; then, the spectrum of the known sample in the existing database is used as the prior knowledge, and the relationship is used for the Fe of the experimental sample in the step 2)2O3Predicting and estimating the content;
3.3) taking the visible-near infrared spectrum characteristics of the known samples in the existing database and corresponding SiO2The content data is used as a training sample in a model training library, and a random forest algorithm is applied to establish the reflection spectrum and SiO of a known sample2The corresponding relation of the contents; then, the spectrum of the known sample in the existing database is used as the prior knowledge, and the SiO of the experimental sample in the step 2) is subjected to the relation2Predicting and estimating the content;
3.4) taking the visible light-near infrared spectrum characteristics of the known sample in the existing database and the data corresponding to the TFe content as training samples in a model training library, and establishing the corresponding relation between the reflection spectrum of the known sample and the TFe content by applying a random forest algorithm; and predicting and estimating the TFe content of the experimental sample in the step 2) by using the spectrum of the known sample in the existing database as a priori knowledge through the relation.
Further, the number of feature variables in the random forest algorithm in the steps 1 and 3) is 31, and the number of decision trees is 500.
Compared with the prior art, the invention has the advantages that:
1. the method can determine Fe in the ore body in situ, in real time, quickly and efficiently on the site of the iron ore open stope3O4、Fe2O3、SiO2And the content of each component of TFe.
2. Method of the invention, predicting estimated Fe3O4、Fe2O3、SiO2And the average error of the contents of TFe and the like is less than 2 percent, and the precision is higher.
Detailed Description
Examples
A method for measuring the components and the content of Anshan-type iron ore based on a random forest algorithm comprises the following steps:
1. database establishment
The random forest algorithm is a machine learning algorithm and is proposed in 2001 by the american scientist Leo Breiman, and the essence of the random forest algorithm is an improvement on the decision tree algorithm. A random forest is established in a random mode, the forest is composed of a plurality of decision trees, and each decision tree is not related; after the forest is obtained, when a new input sample enters, each decision tree in the random forest carries out independent judgment and analysis on the input sample, and then the prediction results of the decision trees are integrated. The integration modes of regression and classification problems are slightly different, the classification problems adopt a voting system, each decision tree votes for a category, and the category which votes most is obtained as a final result; in the regression problem, the prediction result obtained by each decision tree is a real number, and the final prediction result is the average value of the prediction results of the decision trees.
The classification accuracy of the random forest algorithm depends on the parameters k (number of trees) and d (number of feature variables). In this embodiment, since the number of the spectral bands of the experimental sample is 973 bands, that is, the total number of the spectral characteristic variables is 973, the characteristic variable d is 31, and the number k of the trees is 500.
For Anshan iron ore samples with known component content (the selected sample requires Fe)3O4、Fe2O3、SiO2And TFe component content is different from other samples, 150 block samples and 150 powder samples are selected) to carry out visible light-near infrared spectrum test, and then based on a Matlab platform, by compiling related programs, the relation between the reflection spectrum of the sample and the content of each component is analyzed by applying a random forest algorithm, and the relation is subjected to tuning and training to serve as a database.
2. Sampling
In order to ensure the representativeness, the accuracy and the systematicness of the sample, a shovel channel method is adopted for sampling at a blast hole with better exposure of rock ore, namely after the blast hole powder pile is selected, a tool shovel is used for shoveling a groove which runs through the radius of the powder pile so as to completely expose the section of the groove, then a middle section of the groove is cut out to run through the whole layer to collect 1kg of sample, then the sample is uniformly mixed and loaded into a sample box to be used as an experimental sample;
3. visible-near infrared spectroscopy test
The sample spectrum determination method comprises the following steps:
during measurement, the sampling integration time is set to be 3s, the sample observation surface is kept horizontal, the lens of the spectrograph is perpendicular to the sample observation surface, the distance is 0.5-1 m, the angle of view is based on the coverage of the sample test surface, and a sample spectrum curve is collected. In order to reduce the influence of a solar radiation propagation path and aerosol, a spectrum test is carried out at 10: 00-14: 00, the sky is clear and cloudless during measurement, and the solar altitude is 50 degrees.
4. Prediction estimation
And adopting a random forest algorithm, taking the visible light-near infrared spectrum of the Anshan-type iron ore sample in the existing database and the content of the corresponding component as training samples in a model training library, establishing a corresponding relation between the visible light-near infrared spectrum of the Anshan-type iron ore sample and the corresponding component, taking the visible light-near infrared spectrum of the experimental sample to be tested as a sample to be predicted, and applying the relation to predict and estimate the content of the component of the experimental sample to obtain the content of the corresponding component of the experimental sample.
2.1) based on a Matlab platform, adopting a random forest algorithm, and providing the spectrum of the Anshan-type iron ore sample and the corresponding Fe by using a model training library3O4The content is a training sampleSetting the forest tree to be 500 and the number of the characteristic variables to be 31, establishing the corresponding relation between the forest tree and the characteristic variables, taking the spectrum of the experimental sample to be tested as a prediction sample, and carrying out Fe comparison3O4And (4) carrying out prediction estimation on the content.
In this embodiment, the establishment of the correspondence based on the random forest algorithm and the prediction estimation are both realized by a Matlab software platform, and the specific operation steps are as follows:
2.1.1) reading the spectral data of the Anshan-type iron ore and Fe provided in a model training library by writing related programs based on a Matlab platform3O4And (3) content data is used as a model training sample, then a random forest algorithm is applied to set the random forest tree to be 500, the number of characteristic variables is set to be 31, and the model is trained according to the random forest construction steps to establish the corresponding relation between the random forest tree and the characteristic variables.
2.1.2) reading the spectral data of the experimental sample to be tested through a Matlab platform, and then carrying out Fe detection on the experimental sample according to the spectral data of the experimental sample based on the established corresponding relation3O4Predicting and estimating the content to determine Fe in the experimental sample3O4And (4) content.
2.2) based on a Matlab platform, adopting a random forest algorithm and providing the spectrum of the Anshan-type iron ore sample and the corresponding Fe by using a model training library2O3Setting the random forest tree as 500 and the number of characteristic variables as 31 as training samples, establishing the corresponding relation between the random forest tree and the characteristic variables, and taking the spectrum of an experimental sample to be tested as a prediction sample to carry out Fe test2O3And (4) predicting and estimating the content, wherein the specific steps are as above.
2.3) based on a Matlab platform, adopting a random forest algorithm and providing the spectrum of the Anshan type iron ore sample and the corresponding SiO by using a model training library2Setting the random forest tree as 500 and the number of characteristic variables as 31 as training samples, establishing the corresponding relation between the random forest tree and the characteristic variables, and taking the spectrum of the experimental sample to be tested as a prediction sample to test the SiO content2And (4) predicting and estimating the content, wherein the specific steps are as above.
And 2.4) based on a Matlab platform, adopting a random forest algorithm, taking the spectrum of the Anshan-type iron ore sample provided by a model training library and the corresponding TFe content as training samples, setting the random forest particle tree to be 500, setting the number of characteristic variables to be 31, establishing the corresponding relation between the spectrum of the Anshan-type iron ore sample and the corresponding TFe content, and taking the spectrum of an experimental sample to be tested as a prediction sample to predict and estimate the TFe content, wherein the specific steps are as above.
The prediction estimation method is further verified, and the comparison of the chemical actual measurement of 20 groups of samples and the result of the prediction estimation of the invention shows that Fe3O4、Fe2O3、SiO2And the average error of the content prediction estimation of TFe and the like is less than 2%, the precision is high, and the prediction result is ideal. See table 1, table 2, table 3, table 4.
TABLE 1 Fe of samples3O4Data sheet for content measurement
| Sample numbering | Actually measured Fe3O4Content (%) | Prediction of Fe3O4Content (%) | Difference (%) |
| 1 | 0.84 | 1.06 | 0.22 |
| 2 | 13.92 | 13.02 | -0.90 |
| 3 | 7.99 | 5.29 | -2.70 |
| 4 | 7.06 | 7.98 | 0.93 |
| 5 | 16.05 | 14.86 | -1.18 |
| 6 | 12.21 | 11.51 | -0.71 |
| 7 | 0.61 | 1.14 | 0.53 |
| 8 | 14.08 | 15.63 | 1.54 |
| 9 | 27.84 | 25.38 | -2.46 |
| 10 | 27.84 | 26.58 | -1.26 |
| 11 | 2.19 | 2.87 | 0.68 |
| 12 | 19.88 | 19.24 | -0.64 |
| 13 | 20.24 | 19.89 | -0.35 |
| 14 | 8.15 | 8.21 | 0.06 |
| 15 | 9.18 | 8.84 | -0.34 |
| 16 | 34.38 | 35.76 | 1.38 |
| 17 | 34.54 | 35.89 | 1.34 |
| 18 | 51.33 | 45.39 | -5.94 |
| 19 | 44.95 | 40.67 | -4.28 |
| 20 | 0.87 | 1.14 | 0.27 |
TABLE 2 Fe of samples2O3Data sheet for content measurement
| Sample numbering | Actually measured Fe2O3Content (%) | Prediction of Fe2O3Content (%) | Difference (%) |
| 1 | 33.22 | 34.82 | 1.60 |
| 2 | 33.13 | 32.12 | -1.01 |
| 3 | 37.62 | 33.44 | -4.18 |
| 4 | 26.43 | 26.81 | 0.38 |
| 5 | 37.00 | 34.25 | -2.75 |
| 6 | 36.78 | 35.81 | -0.97 |
| 7 | 27.20 | 26.69 | -0.50 |
| 8 | 10.90 | 11.33 | 0.43 |
| 9 | 15.19 | 15.89 | 0.71 |
| 10 | 14.30 | 14.77 | 0.47 |
| 11 | 32.32 | 31.25 | -1.07 |
| 12 | 9.73 | 10.70 | 0.96 |
| 13 | 10.44 | 10.67 | 0.23 |
| 14 | 29.57 | 28.11 | -1.46 |
| 15 | 39.26 | 39.35 | 0.10 |
| 16 | 11.93 | 9.26 | -2.68 |
| 17 | 12.14 | 9.04 | -3.10 |
| 18 | 0.26 | 2.03 | 1.77 |
| 19 | 0.20 | 3.83 | 3.63 |
| 20 | 29.07 | 29.45 | 0.38 |
TABLE 3 SiO of the samples2Data sheet for content measurement
| Sample numbering | Actually measured SiO2Content (%) | Predicting SiO2Content (%) | Difference (%) |
| 1 | 62.78 | 62.27 | -0.51 |
| 2 | 53.39 | 55.03 | 1.64 |
| 3 | 55.01 | 60.07 | 5.06 |
| 4 | 68.91 | 66.65 | -2.26 |
| 5 | 47.39 | 51.02 | 3.63 |
| 6 | 49.96 | 51.39 | 1.43 |
| 7 | 72.91 | 72.88 | -0.03 |
| 8 | 76.30 | 70.64 | -5.66 |
| 9 | 56.23 | 56.29 | 0.06 |
| 10 | 58.85 | 57.69 | -1.16 |
| 11 | 66.03 | 66.01 | -0.02 |
| 12 | 66.77 | 66.19 | -0.58 |
| 13 | 70.71 | 66.08 | -4.63 |
| 14 | 61.12 | 65.75 | 4.63 |
| 15 | 52.24 | 52.30 | 0.06 |
| 16 | 53.28 | 53.57 | 0.29 |
| 17 | 54.07 | 53.11 | -0.96 |
| 18 | 45.68 | 46.92 | 1.24 |
| 19 | 55.29 | 54.39 | -0.90 |
| 20 | 71.40 | 70.53 | -0.87 |
TABLE 4 TFe content measurement data Table of samples
| Sample numbering | Measured TFe content (%) | Predicted TFe content (%) | Difference (%) |
| 1 | 23.86 | 25.21 | 1.35 |
| 2 | 23.61 | 24.40 | 0.79 |
| 3 | 34.99 | 31.15 | -3.84 |
| 4 | 33.27 | 31.96 | -1.31 |
| 5 | 37.52 | 35.02 | -2.50 |
| 6 | 34.59 | 33.68 | -0.91 |
| 7 | 19.48 | 19.91 | 0.43 |
| 8 | 17.83 | 21.97 | 4.14 |
| 9 | 30.79 | 30.82 | 0.03 |
| 10 | 30.17 | 30.59 | 0.42 |
| 11 | 24.21 | 24.15 | -0.06 |
| 12 | 21.21 | 22.55 | 1.34 |
| 13 | 21.96 | 23.42 | 1.46 |
| 14 | 26.60 | 24.12 | -2.48 |
| 15 | 34.13 | 33.99 | -0.14 |
| 16 | 33.25 | 32.87 | -0.38 |
| 17 | 33.51 | 33.07 | -0.44 |
| 18 | 37.35 | 36.66 | -0.69 |
| 19 | 32.69 | 32.06 | -0.63 |
| 20 | 19.68 | 24.23 | 4.55 |
Claims (4)
1. A method for measuring the components and the content of Anshan-type iron ore based on a random forest algorithm is characterized by comprising the following steps of:
1) performing visible light-near infrared spectrum test on a known sample with known component content, and analyzing the reflection spectrum and Fe of the known sample by using a random forest algorithm3O4、Fe2O3、SiO2And the corresponding relation among the contents of the 4 components TFe, and performing adjustment and training on the corresponding relation to serve as a database;
2) sampling a powdery sample at a blast hole after blasting at a certain place on an iron ore stope site, performing spectrum test on the sample by using a visible light-near infrared spectrometer, and taking an obtained spectrum curve as a spectrum curve of an experimental sample at the blast hole;
3) applying a random forest algorithm based on the visible-near infrared spectral characteristics and Fe of the known samples in the existing database of step 1)3O4、Fe2O3、SiO2And the content of TFe component, for Fe in the experimental sample of the step 2)3O4、Fe2O3、SiO2Predicting and estimating the TFe content;
3.1) taking the visible-near infrared spectrum characteristics of the known samples in the existing database and corresponding Fe3O4The content data is used as a training sample in a model training library, and a random forest algorithm is applied to establish the reflection spectrum and Fe of a known sample3O4The corresponding relation of the contents; then, the spectrum of the known sample in the existing database is used as the prior knowledge, and the relationship is used for the Fe of the experimental sample in the step 2)3O4Predicting and estimating the content;
3.2) taking visible-near infrared spectrum characteristics of known samples in the existing database and corresponding Fe2O3The content data is used as a training sample in a model training library, and a random forest algorithm is applied to establish the reflection spectrum and Fe of a known sample2O3The corresponding relation of the contents; then, the spectrum of the known sample in the existing database is used as the prior knowledge, and the relationship is used for the Fe of the experimental sample in the step 2)2O3Predicting and estimating the content;
3.3) taking the visible-near infrared spectrum characteristics of the known samples in the existing database and corresponding SiO2The content data is used as a training sample in a model training library, and a random forest algorithm is applied to establish the reflection spectrum and SiO of a known sample2The corresponding relation of the contents; then, the spectrum of the known sample in the existing database is used as the prior knowledge, and the SiO of the experimental sample in the step 2) is subjected to the relation2Predicting and estimating the content;
3.4) taking the visible light-near infrared spectrum characteristics of the known sample in the existing database and the data corresponding to the TFe content as training samples in a model training library, and establishing the corresponding relation between the reflection spectrum of the known sample and the TFe content by applying a random forest algorithm; and predicting and estimating the TFe content of the experimental sample in the step 2) by using the spectrum of the known sample in the existing database as a priori knowledge through the relation.
2. The assay of claim 1, wherein the database is overlaid with Fe3O4、Fe2O3、SiO2Spectrum of a known sample having different TFe component contents.
3. The method according to claim 1, wherein in the step 2), the sampling method comprises: sampling at the blast hole with good rock ore exposure by adopting a ditch method, and uniformly mixing to obtain an experimental sample.
4. The method for determining the composition and content of the Anshan-type iron ore based on the random forest algorithm, as claimed in claim 1, wherein the number of the characteristic variables in the inverse prediction model in the steps 1) and 3) is 31, and the number of the decision trees is 500.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911334487.6A CN110907379A (en) | 2019-12-23 | 2019-12-23 | Method for measuring components and content of Anshan type iron ore based on random forest algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911334487.6A CN110907379A (en) | 2019-12-23 | 2019-12-23 | Method for measuring components and content of Anshan type iron ore based on random forest algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110907379A true CN110907379A (en) | 2020-03-24 |
Family
ID=69827086
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911334487.6A Pending CN110907379A (en) | 2019-12-23 | 2019-12-23 | Method for measuring components and content of Anshan type iron ore based on random forest algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110907379A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11650147B2 (en) | 2021-03-24 | 2023-05-16 | Caterpillar Inc. | Systems, methods, and apparatuses for real-time characterization of rock cuttings during rock drill cutting |
| CN116148212A (en) * | 2023-04-21 | 2023-05-23 | 矿冶科技集团有限公司 | Method for rapidly determining types and contents of clay minerals in ores based on near infrared spectrum analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103674884A (en) * | 2012-09-17 | 2014-03-26 | 福建中烟工业有限责任公司 | Random forest classification method for tobacco leaf style characteristics based on near infrared spectral information |
| CN107367480A (en) * | 2017-09-20 | 2017-11-21 | 鞍钢集团矿业有限公司 | Silicon dioxide content test method in Anshan type iron mine based on thermal infrared spectrum |
| CN107576630A (en) * | 2017-09-28 | 2018-01-12 | 中国科学院昆明植物研究所 | The middle infrared spectrum portable detection system of plant various heavy content is determined simultaneously |
| CN107589094A (en) * | 2017-09-20 | 2018-01-16 | 鞍钢集团矿业有限公司 | Anshan type iron mine stone type based on spectral signature determines method |
| CN109030388A (en) * | 2018-07-17 | 2018-12-18 | 东北大学 | A kind of iron ore all iron content detection method based on spectroscopic data |
-
2019
- 2019-12-23 CN CN201911334487.6A patent/CN110907379A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103674884A (en) * | 2012-09-17 | 2014-03-26 | 福建中烟工业有限责任公司 | Random forest classification method for tobacco leaf style characteristics based on near infrared spectral information |
| CN107367480A (en) * | 2017-09-20 | 2017-11-21 | 鞍钢集团矿业有限公司 | Silicon dioxide content test method in Anshan type iron mine based on thermal infrared spectrum |
| CN107589094A (en) * | 2017-09-20 | 2018-01-16 | 鞍钢集团矿业有限公司 | Anshan type iron mine stone type based on spectral signature determines method |
| CN107576630A (en) * | 2017-09-28 | 2018-01-12 | 中国科学院昆明植物研究所 | The middle infrared spectrum portable detection system of plant various heavy content is determined simultaneously |
| CN109030388A (en) * | 2018-07-17 | 2018-12-18 | 东北大学 | A kind of iron ore all iron content detection method based on spectroscopic data |
Non-Patent Citations (5)
| Title |
|---|
| LIMING LU: "《Iron Ore：Mineralogy, Processing and Environmental Sustainability》", 24 July 2015, ELSEVIER * |
| YACHUN MAO ET AL.: "Research and Verification of a Remote Sensing BIF Model Based on Spectral Reflectance Characteristics", 《JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING》 * |
| 何群: "基于可见光—近红外光谱特征的BIF铁矿原位测定方法", 《金属矿山》 * |
| 李军华 等: "近红外光谱对天然岩石中矿物成分含量测定的研究", 《光谱学与光谱分析》 * |
| 洪达峰: "近红外光谱对天然岩石中矿物成分含量测定的研究", 《企业技术开发》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11650147B2 (en) | 2021-03-24 | 2023-05-16 | Caterpillar Inc. | Systems, methods, and apparatuses for real-time characterization of rock cuttings during rock drill cutting |
| CN116148212A (en) * | 2023-04-21 | 2023-05-23 | 矿冶科技集团有限公司 | Method for rapidly determining types and contents of clay minerals in ores based on near infrared spectrum analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Lemiere | A review of pXRF (field portable X-ray fluorescence) applications for applied geochemistry | |
| Luffel et al. | New core analysis methods for measuring reservoir rock properties of Devonian shale | |
| Manouchehrian et al. | Application of artificial neural networks and multivariate statistics to estimate UCS using textural characteristics | |
| CN104612675B (en) | A kind of carbonate formation Lithology while drilling method for quickly identifying | |
| CN110333200B (en) | Method for delineating mineralization center based on short-wave infrared spectrum | |
| CN105938099A (en) | Rock character judging method and system based on laser-induced breakdown spectroscopy | |
| CN110967311B (en) | Porphyry deposit alteration zone identification method based on infrared spectrum and magnetic susceptibility measurement | |
| CN104747183A (en) | Carbonate reservoir comprehensive classification method | |
| CN110907379A (en) | Method for measuring components and content of Anshan type iron ore based on random forest algorithm | |
| CN115019171B (en) | Non-contact surrounding rock rapid partition grading method | |
| García-Florentino et al. | A fast in situ non-invasive approach to classify mortars from a construction of high historical value | |
| Lawal et al. | Total organic carbon characterization using neural-network analysis of XRF data | |
| CN105938098A (en) | Rock soil ignition loss prediction method and system based on laser-induced breakdown spectroscopy | |
| Scott et al. | Quantitative chemical analysis of archaeological slag material using handheld X-ray fluorescence spectrometry | |
| CN115078520A (en) | Mineral geochemistry-based porphyry system mineralization evaluation method | |
| CN111272686A (en) | Hyperspectral detection method for iron grade of iron ore mineral powder | |
| CN112946780B (en) | Method and device for determining running and sliding fracture activity period | |
| CN114626014A (en) | Limestone stratum drillability prediction method based on element component multivariate regression | |
| CN106442474A (en) | Cement raw meal three moduli measuring method based on partial least squares | |
| CN111141809B (en) | Soil nutrient ion content detection method based on non-contact type conductivity signal | |
| CN110685676B (en) | Method for quantitatively identifying high-quality shale sections | |
| Parbhakar-Fox et al. | Predictive waste classification using field-based and environmental geometallurgy indicators, Mount Lyell, Tasmania | |
| Zhang et al. | Classification and recognition model of water saturation level of rock based on near-infrared spectroscopy | |
| Tobe et al. | A new method for predicting of rock fall from the tunnel face | |
| CN110927087A (en) | Anshan type iron ore magnetic susceptibility inversion method based on visible light-near infrared spectrum |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200324 |
|
| RJ01 | Rejection of invention patent application after publication |