A method for enriching strategic metal elements based on mineral x-ray response characteristics

Abstract

The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategic metal sorting and pre-enrichment. The application discloses a strategic metal element enrichment method based on mineral X-ray response characteristics, and belongs to the technical field of coal-based strategicmetal sorting and pre-enrichment.

Description

Technical Field

[0001] This invention relates to the field of coal-based strategic metal sorting and pre-enrichment technology, and in particular to a method for enriching strategic metal elements based on the X-ray response characteristics of minerals. Background Technology

[0002] Strategic metal mineral resources such as lithium, gallium, and titanium play a vital and irreplaceable role in ensuring national economic security, national defense security, and the sustainable and healthy development of strategic emerging industries, making them a frontier area of ​​resource competition among nations worldwide. Among these, gallium (Ga) is a widely used and strategically valuable metallic element used in both military and civilian industries, often referred to as the "food of the electronics industry." Currently, it is primarily used in semiconductors, catalysis, and medical fields, while also finding new applications and developments in thermal interface materials, welding, optics, and shape memory alloys.

[0003] The primary mineral for gallium extraction is bauxite. However, with the booming development of gallium applications, the consumption of gallium resources is increasing daily. Bauxite faces problems such as insufficient reserves and rising mining costs. Finding stable and reliable alternative renewable sources is the only way to solve the resource shortage. Although coal gangue has a low gallium content, it has abundant reserves and huge potential economic value. The effective utilization of gallium-rich gangue can not only reduce landfill costs but also alleviate the problem of insufficient primary gallium reserves.

[0004] Coal gangue has complex chemical properties, diverse occurrence phases, embedded non-occurring phases, and a wide particle size range, presenting challenges such as difficulty in identifying coarse-grained responses. This results in low efficiency and poor accuracy in conventional pre-enrichment methods for separating occurrence phases. Currently, mainstream intelligent gangue separators are mainly used for identifying different minerals, employing the R-value method to reduce the influence of thickness on mineral imaging. However, the effect is poor, and the identification accuracy is insufficient for separating minerals with multiple ore phases and complex embedding. Summary of the Invention

[0005] In view of the above analysis, the present invention aims to provide a method for enriching strategic metal elements based on the X-ray response characteristics of minerals, so as to solve the problems of poor sorting effect and low accuracy of existing methods for enriching strategic metal occurrence phases.

[0006] The objective of this invention is mainly achieved through the following technical solutions:

[0007] This invention provides a method for enriching strategic metallic elements based on the X-ray response characteristics of minerals, comprising the following steps:

[0008] Step 1: Determine the research area based on the target mining area and clarify the material composition and physicochemical properties of strategic metal minerals;

[0009] Step 2: Prepare a regular test sample with controllable thickness;

[0010] Step 3: Identify and detect the test samples, and establish a single mineral thickness-grayscale feature identification library based on the detection results;

[0011] Step 4: Prepare mixed samples and test them. Based on the test results, establish a thickness-grayscale feature identification library for mixed minerals with and without extrinsic phases.

[0012] Step 5: Use Python to build a support vector machine model, use the thickness-grayscale features of mixed minerals as the training set input of the model, generate the sorting threshold of the intelligent gangue sorting machine, and complete the establishment of the gangue prediction model;

[0013] Step 6: Based on the sorting threshold, sort the actual minerals in the study area to obtain concentrate and tailings. The concentrate is the occurrence phase of strategic metals, thus completing the pre-enrichment of strategic metals.

[0014] Furthermore, in step 1, after determining the study area, random samples of ore from the study area are taken for physicochemical analysis to clarify the material composition and physicochemical properties of strategic metal minerals.

[0015] Furthermore, in step 2, research samples are collected, and the strategic metal mineral composition and physicochemical properties are analyzed to determine the strategic metal occurrence / non-occurrence phase category.

[0016] Furthermore, in step 2, single minerals corresponding to the occurrence phase / non-occurrence phase are procured based on the composition of strategic metal mineral materials.

[0017] Furthermore, in step 2, the purchased single minerals are blended according to the actual composition ratio of strategic metal minerals, and then pressed into tablets to obtain regular test samples with controllable thickness.

[0018] Furthermore, in step 2, the strategic metal-bearing / non-bearing phase minerals obtained from procurement include both block and powder forms. The block strategic metal-bearing / non-bearing phase minerals for identification and detection are pre-processed and cut into regular samples of fixed thickness.

[0019] Furthermore, in step 2, the powdered strategic metal-bearing / non-bearing minerals for identification and detection are pressed into regular samples of fixed thickness using an infrared tablet press.

[0020] Furthermore, in step 2, each mineral is pressed to a fixed thickness of 10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm.

[0021] Furthermore, in step 3, an intelligent gangue separator is used to identify and detect the regular, thickness-controllable strategic metal mineral deposits in the presence / non-presence phases obtained from the tablet preparation, and to obtain mineral thickness characteristics and mineral grayscale characteristics.

[0022] Furthermore, in step 3, based on the mineral thickness and grayscale characteristics detected by the intelligent gangue separator, thickness-grayscale characteristic curves of the occurrence phase / non-occurrence phase of strategic metal minerals are plotted, the curves are fitted, and a single mineral thickness-grayscale characteristic identification library for occurrence phase / non-occurrence phase is established.

[0023] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0024] (1) The method of the present invention uses the mineral thickness-grayscale correspondence method to replace the existing R value method, and establishes a mineral thickness-grayscale recognition library, which can effectively solve the imaging feature differences caused by mineral thickness differences and achieve accurate mineral recognition.

[0025] (2) The method of this invention is the first to propose the use of intelligent gangue separator for the sorting and pre-enrichment of strategic metal minerals, providing a new method for the sorting and pre-enrichment of strategic metal minerals with complex embedding and diverse occurrence phases.

[0026] In this invention, the above-described technical solutions can be combined with each other to achieve more preferred combinations. Other features and advantages of this invention will be set forth in the following description, and some advantages may become apparent from the description or be learned by practicing the invention. The objects and other advantages of this invention can be realized and obtained through the embodiments described and the accompanying drawings, which are particularly pointed out. Attached Figure Description

[0027] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts.

[0028] Figure 1 This describes the process for establishing the sorting threshold for the intelligent gangue sorting machine of the present invention. Detailed Implementation

[0029] The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which constitute a part of the present invention and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0030] Strategic metals refer to metals used extensively in military manufacturing industries and play a crucial role in national defense. With the continuous development of the global economy, strategic metals have become an important pillar of national security and economic development. Strategic metal mineral resources such as lithium, gallium, and titanium play a vital and irreplaceable role in ensuring national economic security, national defense security, and the sustainable and healthy development of strategic emerging industries, and are at the forefront of resource competition among nations worldwide.

[0031] To pre-enrich strategic metals (including lithium, gallium, titanium, and rhenium), this invention provides a method for enriching strategic metal elements based on the X-ray response characteristics of minerals, comprising the following steps:

[0032] Step 1: Determine the research area based on the target mining area and clarify the material composition and physicochemical properties of strategic metal minerals;

[0033] In step 1 above, after the study area is determined, random samples of ore from the study area are taken for physicochemical analysis to clarify the material composition and physicochemical properties of strategic metal minerals.

[0034] Step 2: Prepare a regular test sample with controllable thickness;

[0035] The process of preparing test samples includes: collecting research samples, determining the strategic metal occurrence / non-occurrence phase category by analyzing the composition and physicochemical properties of strategic metal minerals, purchasing single minerals corresponding to the occurrence / non-occurrence phases based on the composition of strategic metal minerals, blending the purchased single minerals according to the actual composition ratio of strategic metal minerals, and preparing test samples by pressing them into tablets using an infrared tablet press to obtain regular test samples with controllable thickness.

[0036] In step 2 above, the strategic metal-bearing / non-bearing phase minerals obtained from procurement are available in both block and powder forms. The block-shaped strategic metal-bearing / non-bearing phase minerals for identification and testing are pre-processed and cut into regular samples of fixed thickness. The powder-shaped strategic metal-bearing / non-bearing phase minerals for identification and testing are pressed into regular samples of fixed thickness using an infrared tablet press. Each type of mineral is pressed into fixed thicknesses of 10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm.

[0037] Step 3: Identify and detect the test samples, and establish a single mineral thickness-grayscale feature identification library based on the detection results;

[0038] The specific process includes: using an intelligent gangue separator to identify and detect regular, thickness-controllable strategic metal mineral deposits in both the occurrence and non-occurrence phases obtained from pressing and preparing samples; the X-ray sensor and receiver of the intelligent gangue separator can automatically extract the grayscale features of the detected minerals and output the grayscale feature results; the laser sensor built into the intelligent gangue separator reads the mineral thickness; based on the mineral thickness and grayscale features detected by the intelligent gangue separator, a thickness-grayscale feature curve of the occurrence / non-occurrence phase of strategic metal minerals is plotted; the curve is fitted; and a single mineral thickness-grayscale feature identification library for occurrence / non-occurrence phases is established.

[0039] Step 4: Prepare mixed samples and test them. Based on the test results, establish a thickness-grayscale feature identification library for mixed minerals with and without extrinsic phases.

[0040] Massive strategic metal-bearing / non-bearing minerals were ground into powder with a particle size range of 0.125-0.5 mm. Using an infrared compressor, the powder was mixed according to the content ratio of each mineral component in coal gangue and pressed into mixed samples with different contents and thicknesses. The sample thickness was controlled at 10 mm, 15 mm, 20 mm, 25 mm, 30 mm, 35 mm, 40 mm, 45 mm, 50 mm, 55 mm, 60 mm, 65 mm, and 70 mm. The purpose was to use the prepared mixed minerals to imitate actual coal gangue in order to obtain the X-ray image recognition characteristics of actual coal gangue under standard thickness conditions, and to adjust the content based on this.

[0041] In step 4 above, the prepared mixed sample is tested using an intelligent gangue separator, and the thickness and grayscale of the mineral image feature values ​​are extracted. The thickness-grayscale feature curve of the mixed minerals of the strategic metal mineral occurrence phase / non-occurrence phase is plotted, the curve is fitted, and a thickness-grayscale feature recognition library of the mixed minerals of the occurrence phase / non-occurrence phase is established.

[0042] Step 5: Use Python to build a support vector machine model, use the thickness-grayscale features of mixed minerals as the training set input of the model, generate the sorting threshold of the intelligent gangue sorting machine, and complete the establishment of the gangue prediction model;

[0043] The specific process includes: using Python to build a support vector machine model, taking the thickness-grayscale features of the mixed minerals as the training set input, and using the training set data to generate sorting thresholds for the intelligent gangue sorting machine, thus establishing a gangue prediction model. This model divides the minerals into two parts, clarifying which part is concentrate and which part is tailings based on the image features of a single mineral. The sorting thresholds are as follows: Figure 1 As shown, the subsequent gangue sorting work will be carried out.

[0044] Step 6: Based on the sorting threshold, sort the actual minerals in the study area to obtain concentrate and tailings. The concentrate is the occurrence phase of strategic metals, thus completing the pre-enrichment of strategic metals.

[0045] The specific process includes: sorting the actual minerals in the study area according to the sorting threshold of the intelligent gangue separator; after the ore is mined, it is crushed to 10-70mm by a crusher and transported to the intelligent gangue separator by a conveyor belt; the intelligent gangue separator identifies strategic metal-bearing minerals / non-bearing minerals based on the sorting threshold given by the gangue prediction model, and obtains concentrate and tailings, completing the automatic sorting operation; the grade of strategic metals in the concentrate is increased from 19ppm in the original ore to 26ppm, and the concentrate recovery rate exceeds 65%, achieving the purpose of pre-enrichment.

[0046] It should be noted that the intelligent gangue separator used in this invention is equipped with a high-precision laser sensor, which can accurately read ore thickness information, while existing gangue separators can only read grayscale feature values. Furthermore, the intelligent gangue separator of this invention includes an identification module, whose function is to overcome the shortcomings of traditional pseudo-dual-energy X-ray identification modules that only process mineral grayscale features, by simultaneously processing both mineral thickness and grayscale features, performing corresponding analysis of these two features to achieve accurate identification of mineral image features, and improving the accuracy of gangue sorting. Traditional pseudo-dual-energy X-rays only read grayscale features and sort based on a single feature. Compared with existing technologies, the intelligent gangue separator of this invention uses both thickness and grayscale features for sorting, resulting in more accurate results.

[0047] It is important to emphasize that current intelligent gangue separators are based on the Lambert-Beer law, using dual-energy X-rays to eliminate the influence of thickness on mineral identification. However, in industrial applications, it is difficult to achieve true dual-energy X-ray transmission of minerals; only pseudo-dual-energy X-rays can be used. This means that the influence of thickness can only be weakened, not eliminated. While this method is effective for separating minerals with large differences in grayscale characteristics, such as coal gangue separation and gold ore separation, it is very ineffective for separating minerals with small differences in physicochemical properties, such as gangue-gangue separation. This invention uses X-ray and laser sensors to collect the thickness and grayscale features of mineral images and improves the identification module of the intelligent gangue separator. It analyzes the thickness and grayscale features in a corresponding manner, uses a support vector machine to train an image prediction model, and generates a binary sorting threshold of mineral grayscale-thickness to achieve accurate identification of mineral components.

[0048] It should be noted that the strategic metal element enrichment method based on the mineral X-ray response characteristics of this invention is applicable to strategic metal elements including lithium, gallium, titanium, and rhenium.

[0049] Example 1

[0050] The first step involved identifying the study area as a mining area in Pingshuo County, Shanxi Province. The strategic metal mineral resource was coal gangue from this mining area, with gallium being the primary strategic metal to be separated. Physicochemical property analysis of the coal gangue from the mining area revealed that its main mineral components included kaolinite, boehmite, quartz, pyrite, and illite. Among these, the gallium-bearing phases were primarily kaolinite and boehmite.

[0051] The second step involves purchasing raw ores such as kaolinite, boehmite, quartz, pyrite, and illite. These raw ores are then cut and processed into ores of regular thickness, with thicknesses set between 10-70 mm (10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm respectively). An intelligent gangue separator is used to inspect the ores. The intelligent dry separator extracts the image feature values ​​of the raw ore, namely ore thickness and grayscale. The ore thickness feature is obtained by a laser sensor, while the ore grayscale feature is obtained by processing with an X-ray excitation / receiving device. Based on the ore thickness and grayscale features, the ore feature data is fitted to obtain the distribution law of different ore feature values. The study found that under the same thickness conditions, kaolinite has the highest image grayscale value, followed by boehmite, while other minerals have relatively low grayscale values. The feature value differences between kaolinite, boehmite, and other minerals meet the sorting standards of the intelligent gangue separator.

[0052] The third step involves using an intelligent gangue separator to identify and detect the regular, thickness-controllable strategic metal mineral deposits in both the present and non-present phases obtained from the pressing and sample preparation. The X-ray sensor and receiver of the intelligent gangue separator can automatically extract the grayscale features of the detected minerals and output the grayscale feature results. The built-in laser sensor of the intelligent gangue separator reads the mineral thickness. Based on the mineral thickness and grayscale features detected by the intelligent gangue separator, a thickness-grayscale feature curve of the present / non-present phases of strategic metal minerals is plotted. The curve is fitted, and a single mineral thickness-grayscale feature identification library for the present / non-present phases is established.

[0053] The fourth step involves grinding raw ores such as kaolinite, boehmite, quartz, pyrite, and illite into powder, with the particle size controlled between 0.125-0.5 mm. Particle sizes that are too large or too small will affect the sample pressing results and cause errors. Adjustments are made based on the average content of each mineral in the coal gangue in the mining area. An infrared pressing sample preparation machine is used to press mixed minerals with different contents to simulate the mineral composition of actual coal gangue. Different thicknesses are adjusted. The infrared pressing sample preparation machine is kept at 50 tons for 2 minutes. The pressed simulated minerals are then sent to an intelligent gangue separator to extract the image feature values ​​and thickness features of the simulated mixed minerals.

[0054] The fifth step involves dividing the extracted image features into two parts and inputting them into a support vector machine (SVM) prediction model. The SVM uses 80% of the data as a training set and 20% as a test set to build an image prediction model and generate a binary sorting threshold. The test set shows a prediction accuracy of 96.11%, proving the accuracy of the model. This yields an image prediction model for the minerals in the mining area. The model and the binary sorting threshold are then input into the central database of the intelligent gangue sorting machine to prepare for sorting operations.

[0055] Step 6: The coal gangue from the mining area is crushed to 10-70mm, specifically 10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm. It is then conveyed to an intelligent gangue separator via a conveyor belt. The X-ray identification device of the intelligent gangue separator extracts the grayscale characteristics of the minerals, and the laser sensor extracts the thickness characteristics. The data is transmitted to the central processing unit. The central processing unit uses an image prediction model to classify the minerals into concentrate (strategic gallium-bearing phase minerals) and tailings (non-strategic gallium-bearing phase minerals). The identification results are transmitted to the injection system. The injection system adopts different injection measures for the concentrate and tailings. Finally, the concentrate is collected in the concentrate trough, and the tailings are collected in the tailings trough.

[0056] Physical property testing of the raw ore and concentrate revealed that the average gallium content in the raw ore was 19.72 ppm, while the average gallium content in the concentrate after pre-enrichment was 26.2 ppm. The concentrate recovery rate was higher than 65%, and the gallium enrichment ratio was 1.33.

[0057] Example 2

[0058] This embodiment focuses on the pre-enrichment of the strategic metal titanium, and the specific process is as follows:

[0059] The first step was to determine the study area as a mining area in Hebei Province, with the strategic metal mineral being ilmenite from that area. The main strategic metal to be separated was titanium. Physicochemical properties of the ore from the mining area were tested to clarify its main mineral components, including feldspar, pyroxene, quartz, montmorillonite, and titanomagnetite.

[0060] The second step involves procuring raw ores such as feldspar, pyroxene, quartz, montmorillonite, and titanomagnetite. These raw ores are then cut and processed into ores of regular thickness, with thicknesses set between 10-70 mm (10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm respectively). An intelligent gangue separator is used to inspect the ores. The intelligent dry separator extracts the image feature values ​​of the raw ore, namely ore thickness and grayscale. The ore thickness feature is obtained by a laser sensor, while the ore grayscale feature is obtained by processing with an X-ray excitation / receiving device. Based on the ore thickness and grayscale features, the ore feature data is fitted to obtain the distribution law of different ore feature values. The study found that under the same thickness conditions, titanomagnetite has the lowest image grayscale value, while other minerals have relatively high grayscale values. The difference in feature values ​​between titanomagnetite and other minerals meets the sorting standards of the intelligent gangue separator.

[0061] The third step involves using an intelligent gangue separator to identify and detect the regular, thickness-controllable strategic metal mineral deposits in both the present and non-present phases obtained from the pressing and sample preparation. The X-ray sensor and receiver of the intelligent gangue separator can automatically extract the grayscale features of the detected minerals and output the grayscale feature results. The built-in laser sensor of the intelligent gangue separator reads the mineral thickness. Based on the mineral thickness and grayscale features detected by the intelligent gangue separator, a thickness-grayscale feature curve of the present / non-present phases of strategic metal minerals is plotted. The curve is fitted, and a single mineral thickness-grayscale feature identification library for the present / non-present phases is established.

[0062] The fourth step involves grinding the raw ores, such as feldspar, pyroxene, quartz, montmorillonite, and titanomagnetite, into powder with a particle size controlled between 0.125-0.5 mm. Particle sizes that are too large or too small will affect the sample pressing results and cause errors. Adjustments are made based on the average content of each mineral in the coal gangue of the mining area. An infrared pressing sample preparation machine is used to press mixed minerals with different contents to simulate the mineral composition of actual ilmenite. Different thicknesses are adjusted, and the infrared pressing sample preparation machine is kept at 50 tons with a holding time of 2 minutes. The pressed simulated minerals are then sent to an intelligent gangue separator to extract the image feature values ​​and thickness features of the simulated mixed minerals.

[0063] The fifth step involves dividing the extracted image features into two parts and inputting them into a support vector machine (SVM) prediction model. The SVM uses 80% of the data as a training set and 20% as a test set to build an image prediction model and generate a binary sorting threshold. The test set shows a prediction accuracy of 95.71%, proving the accuracy of the model. This yields an image prediction model for the minerals in the mining area. The model and the binary sorting threshold are then input into the central database of the intelligent gangue sorting machine to prepare for sorting operations.

[0064] Step 6: The ore from the mining area is crushed to 10-70mm, specifically 10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm. It is then conveyed to an intelligent gangue separator via a conveyor belt. The X-ray identification device of the intelligent gangue separator extracts the grayscale characteristics of the minerals, and the laser sensor extracts the thickness characteristics. The data is transmitted to the central processing unit. The central processing unit uses an image prediction model to classify the minerals into concentrate (ilmenite) and tailings (feldspar, quartz). The identification results are transmitted to the injection system. The injection system adopts different injection measures for the concentrate and tailings. Finally, the concentrate is collected in the concentrate bin and the tailings are collected in the tailings bin.

[0065] Physical property testing of the raw ore and concentrate revealed that the TiO2 grade in the raw ore was 1.87% and the TFe grade was 8.53%, while the TiO2 grade in the concentrate after pre-enrichment was 27.82% and the TFe grade was 58.53%, with a concentrate recovery rate of over 70%.

[0066] Example 3

[0067] This embodiment focuses on the pre-enrichment of strategic metal lithium, and the specific process is as follows:

[0068] The first step was to determine the research area as a mining area in Ganzi, Sichuan Province, with spodumene as the strategic metal mineral and lithium as the main strategic metal to be separated. Physicochemical properties of the ore from the mining area were tested to clarify that its main mineral components include feldspar, spodumene, and quartz.

[0069] The second step involves procuring raw ores such as feldspar, spodumene, quartz, mica, and beryl. These raw ores are then cut and processed into ores of regular thickness, with thicknesses set between 10-70 mm (10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm respectively). An intelligent gangue separator is used to inspect the ores. The intelligent dry separator extracts the image feature values ​​of the raw ore, namely ore thickness and grayscale. The ore thickness feature is obtained by a laser sensor, while the ore grayscale feature is obtained by processing with an X-ray excitation / receiving device. Based on the ore thickness and grayscale features, the ore feature data is fitted to obtain the distribution law of different ore feature values. The study found that under the same thickness conditions, spodumene has the lowest image grayscale value, while other minerals have relatively high grayscale values. The difference in feature values ​​between spodumene and other minerals meets the sorting standards of the intelligent gangue separator.

[0070] The third step involves using an intelligent gangue separator to identify and detect the regular, thickness-controllable strategic metal mineral deposits in both the present and non-present phases obtained from the pressing and sample preparation. The X-ray sensor and receiver of the intelligent gangue separator can automatically extract the grayscale features of the detected minerals and output the grayscale feature results. The built-in laser sensor of the intelligent gangue separator reads the mineral thickness. Based on the mineral thickness and grayscale features detected by the intelligent gangue separator, a thickness-grayscale feature curve of the present / non-present phases of strategic metal minerals is plotted. The curve is fitted, and a single mineral thickness-grayscale feature identification library for the present / non-present phases is established.

[0071] The fourth step involves grinding the raw ores, such as feldspar, spodumene, quartz mica, and beryl, into powder with a particle size controlled between 0.125 and 0.5 mm. Particle sizes that are too large or too small will affect the sample pressing results and cause errors. Adjustments are made based on the average content of each mineral in the coal gangue from the mining area. An infrared pressing sample preparation machine is used to press mixed minerals with different contents to simulate the mineral composition of actual ilmenite. Different thicknesses are adjusted. The infrared pressing sample preparation machine is kept at 50 tons for 2 minutes. The pressed simulated minerals are then sent to an intelligent gangue separator to extract the image feature values ​​and thickness features of the simulated mixed minerals.

[0072] The fifth step involves dividing the extracted image features into two parts and inputting them into a support vector machine (SVM) prediction model. The SVM uses 80% of the data as a training set and 20% as a test set to build an image prediction model and generate a binary sorting threshold. The test set shows a prediction accuracy of 94.62%, proving the accuracy of the model. This yields an image prediction model for the minerals in the mining area. The model and the binary sorting threshold are then input into the central database of the intelligent gangue sorting machine to prepare for sorting operations.

[0073] Step 6: The ore from the mining area is crushed to 10-70mm, specifically 10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm. It is then conveyed to an intelligent gangue separator via a conveyor belt. The X-ray identification device of the intelligent gangue separator extracts the grayscale characteristics of the minerals, and the laser sensor extracts the thickness characteristics. The data is transmitted to the central processing unit. The central processing unit uses an image prediction model to classify the minerals into concentrate (spodumene) and tailings (feldspar, quartz). The identification results are transmitted to the injection system. The injection system adopts different injection measures for the concentrate and tailings. Finally, the concentrate is collected in the concentrate tank and the tailings are collected in the tailings tank.

[0074] Physical property testing of the raw ore and concentrate revealed that the lithium content in the raw ore was 1.65%, while the lithium content in the concentrate after pre-enrichment was 8.7%, with a concentrate recovery rate of over 65%.

[0075] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for enriching strategic metallic elements based on the X-ray response characteristics of minerals, characterized in that, Includes the following steps: Step 1: Determine the research area based on the target mining area and clarify the material composition and physicochemical properties of strategic metal minerals; Step 2: Prepare a regular test sample with controllable thickness; Step 3: Identify and detect the test samples, and establish a single mineral thickness-grayscale feature identification library based on the detection results; In step 3, an intelligent gangue separator is used to identify and detect the regular and thickness-controllable strategic metal mineral deposits in the presence / non-presence phases obtained from the pressing and sampling process. The X-ray sensor and receiver of the intelligent gangue separator can automatically extract the grayscale features of the detected minerals and output the grayscale feature results. The laser sensor built into the intelligent gangue separator reads the mineral thickness and obtains the mineral thickness features and mineral grayscale features. Based on the mineral thickness and grayscale characteristics detected by the intelligent gangue separator, thickness-grayscale characteristic curves of the occurrence phase / non-occurrence phase of strategic metal minerals were plotted, the curves were fitted, and a single mineral thickness-grayscale characteristic identification library for occurrence phase / non-occurrence phase was established. Step 4: Prepare mixed samples and test them. Based on the test results, establish a thickness-grayscale feature identification library for mixed minerals with and without extrinsic phases. In step 4, the prepared mixed sample is tested using an intelligent gangue sorting machine, the thickness and grayscale of the mineral image feature values ​​are extracted, the thickness-grayscale feature curve of the mixed mineral of the strategic metal mineral occurrence phase / non-occurrence phase is plotted, the curve is fitted, and a thickness-grayscale feature recognition library of the mixed mineral of occurrence phase / non-occurrence phase is established. Step 5: Use Python to build a support vector machine model, use the thickness-grayscale features of mixed minerals as the training set input of the model, generate the sorting threshold of the intelligent gangue sorting machine, and complete the establishment of the gangue prediction model; A support vector machine model is built using Python. The thickness-grayscale features of the mixed minerals are used as the training set input for the model. Based on the training set data, the support vector machine model can generate the sorting threshold of the intelligent gangue sorting machine, complete the establishment of the gangue prediction model, divide the minerals into two parts, and clarify which part is the concentrate and which part is the tailings based on the image features of a single mineral, and then carry out the subsequent gangue sorting work. Step 6: Based on the sorting threshold, sort the actual minerals in the study area to obtain concentrate and tailings. The concentrate is the occurrence phase of strategic metals, thus completing the pre-enrichment of strategic metals.

2. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 1, characterized in that, In step 1, after the study area is determined, random samples of ore from the study area are taken for physicochemical analysis to clarify the material composition and physicochemical properties of strategic metal minerals.

3. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 1, characterized in that, In step 2, research samples are collected, and the strategic metal mineral composition and physicochemical properties are analyzed to determine the strategic metal occurrence / non-occurrence phase category.

4. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 3, characterized in that, In step 2, single minerals corresponding to the occurrence phase / non-occurrence phase are procured based on the composition of strategic metal mineral materials.

5. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 4, characterized in that, In step 2, the purchased single minerals are blended according to the actual composition ratio of strategic metal minerals, and then pressed into tablets to obtain regular test samples with controllable thickness.

6. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 1, characterized in that, In step 2, the strategic metal-bearing / non-bearing phase minerals obtained from procurement include both block and powder forms. The block strategic metal-bearing / non-bearing phase minerals for identification and detection are pre-processed and cut into regular samples of fixed thickness.

7. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 6, characterized in that, In step 2, the powdered strategic metal-bearing / non-bearing minerals for identification and detection are pressed into regular samples of fixed thickness using an infrared tablet press.

8. The method for enriching strategic metallic elements based on the X-ray response characteristics of minerals according to claim 7, characterized in that, In step 2, each mineral is pressed to a fixed thickness of 10mm, 15mm, 20mm, 25mm, 30mm, 35mm, 40mm, 45mm, 50mm, 55mm, 60mm, 65mm, and 70mm.

Images