Method and system for detecting total iron content of iron ore based on spectral analysis

Abstract

The invention relates to the technical field of iron content detection, in particular to an iron ore total iron content detection method and system based on spectral analysis, and the method comprises the steps that after spectral data of iron ore are obtained, an improved envelope line method is adopted for preprocessing the spectral data, determining the local maximum value point of the spectral data by adopting a sliding window method, specifically, determining the size of a window for searching the local maximum value point of the spectral data by adopting the sliding window method according to the number of data points of a maximum characteristic peak of a spectral curve of the spectral data, so that the robustness of determining the local maximum value point of the spectral data is improved; when the envelope line node is determined, not only is the local maximum value considered, but also the intersection point of the fitting curve of the spectrum curve is considered, and the feature points in the two aspects are taken as the envelope line node, so that the envelope line node is determined more accurately, and the characteristics of the spectrum curve can be reflected better.

Description

Technical Field

[0001] The present invention relates to the technical field of iron content detection, and in particular to a method and system for detecting the total iron content of iron ore based on spectral analysis. Background Art

[0002] Iron ore is an important mineral resource in modern society. It is the earliest and most widely used metal mineral resource discovered by humans. In order to realize the mutual integration of mines and intelligent technology and promote the development of intelligent mines, quickly and accurately obtaining the total iron content of iron ore has become the key to realizing intelligent geological exploration and accelerating the construction of intelligent mines. Traditionally, the information on the total iron content of iron ore is mainly obtained through a combination of on-site sampling and indoor testing. There are many methods for testing iron ore indoors, including chemical analysis, electrochemical analysis, chromatography proton method, mass spectrometry, neutron activation analysis, phase analysis, photometry and atomic spectroscopy; although there are many traditional methods for iron ore determination, different methods have their specific limitations, such as low sampling density of iron ore samples, sparse measurement points, large workload, long cycle, etc. Therefore, the traditional iron ore testing method cannot meet the requirements of modern intelligent mine construction.

[0003] There are technical solutions for detecting the total iron content of iron ore based on spectral data in the prior art. For example, a Chinese invention patent (CN109030388A) discloses a method for detecting the total iron content of iron ore based on spectral data, comprising the following steps: obtaining spectral data of an iron ore sample to be detected, wherein the spectral data includes m spectral features; inputting the spectral data into an iron ore classification model to obtain the iron ore type of the iron ore sample to be detected; according to the obtained iron ore type, inputting the spectral data into an iron ore total iron content detection model of the corresponding iron ore type to obtain the total iron content of the iron ore corresponding to the spectral data. However, the above scheme does not perform preprocessing operations on the spectral data, resulting in low accuracy in iron content detection. At the same time, there is a scheme for preprocessing spectral data using an improved envelope method in the prior art, which uses the first-order result of the spectral data to determine the envelope nodes. However, the above method has the problems of poor applicability to different spectral curves and inaccurate determination of the envelope nodes, resulting in poor preprocessing effect. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a method and system for detecting the total iron content of iron ore based on spectral analysis, which are used to solve the problems existing in the prior art.

[0005] The present invention provides a method for detecting the total iron content of iron ore based on spectral analysis, comprising the following steps: S1: performing pretreatment operation on the iron ore sample to obtain iron ore powder; S2: collecting spectral data of the iron ore powder; S3: performing data preprocessing operation on the spectral data; Wherein, the preprocessing operation on the spectral data is to perform data preprocessing operation on the spectral data using an improved envelope method; The preprocessing operation for the spectral data is to use the improved envelope method to perform data preprocessing operation on the spectral data, specifically: S3.1: determining a local maximum point of the spectral data in the spectral data, and using the local maximum point as a marking point; S3.2: performing curve fitting on the spectral curve formed by the spectral data, and taking the intersection of the fitting curve and the spectral curve as a marking point; S3.3: De-duplication processing is performed on the marking points determined in S3.1 and S3.2 to obtain final marking points; S3.4: Determine the envelope node of the spectral data according to the final marking point; S3.5: Determine the envelope of the spectral data according to the envelope node; S3.6: performing a preprocessing operation on the spectral data according to the envelope; S4: performing a characteristic band extraction operation on the spectral data preprocessed by S3; S5: Inputting the spectral data of the characteristic band into a total iron content prediction model to obtain a total iron content detection result of the iron ore.

[0006] Preferably, in S3.1, a sliding window method is used to determine the local maximum point of the spectral data; Preferably, the local maximum point of the spectral data is determined by a sliding window method as follows: S3.1.1: Convert the spectral data into a spectral curve; S3.1.2: Determine the number of data points of the maximum characteristic peak of the spectral curve; S3.1.3: Determine the window size of the sliding window according to the number of data points of the maximum characteristic peak of the spectral curve; The window size s of the sliding window is determined by: ; Wherein, m is the number of data points of the maximum characteristic peak of the spectral curve, a is the adjustment coefficient, 0<a<1, and round() is the rounding function; S3.1.4: Determine the local maximum point of the spectral data according to the window determined in S3.1.3.

[0007] Preferably, in S3.1.3, the value of a is related to the characteristic quantity of the spectral curve, and the characteristic quantity includes a characteristic peak, a reflection peak, and an inflection point of the spectral curve.

[0008] Preferably, the value formula of a is: ; Wherein, b is the characteristic number of the spectral curve.

[0009] Preferably, in S3.1.1, the spectral data is converted into a spectral curve by ASD ViewSpec Pro software.

[0010] Preferably, in S3.1.2, the starting point and the end point of the maximum characteristic peak of the spectral curve are marked in the ASD ViewSpec Pro software, and then the statistical function of the ASD ViewSpec Pro software is used to determine the number of data points of the maximum characteristic peak of the spectral curve.

[0011] Preferably, in S3.1.4, the first-order derivative of the spectral curve of the spectral data in the window is calculated, and the zero-crossing point of the first-order derivative of the spectral curve is taken as the local maximum point of the spectral data in the window.

[0012] Preferably, S3.4 is specifically as follows: if the line connecting adjacent final marking points has no intersection with the spectral curve of the spectral data, then the two final marking points are both envelope nodes; if the line connecting adjacent final marking points has an intersection with the spectral curve of the spectral data, then a tangent to the spectral curve of the spectral data is made from the lower final marking point to the higher final marking point, wherein the intersection point of the tangent with the spectral curve of the spectral data is the tangent point, and the lower final marking point and the tangent point are used as envelope nodes.

[0013] According to another aspect of the present invention, a system for detecting the total iron content of iron ore based on spectral analysis is provided. The system adopts the above-mentioned method for detecting the total iron content of iron ore based on spectral analysis. The system comprises: The iron ore pretreatment module is used to perform pretreatment operations on the iron ore sample to obtain iron ore powder; A spectral data acquisition module, used for acquiring spectral data of the iron ore powder; A data preprocessing module, used for performing data preprocessing operations on the spectral data; A characteristic band extraction module, used for performing a characteristic band extraction operation on the spectral data passed through the data preprocessing module; The iron ore total iron content detection module is used to input the spectral data of the characteristic band into the total iron content prediction model to obtain the iron ore total iron content detection result.

[0014] The embodiments of the present invention have the following technical effects: After obtaining the spectrum data of iron ore, the present invention adopts an improved envelope method to pre-process the spectrum data, wherein a sliding window method is adopted to determine the local maximum point of the spectrum data, specifically, the size of the window for finding the local maximum point of the spectrum data by the sliding window method is determined by the number of data points of the maximum characteristic peak of the spectrum curve of the spectrum data, thereby improving the robustness of determining the local maximum point of the spectrum data; When determining the envelope nodes, not only the local maximum values ​​but also the intersection points of the fitting curves of the spectral curves are considered, and the characteristic points of both aspects are used as envelope nodes, so that the envelope nodes are determined more accurately and can better reflect the characteristics of the spectral curve. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 It is a flow chart of a method for detecting the total iron content of iron ore based on spectral analysis provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0018] Embodiment 1, attached Figure 1 A flow chart of a method for detecting the total iron content of iron ore based on spectral analysis is shown in the attached Figure 1 As shown, a method for detecting the total iron content of iron ore based on spectral analysis comprises the following steps: S1: performing pretreatment operation on the iron ore sample to obtain iron ore powder; The collected iron ore samples were sealed with black ziplock bags and then taken back to the laboratory. After the iron ore samples were brought back to the laboratory, they were air-dried, ground, and sieved through a nylon sieve with a mesh size of 100 to obtain iron ore powder.

[0019] S2: collecting spectral data of the iron ore powder; In this step, the ASD FieldSpec4 ground object spectrometer is used to collect spectral data in a dark room in the laboratory. The ASD FieldSpec4 ground object spectrometer is specially used for spectral measurement and can measure the reflectance, transmittance, radiometry and irradiance spectra of ground objects. The device is known for its high signal-to-noise ratio, fast scanning capability and wide wavelength coverage (350 nm - 2500 nm). The spectral range of the ground object spectrometer is 350-2500 nm, the spectral resolution is 3 nm in the range of 350-1000 nm, and is 6 nm in the range of 1000-2500 nm. During measurement, the iron ore powder is spread on a 10 cm*10 cm black non-reflective cardboard, and a reference white board is placed horizontally. The end of the probe optical fiber of the ground object spectrometer is located directly above the sample. In this step, a contact measurement method is adopted.

[0020] In the process of spectrum data collection, in order to ensure the reliability of spectrum data, the iron ore powder is repeatedly collected 5 times. When the reflectance difference of repeated collection is within 5%, the spectrum data is considered qualified, and then the average value of the qualified spectrum data is taken as the spectrum data of the iron ore powder.

[0021] S3: performing data preprocessing operation on the spectral data; Wherein, the preprocessing operation on the spectral data is to perform data preprocessing operation on the spectral data using an improved envelope method; The envelope method is a commonly used spectral data preprocessing method, which is mainly used to highlight the absorption and reflection characteristics of spectral data and normalize them to a consistent spectral background for subsequent analysis and comparison. The core idea of ​​the envelope method is to fit the "shell" (i.e., envelope) of the spectral data, and then perform a ratio operation between the reflectance value of the original spectral data and the reflectance value on the envelope to obtain normalized spectral data. This method can effectively eliminate background noise and non-target information, strengthen the absorption characteristics, and make the morphological characteristics of the spectral data more obvious; although the envelope method can make the absorption characteristics of the spectral data more obvious, for absorption characteristics that are not particularly obvious, there may be a situation where the characteristics are lost. Therefore, this embodiment improves the traditional envelope method in the data preprocessing link; Specifically, the preprocessing operation on the spectral data is to use the improved envelope method to perform data preprocessing on the spectral data. Specifically, the preprocessing operation is: S3.1: determining a local maximum point of the spectral data in the spectral data, and using the local maximum point as a marking point; Wherein, in this step, a sliding window method is used to determine the local maximum point of the spectral data; Specifically, the local maximum point of the spectral data is determined by the sliding window method as follows: S3.1.1: Convert the spectral data into a spectral curve; In this step, the spectral data is converted into a spectral curve by ASD ViewSpec Pro software; S3.1.2: Determine the number of data points of the maximum characteristic peak of the spectral curve; In the step, the starting point and the end point of the maximum characteristic peak of the spectrum curve are marked in the ASD ViewSpec Pro software, and then the number of data points of the maximum characteristic peak of the spectrum curve is determined using the statistical function of the ASD ViewSpec Pro software; S3.1.3: Determine the window size of the sliding window according to the number of data points of the maximum characteristic peak of the spectral curve; The window size s of the sliding window is determined by: ; Wherein, m is the number of data points of the maximum characteristic peak of the spectral curve, a is the adjustment coefficient, 0<a<1, and round() is the rounding function; The value of a is related to the number of features of the spectral curve, which includes a characteristic peak, a reflection peak, and an inflection point of the spectral curve; As a preferred embodiment, the value formula of a is: ; Wherein, b is the characteristic number of the spectral curve; S3.1.4: Determine the local maximum point of the spectral data according to the window determined in S3.1.3; Wherein, the first-order derivative of the spectrum curve of the spectrum data in the window is calculated, and the zero-crossing point of the first-order derivative of the spectrum curve is taken as the local maximum point of the spectrum data in the window; In step S3.1, the size of the window for finding the local maximum point of the spectral data by the sliding window method is determined by the number of data points of the maximum characteristic peak of the spectral curve of the spectral data, thereby improving the robustness of determining the local maximum point of the spectral data; S3.2: performing curve fitting on the spectral curve formed by the spectral data, and taking the intersection of the fitting curve and the spectral curve as a marking point; Wherein, in this step, the spectrum curve is fitted by using the Lorentz fitting method; S3.3: De-duplication processing is performed on the marking points determined in S3.1 and S3.2 to obtain final marking points; S3.4: Determine the envelope node of the spectral data according to the final marking point; In this step, S3.4 is specifically as follows: if the line connecting the adjacent final marking points has no intersection with the spectral curve of the spectral data, then the two final marking points are both envelope nodes; if the line connecting the adjacent final marking points has an intersection with the spectral curve of the spectral data, then a tangent line of the spectral curve of the spectral data is drawn from the lower final marking point to the higher final marking point, wherein the intersection point of the tangent line with the spectral curve of the spectral data is the tangent point, and the lower final marking point and the tangent point are taken as envelope nodes; In this embodiment, when determining the envelope nodes, not only the local maximum values ​​but also the intersection points of the fitting curves of the spectral curves are considered, and the characteristic points of both aspects are used as envelope nodes, so that the envelope nodes are determined more accurately and can better reflect the characteristics of the spectral curve.

[0022] S3.5: Determine the envelope of the spectral data according to the envelope node; S3.6: performing a preprocessing operation on the spectral data according to the envelope; In this step, the spectrum data is divided by the envelope to achieve a preprocessing operation.

[0023] S4: performing a characteristic band extraction operation on the spectral data preprocessed by S3; The mineral composition of iron ore is complex and contains many impurities. During the detection of total iron content, the spectral information of various impurities will overlap, which will cause a lot of noise in the detected spectral data, thus having a great impact on the detection accuracy. Therefore, it is necessary to extract the characteristic bands related to the properties of the iron ore to be detected to improve the detection accuracy; In this step, the characteristic bands of the spectral data are extracted by using correlation analysis method; Correlation analysis is a commonly used method for extracting spectral characteristic bands. It calculates the correlation between spectral data and target variables (characteristic bands) and selects the bands most relevant to the target variables, thereby realizing the extraction of characteristic bands. The correlation analysis method is used to perform correlation analysis on the spectral data and the iron ore attribute information band by band, the correlation coefficient between each band in the spectral data and the iron ore attribute data is calculated respectively, and the band with a larger absolute value of the correlation coefficient is used as the characteristic band of the spectral data.

[0024] S5: inputting the spectral data of the characteristic band into a total iron content prediction model to obtain a total iron content detection result of the iron ore; Wherein, the total iron content detection model is a deep learning model, and the deep learning model is one of an artificial neural network model, a deep neural network model, and a recurrent neural network model; The input of the deep learning model is the spectral data of the characteristic band, and the output is the total iron content detection value.

[0025] Embodiment 2, the present invention further provides a system for detecting the total iron content of iron ore based on spectral analysis, the system adopts a method for detecting the total iron content of iron ore based on spectral analysis in embodiment 1, and the system comprises: The iron ore pretreatment module is used to perform pretreatment operations on the iron ore sample to obtain iron ore powder; A spectral data acquisition module, used for acquiring spectral data of the iron ore powder; A data preprocessing module, used for performing data preprocessing operations on the spectral data; A characteristic band extraction module, used for performing a characteristic band extraction operation on the spectral data passed through the data preprocessing module; The iron ore total iron content detection module is used to input the spectral data of the characteristic band into the total iron content prediction model to obtain the iron ore total iron content detection result.

[0026] Embodiment 3, the present invention also provides an electronic device, including one or more processors and a memory.

[0027] The processor may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0028] The memory may include one or more computer program products, and the computer program product may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may run the program instructions to implement a method for detecting the total iron content of iron ore based on spectral analysis and / or other desired functions of any embodiment of the present application described above. Various contents such as initial external parameters, thresholds, etc. may also be stored in the computer-readable storage medium.

[0029] In one example, the electronic device may further include: an input device and an output device, and these components are interconnected through a bus system and / or other forms of connection mechanisms (not shown). The input device may include, for example, a keyboard, a mouse, etc. The output device may output various information to the outside, including early warning information, braking force, etc. The output device may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected thereto, etc.

[0030] Of course, for the sake of simplicity, components such as buses, input / output interfaces, etc. are omitted. In addition, the electronic device may further include any other appropriate components according to specific application conditions.

[0031] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to implement the functions of a method for detecting the total iron content of iron ore based on spectral analysis provided in any embodiment of the present application.

[0032] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user computing device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0033] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor implements a method for detecting the total iron content of iron ore based on spectral analysis provided in any embodiment of the present application.

[0034] The computer readable storage medium can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0035] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the total iron content of iron ore based on spectral analysis, characterized in that: The following steps are involved: S1: performing pretreatment operation on the iron ore sample to obtain iron ore powder; S2: collecting spectral data of the iron ore powder; S3: performing data preprocessing operation on the spectral data; Wherein, the data preprocessing operation on the spectral data is to perform data preprocessing operation on the spectral data using an improved envelope method; The preprocessing operation for the spectral data is to use an improved envelope method to perform data preprocessing operation on the spectral data, specifically: S3.1: determining a local maximum point of the spectral data in the spectral data, and using the local maximum point as a marking point; S3.2: performing curve fitting on the spectral curve formed by the spectral data, and taking the intersection of the fitting curve and the spectral curve as a marking point; S3.3: Deduplication processing is performed on the marking points determined in S3.1 and S3.2 to obtain final marking points; S3.4: Determine the envelope node of the spectral data according to the final marking point; S3.5: Determine the envelope of the spectral data according to the envelope node; S3.6: performing a preprocessing operation on the spectral data according to the envelope; S4: performing a characteristic band extraction operation on the spectral data preprocessed by the S3 data; S5: Input the spectral data of the characteristic band into the total iron content prediction model to obtain the total iron content detection result of the iron ore.

2. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 1, characterized in that: In S3.1, a sliding window method is used to determine the local maximum point of the spectral data.

3. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 2, characterized in that: The local maximum point of the spectral data is determined by the sliding window method as follows: S3.1.1: Convert the spectral data into a spectral curve; S3.1.2: Determine the number of data points of the maximum characteristic peak of the spectral curve; S3.1.3: Determine the window size of the sliding window according to the number of data points of the maximum characteristic peak of the spectral curve; The window size s of the sliding window is determined by: ； Wherein, m is the number of data points of the maximum characteristic peak of the spectral curve, a is the adjustment coefficient, 0<a<1, and round() is the rounding function; S3.1.4: Determine the local maximum point of the spectral data according to the window size determined in S3.1.

3.

4. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 3, characterized in that: In S3.1.3, the value of a is related to the number of features of the spectral curve, and the number of features includes a characteristic peak, a reflection peak, and an inflection point of the spectral curve.

5. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 4, characterized in that: The value formula of a is: ； Wherein, b is the characteristic number of the spectral curve.

6. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 3, characterized in that: In S3.1.1, the spectral data is converted into a spectral curve using ASD ViewSpec Pro software.

7. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 6, characterized in that: In S3.1.2, the starting point and the end point of the maximum characteristic peak of the spectral curve are marked in the ASD ViewSpec Pro software, and then the statistical function of the ASD ViewSpec Pro software is used to determine the number of data points of the maximum characteristic peak of the spectral curve.

8. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 3, characterized in that: In S3.1.4, the first-order derivative of the spectrum curve of the spectrum data in the window is calculated, and the zero-crossing point of the first-order derivative of the spectrum curve is taken as the local maximum point of the spectrum data in the window.

9. The method for detecting the total iron content of iron ore based on spectral analysis according to claim 1, characterized in that: Specifically, S3.4 is as follows: if the line connecting adjacent final marking points has no intersection with the spectral curve of the spectral data, then the two final marking points are both envelope nodes; if the line connecting adjacent final marking points has an intersection with the spectral curve of the spectral data, then a tangent to the spectral curve of the spectral data is drawn from the lower final marking point to the higher final marking point, wherein the intersection point of the tangent with the spectral curve of the spectral data is the tangent point, and the lower final marking point and the tangent point are used as envelope nodes.

10. A system for detecting the total iron content of iron ore based on spectral analysis, characterized in that: The system adopts a method for detecting the total iron content of iron ore based on spectral analysis as described in any one of claims 1 to 9, and the system comprises: The iron ore pretreatment module is used to perform pretreatment operations on the iron ore sample to obtain iron ore powder; A spectral data acquisition module, used for acquiring spectral data of the iron ore powder; A data preprocessing module, used for performing data preprocessing operations on the spectral data; A characteristic band extraction module, used for performing a characteristic band extraction operation on the spectral data passed through the data preprocessing module; The iron ore total iron content detection module is used to input the spectral data of the characteristic band into the total iron content prediction model to obtain the iron ore total iron content detection result.

Images