Raw coal effective spectrum screening method and system based on CN and C2 characteristic peaks

Abstract

The invention provides a raw coal effective spectrum screening method and system based on CN and C2 characteristic peaks, and the method comprises the steps: obtaining standard spectrums of a first wave band where the CN characteristic peak is located and a second wave band where the C2 characteristic peak is located based on a standard coal sample, and constructing a standard spectrum characteristic matrix; collecting spectral data of a to-be-detected raw coal sample, extracting a to-be-compared spectrum with the same wave band as the standard spectrum, and constructing a to-be-compared spectrum characteristic matrix; and calculating a correlation coefficient between the standard spectral feature matrix and the spectral feature matrix to be compared, comparing the correlation coefficient with a preset threshold value, and screening an effective spectrum according to a comparison result. The invention provides a method for optimizing and screening effective spectral data of raw coal based on CN and C2 characteristic peaks, which is suitable for scenes of laser-induced breakdown spectroscopy (LIBS) on-line detection of coal quality industrial indexes and element contents, and aims to improve the effectiveness and representativeness of LIBS spectral data so as to improve the repeatability and accuracy of coal structure analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of spectral analysis, and in particular to a raw coal effective spectrum screening method and system based on CN and C2 characteristic peaks. BACKGROUND

[0002] Coal, as a major energy source, its organic component structure directly determines the combustion efficiency, pollution emission and high value-added utilization potential. Laser-induced breakdown spectroscopy (LIBS) technology has become an important means of coal structure analysis due to its advantages of rapidity and non-destructiveness. In the process of LIBS online detection, a large amount of gangue, metal minerals and other non-coal substances are mixed in the raw coal blocks, resulting in the collected spectrum containing the characteristic spectrum of gangue and metal substances during detection, affecting the detection accuracy. Spectrum data optimization screening can identify abnormal signals caused by sample abnormalities, changes in experimental environment or other errors through the fluctuation characteristics of spectral intensity, and quickly identify effective spectra and eliminate invalid spectra in data preprocessing. Coal contains a large amount of carbon elements (C), and after laser-induced breakdown, the excited carbon elements (C) and excited state nitrogen elements (N) rapidly combine to form free radicals CN in the air environment, and its unique multi-peak characteristics are significantly different from the spectra of gangue and metal minerals. The C2 molecular radiation characteristic peak is mainly derived from the sample itself, and the functional group breakage is the main mechanism for the formation of C2 molecular characteristic peaks. Different coal samples all have C2 molecular radiation in their spectra. The characteristics of CN and C2 characteristic peaks are obviously different from those of gangue and metal minerals, which can be used to accurately identify invalid spectra, so as to eliminate abnormal data and improve the effectiveness and representativeness of spectral data. The effectiveness and representativeness of the detection data are of great significance to improve the repeatability and accuracy of LIBS online detection of industrial indicators and element content, and to guide the mixing of coal and coal blending in the plant.

[0003] The traditional coal quality spectrum data optimization screening method mainly includes outlier detection, background removal, normalization, etc. The effective data can be screened by comparing the ratio of the intensity of the characteristic spectral line to the intensity of the adjacent non-characteristic continuous spectral region, and when the ratio is higher than the set threshold, it is determined as effective data, which can screen abnormal spectra, but cannot distinguish between effective and invalid spectra. In order to improve the detection accuracy and repeatability, it is urgent to develop a systematic effective spectrum data optimization screening method based on physics. SUMMARY

[0004] The purpose of the present application is to provide a raw coal effective spectrum screening method and system based on CN and C2 characteristic peaks, which aims to solve the above problems in the prior art.

[0005] The present application provides a raw coal effective spectrum screening method based on CN and C2 characteristic peaks, comprising: The standard spectrum feature matrix is constructed based on the standard spectrum of the first wave band where the CN characteristic peak is located and the second wave band where the C2 characteristic peak is located. Spectrum data of the to-be-tested raw coal sample is collected, a to-be-compared spectrum of the same wave band as the standard spectrum is extracted, and a to-be-compared spectrum feature matrix is constructed. The correlation coefficient of the standard spectrum feature matrix and the to-be-compared spectrum feature matrix is calculated, the correlation coefficient is compared with a preset threshold, and effective spectrum is screened according to a comparison result.

[0006] The embodiment of the present application provides a raw coal effective spectrum screening system based on CN and C2 characteristic peaks, which comprises: The standard spectrum construction module is configured to construct a standard spectrum feature matrix based on the standard spectrum of the first wave band where the CN characteristic peak is located and the second wave band where the C2 characteristic peak is located. The to-be-tested spectrum processing module is configured to collect spectrum data of the to-be-tested raw coal sample, extract a to-be-compared spectrum of the same wave band as the standard spectrum, and construct a to-be-compared spectrum feature matrix. The effective spectrum screening module is configured to calculate the correlation coefficient of the standard spectrum feature matrix and the to-be-compared spectrum feature matrix, compare the correlation coefficient with a preset threshold, and screen effective spectrum according to a comparison result.

[0007] The embodiment of the present application further provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is executed by the processor to implement the steps of the raw coal effective spectrum screening method based on CN and C2 characteristic peaks.

[0008] The embodiment of the present application further provides a computer readable storage medium, and the computer readable storage medium stores an implementation program of information transmission, and the program is executed by a processor to implement the steps of the raw coal effective spectrum screening method based on CN and C2 characteristic peaks.

[0009] The embodiment of the present application can have the following beneficial effects: the embodiment of the present application proposes a method for optimizing and screening raw coal effective spectrum data based on CN and C2 characteristic peaks, which is suitable for the scene of laser-induced breakdown spectroscopy (LIBS) online detection of coal quality industrial indexes and element content, and aims to improve the effectiveness and representativeness of LIBS spectrum data, thereby improving the repeatability and accuracy of coal structure analysis. BRIEF DESCRIPTION OF DRAWINGS

[0010] In order to make one or more embodiments of the present specification or the prior art clearer, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present specification, and other drawings can be obtained by those skilled in the art without creative labor.

[0011] Figure 1 is a method flowchart of the raw coal effective spectrum screening method based on CN and C2 characteristic peaks of the embodiment of the present application. Figure 2 is a method flowchart of the embodiment of the present application. Figure 3 is a 378.27 nm-388.70 nm waveband spectrum comparison schematic diagram of different substances in raw coal of the embodiment of the present application. Figure 4 is a standard spectrum and a to-be-compared spectrum correlation coefficient change and CI 247.856 nm characteristic peak signal intensity and RSD change schematic diagram under different focusing distances of the embodiment of the present application. Figure 5 is a 455.43 nm-565.01 nm waveband spectrum comparison schematic diagram of different focal length positions in raw coal of the embodiment of the present application. Figure 6 is a raw coal effective spectrum screening system schematic diagram based on CN and C2 characteristic peaks of the embodiment of the present application. DETAILED DESCRIPTION

[0012] In order to make one or more embodiments of the present specification or the prior art clearer, the drawings needed in the embodiment or prior art description will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present specification, and other drawings can be obtained by those skilled in the art without creative labor.

[0013] Method embodiment According to the embodiment of the present application, a raw coal effective spectrum screening method based on CN and C2 characteristic peaks is provided. Figure 1 is a method flowchart of the raw coal effective spectrum screening method based on CN and C2 characteristic peaks of the embodiment of the present application, as Figure 1 shown, the raw coal effective spectrum screening method based on CN and C2 characteristic peaks according to the embodiment of the present application specifically includes: In step S101, the standard spectrum feature matrix is constructed based on the standard coal sample to obtain the standard spectrum of the first wave band where the CN characteristic peak is located and the second wave band where the C2 characteristic peak is located, and specifically includes: The standard coal sample is detected by laser-induced breakdown spectroscopy to collect the original spectrum data thereof in the first wave band and the second wave band; The spectrum signals of the CN characteristic peak and the C2 characteristic peak are extracted from the original spectrum data; The extracted spectrum signals are combined to obtain the standard spectrum feature matrix.

[0014] In step S102, the spectrum data of the to-be-detected raw coal sample is collected, the to-be-compared spectrum of the same wave band as the standard spectrum is extracted, and the to-be-compared spectrum feature matrix is constructed; The to-be-compared spectrum includes at least one of a rock spectrum, a metal mineral spectrum, an over-focused spectrum, an off-focused spectrum and a shielding spectrum; The to-be-compared spectrum feature matrix has the same dimension as the standard spectrum feature matrix.

[0015] In step S103, the correlation coefficient of the standard spectrum feature matrix and the to-be-compared spectrum feature matrix is calculated, the correlation coefficient is compared with a preset threshold, and the effective spectrum is screened according to the comparison result.

[0016] The method further includes: The screened spectrum data is subjected to multi-dimensional evaluation; The multi-dimensional evaluation includes at least one evaluation parameter of the relative standard deviation of the spectrum line intensity, the sensitivity, the specificity or the accuracy.

[0017] The above technical solutions of the embodiments of the present application are described in detail in combination with the specific conditions of the raw coal effective spectrum screening method based on the CN and C2 characteristic peaks of the embodiments of the present application.

[0018] The embodiments of the present application propose a raw coal effective spectrum data optimization screening method based on CN and C2 characteristic peaks, utilize pulsed laser to load the coal sample to generate laser-induced breakdown spectroscopy, collect the plasma radiation spectrum data, select the spectrum features of the CN and C2 characteristic peaks as the standard spectrum, construct the spectrum feature matrix, extract the to-be-compared spectrum corresponding to the same wave band from the spectrum collection, construct the corresponding spectrum feature matrix, calculate the correlation coefficient of the standard spectrum and the to-be-compared spectrum , retain the effective spectrum, and eliminate the rest of the spectrum because the CN characteristic peak does not exist or the CN / C2 characteristic peak shape is irregular. The high-precision identification of the coal sample, the gangue and the metal mineral is realized, and the RSD values of the CI 247.856 nm, the Si I 288.165 nm and the Na I 589.024 nm characteristic peaks after optimization screening are low. The method includes the following steps: Step 1: focus the pulsed laser on the surface of the raw coal sample, and excite to generate plasma; Step 2: collect the plasma radiation spectrum through a one-to-five optical fiber and transmit it to a five-channel optical fiber spectrometer to obtain spectral data; Step 3: extract the CN free radical characteristic peak in the 378.27 nm~388.70 nm waveband, and construct a standard spectral feature matrix; Step 4: extract the C2 molecular characteristic peak in the 455.43 nm~565.01 nm waveband, and construct a standard spectral feature matrix; Step 5: construct a spectral feature matrix with the same dimension as the standard spectral feature matrix for the same waveband of the spectrum to be compared; Step 6: calculate the correlation coefficient of the standard spectral feature matrix and the spectral feature matrix to be compared , retain the spectral data greater than 0.9, and eliminate the remaining spectral data; Step 7: calculate the relative standard deviation, sensitivity, specificity and accuracy of the retained spectral data for quantitative analysis modeling.

[0019] Among them, the spectrum to be compared includes rock spectrum, metal spectrum, over-focused spectrum, off-focused spectrum, shielding spectrum and effective spectrum.

[0020] The screened spectral data can be directly used to establish a quantitative analysis model of the industrial index and element content of raw coal.

[0021] That is, the raw coal effective spectrum data optimization screening method based on CN and C2 characteristic peaks proposed in the embodiment of the application uses the same optical path system to realize the collection of raw coal sample spectral data, selects the CN (378.27 nm~388.70 nm) and C2 (455.43 nm~565.01 nm) wavelength range as the analysis characteristic, and constructs a spectral feature matrix ; extract the same waveband of the spectrum to be compared (including rock spectrum, coal spectrum and metal spectrum; over-focused spectrum, off-focused spectrum, shielding spectrum, effective spectrum) from the spectral collection, construct the corresponding spectral feature matrix ; calculate the correlation coefficient of the standard spectrum and the spectrum to be compared according to the formula ; retain >0.9 spectrum, and eliminate the remaining spectrum because there is no CN characteristic peak or the CN / C2 characteristic peak shape is irregular. After identifying and eliminating abnormal data by statistical and physical methods, the normal distribution test of the average spectral intensity of the sample is performed, and the corresponding RSD, S, SP, ACC and other evaluation parameters are calculated to evaluate the representativeness and effectiveness of the data. The spectral data screened by the data can be directly used for quantitative analysis modeling. The embodiment of the application specifically includes the following steps: Step one: to realize fast and effective coal quality spectrum screening, the spectral features of standard coal samples in the wavelength range of 378.27 nm to 388.70 nm are selected as standard spectra by using the characteristic comparison method, and a spectral feature matrix is constructed ; the same wavelength range of the to-be-compared spectra (including rock spectra, coal spectra and metal spectra) is extracted from the spectrum collection, and a corresponding spectral feature matrix is constructed ; the correlation coefficient of the standard spectrum and the to-be-compared spectrum is calculated according to formula (1) ; the spectrum with a correlation coefficient greater than 0.9 is retained, and the rest of the spectra is removed because the CN characteristic peak does not exist or the CN characteristic peak shape is irregular. >0.9.

[0022] (1); In the formula, is the matrix element of the standard spectral feature, is the average value of the standard spectral feature, is the matrix element of the to-be-compared spectral feature, is the average value of the to-be-compared spectral feature matrix, and the spectrum is a coal sample spectrum.

[0023] Step two: to quickly screen the effective focal length area coal quality spectrum, the spectral features of standard coal samples in the wavelength range of 455.43 nm to 565.01 nm are selected as standard spectra by using the characteristic comparison method, and a spectral feature matrix is constructed ; the same wavelength range of the to-be-compared spectra (over-focused spectra, off-focused spectra, blocked spectra and effective spectra) is extracted from the spectrum collection, and a spectral feature matrix is constructed ; the correlation coefficient of the standard spectrum and the to-be-compared spectrum is calculated according to formula (1) ; the spectrum with a correlation coefficient greater than 0.9 is retained, and the rest of the spectra is removed because the C2 characteristic peak shape is irregular, and the spectrum is an effective spectrum. >0.9.

[0024] Step three: the relative standard deviation (RSD) of the spectral line intensity, the sensitivity S (Sensitivity), the specificity SP (Specificity) and the accuracy ACC (Accuracy) are calculated to evaluate the accuracy and effect of data removal.

[0025] The specific detection process of the embodiment of the application is shown in Figure 2 . Step one: select the characteristic peak in the CN (378.27 nm to 388.70 nm) wavelength range, construct a spectral feature matrix ; then, the same wavelength range of the to-be-compared spectra (including rock spectra, coal spectra and metal spectra) is extracted from the spectrum collection, and a corresponding spectral feature matrix is constructed , the correlation coefficient of the standard spectrum and the spectrum to be compared is calculated , and the spectrum with a correlation coefficient >0.9 is reserved. Figure 3 (a) shows the comparison of the gangue spectrum and the standard spectrum. Although the CN free radical characteristic peak exists in the gangue spectrum, the CN characteristic peak is not smooth due to the large amount of impurity elements in the gangue, and is accompanied by multiple strong matrix element peaks. The correlation coefficient is 0.813, and the spectrum is determined as an invalid spectrum; Figure 3 (b) is the comparison of the coal spectrum and the standard spectrum. The shapes of the two are highly consistent, and the correlation coefficient is as high as 0.993, which is determined as a valid spectrum of the coal sample; Figure 3 (c) shows the comparison of the metal spectrum and the standard spectrum. In the waveband of 378.27 nm~388.70 nm, the metal spectrum does not show the CN characteristic peak, but shows multiple metal characteristic peaks. The correlation coefficient is as low as 0.293, which is determined as an invalid spectrum. Through effective coal quality spectrum screening, the finally reserved spectrum has the coal quality characteristic peak as shown in Figure 3 (b).

[0026] Step two: as shown in Figure 4 , based on the characteristics of the C2 molecular peak, the effective focal length spectrum of the coal LIBS spectrum is obviously different from the over-focusing, defocusing and shielding spectrum. The spectral characteristics of the standard coal sample in the C2 (455.43 nm~565.01 nm) waveband are selected as the standard spectrum, and the characteristic spectrum matrix is constructed. The same waveband of the spectrum to be compared (over-focusing spectrum, defocusing spectrum, shielding spectrum and effective spectrum) is extracted from the spectrum collection, and the characteristic spectrum matrix is constructed. The correlation coefficient of the standard spectrum and the spectrum to be compared is calculated according to formula (1) , and the spectrum with a correlation coefficient >0.9 is reserved. Figure 5 (a) shows the comparison of the effective focal length spectrum and the standard spectrum. The shapes of the two are well consistent, and the correlation coefficient is 0.991, which is determined as a valid spectrum of the coal sample; Figure 5 (b) is the comparison of the shielding spectrum and the standard spectrum. Due to the shielding of the stacked raw coal blocks, the laser energy acting on the sample surface is significantly reduced, the C2 molecular peak cannot be effectively excited, and the correlation coefficient is 0.569, which is determined as an invalid spectrum of the coal sample; Figure 5 (c) shows the comparison of the defocusing spectrum and the standard spectrum. Due to the formation of air plasma, part of the pulse laser energy is absorbed, which reduces the laser energy acting on the sample surface, the C2 molecular peak cannot be effectively excited, and the correlation coefficient is 0.759, which is determined as an invalid spectrum of the coal sample; Figure 5(d) For the comparison of the over-focused spectrum and the standard spectrum, the over-focused spot diameter increases, the energy density decreases, and the C2 molecular peak cannot be effectively excited. The correlation coefficient is 0.886, and it is determined that the spectrum of the coal sample is invalid. After the effective focal length spectrum screening, the spectra retained have the coal quality characteristic peaks as shown in Figure 5 (a).

[0027] Step three: the relative standard deviation (RSD) of the spectral line intensity is used to evaluate the fluctuation of the spectral signal. A smaller RSD value indicates that the LIBS measured spectral signal is more stable.

[0028] (2). wherein, is the number of measurements, is the absolute line intensity of the element in each measurement, is the arithmetic mean line intensity of the repeated measurements.

[0029] Sensitivity S (Sensitivity), specificity SP (Specificity) and accuracy ACC (Accuracy) are used to evaluate the accuracy and effect of data rejection. Sensitivity S, also known as the true rate, measures the ability to correctly identify valid data, that is, the proportion of actual valid data that is correctly identified as valid data, that is, the retention rate of valid data, as shown in equation (3): (3). wherein, TP is the true positive, indicating the number of valid data correctly identified; FN is the false negative, indicating the number of actual valid data that is incorrectly rejected. The higher the sensitivity, the higher the proportion of valid data correctly retained in the data screening process. Specificity SP, also known as the true negative rate, measures the ability to correctly identify invalid data. Specificity SP indicates the proportion of actual invalid data that is correctly identified as invalid data, that is, the rejection rate of invalid data, as shown in equation (4): (4). wherein, TN is the true negative, indicating the number of invalid data correctly identified; FP is the false positive, indicating the number of actual invalid data that is incorrectly identified as valid data. The higher the specificity, the higher the proportion of invalid data correctly rejected. Accuracy ACC measures the proportion of correctly screened data in all data. The accuracy can comprehensively reflect the overall effect of data screening, as shown in equation (5): (5). The higher the accuracy, the more accurate the overall screening process in judging valid and invalid data.

[0030] After screening by C2 and CN characteristic peaks, the RSD values of CI 247.856 nm, Si I 288.165 nm and Na I 589.024 nm characteristic peaks corresponding to the characteristic peaks were significantly reduced, and the average values were reduced from 66.53%, 69.93% and 46.38% to 41.26%, 49.13% and 32.88% respectively, as shown in Table 1.

[0031] Table 1 Optimization effect of RSD average value of characteristic peaks according to statistical and physical methods for screening effective spectra

[0032] As shown in Table 2, the specificity, sensitivity and accuracy of the sample are high, the average rejection rate of invalid data is 93%, the average retention rate of effective data is 93%, and the overall accuracy of accurately identifying effective and invalid data during the screening process is 93.7%. Combined with statistical and physical methods for screening effective spectral data, not only the signal stability can be improved, but also 93% of invalid spectra can be rejected and 93% of effective spectra can be retained during the screening process, thereby improving the sample representativeness of the spectrum.

[0033] Table 2 Evaluation parameters of statistical and physical methods for screening 10 groups of samples

[0034] In summary, the embodiment of the present application provides a raw coal effective spectral data screening method based on CN and C2 characteristic peaks suitable for online coal quality detection process, which uses pulsed laser to load coal sample to generate laser-induced breakdown spectroscopy (LIBS), and uses a five-way optical fiber spectrometer to collect spectrum to five channels. The spectral characteristics of the standard coal sample in the waveband of 378.27 nm to 388.70 nm and 455.43 nm to 565.01 nm are selected as the standard spectrum, and a spectral feature matrix is constructed ; the same waveband of the to-be-compared spectrum (including rock spectrum, coal spectrum and metal spectrum; over-focus spectrum, off-focus spectrum, shielding spectrum and effective spectrum) is extracted from the spectrum collection, and a corresponding spectral feature matrix ; the correlation coefficient of the standard spectrum and the to-be-compared spectrum is calculated according to the formula ; the The spectrum of 0.9 is removed, and the rest of the spectrum is removed because there is no CN characteristic peak or the CN / C2 characteristic peak shape is irregular. The embodiment of the present application does not need to take samples and only consumes micrograms-milligrams of coal samples. After screening by C2 and CN characteristic peaks, the RSD values of CI 247.856 nm, Si I 288.165 nm and Na I 589.024 nm characteristic peaks are significantly reduced, and the average values are reduced from 66.53%, 69.93% and 46.38% to 41.26%, 49.13% and 32.88% respectively. This is of great significance to improve the accuracy and effectiveness of online detection of effective spectral data acquisition and quantitative analysis of coal quality LIBS.

[0035] System embodiment According to the embodiment of the present application, an effective spectrum screening system for raw coal based on CN and C2 characteristic peaks is provided, Figure 6 The effective spectrum screening system for raw coal based on CN and C2 characteristic peaks is a schematic diagram of the embodiment of the present application, as Figure 6 The effective spectrum screening system for raw coal based on CN and C2 characteristic peaks according to the embodiment of the present application specifically comprises: The standard spectrum construction module 60 is used to obtain the standard spectrum of the first wave band where the CN characteristic peak is located and the second wave band where the C2 characteristic peak is located based on the standard coal sample, and to construct a standard spectrum feature matrix, specifically for: Performing laser-induced breakdown spectroscopy detection on the standard coal sample, and collecting the original spectrum data of the first wave band and the second wave band; Extracting the spectrum signals of the CN characteristic peak and the C2 characteristic peak from the original spectrum data; Combining the extracted spectrum signals to obtain a standard spectrum feature matrix.

[0036] The to-be-detected spectrum processing module 62 is used to collect the spectrum data of the to-be-detected raw coal sample, extract the to-be-compared spectrum of the same wave band as the standard spectrum, and construct a to-be-compared spectrum feature matrix; The to-be-compared spectrum includes at least one of rock spectrum, metal mineral spectrum, over-focus spectrum, off-focus spectrum and shielding spectrum; The to-be-compared spectrum feature matrix has the same dimension as the standard spectrum feature matrix.

[0037] The effective spectrum screening module 64 is used to calculate the correlation coefficient of the standard spectrum feature matrix and the to-be-compared spectrum feature matrix, and compare the correlation coefficient with a preset threshold, and screen the effective spectrum according to the comparison result.

[0038] The system further comprises: The evaluation module is used to perform multi-dimensional evaluation on the screened spectrum data; The multi-dimension evaluation includes at least one of the evaluation parameters of relative standard deviation of spectral line intensity, sensitivity, specificity or accuracy.

[0039] The embodiment of the present application is a system embodiment corresponding to the above-mentioned method embodiment, and the specific operation of each module can be understood with reference to the description of the method embodiment, which will not be repeated here.

[0040] In summary, the raw coal effective spectrum data optimization screening method based on CN and C2 characteristic peaks proposed in the embodiment of the present application selects CN and C2 characteristic peaks in a specific wave band, constructs a standard spectral feature matrix, and calculates the correlation coefficient with the spectrum to be compared, and retains the spectrum with a correlation coefficient greater than 0.9, effectively eliminating gangue, metallic minerals, overfocus, defocus and shielding spectrum. The abnormal data is identified and eliminated by combining statistical and physical methods, which significantly improves the stability and representativeness of the spectral data. This method not only reduces the volatility of the spectral data, but also improves the data representativeness. The optimized spectral data can be directly used for quantitative analysis modeling, which can significantly improve the analysis accuracy and model robustness. At the same time, the screening effect is evaluated by sensitivity (S), specificity (SP) and accuracy (ACC), to ensure the accuracy and efficiency of the screening process, and to provide an efficient and reliable solution for raw coal component analysis in complex industrial scenarios.

[0041] Device embodiment one The embodiment of the present application provides an electronic device, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the computer program is executed by the processor to realize the steps as described in the method embodiment.

[0042] Device embodiment two The embodiment of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores an implementation program of information transmission, and the program is executed by a processor to realize the steps as described in the method embodiment.

[0043] The computer readable storage medium described in the embodiment includes but is not limited to ROM, RAM, magnetic disk or optical disk, etc.

[0044] Finally, it should be pointed out that: the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for effective spectral screening of raw coal based on CN and C2 characteristic peaks, characterized in that, include: Standard spectra of the CN characteristic peak in the first band and the C2 characteristic peak in the second band were obtained from standard coal samples, and a standard spectral feature matrix was constructed. Collect spectral data of the raw coal sample to be tested, extract the spectrum to be compared in the same band as the standard spectrum, and construct the feature matrix of the spectrum to be compared. Calculate the correlation coefficient between the standard spectral feature matrix and the spectral feature matrix to be compared, and compare the correlation coefficient with a preset threshold. Filter valid spectra based on the comparison results.

2. The method according to claim 1, characterized in that, The method further includes: The selected spectral data were evaluated from multiple dimensions. The multidimensional evaluation includes at least one evaluation parameter among the relative standard deviation, sensitivity, specificity, or accuracy of the spectral line intensity.

3. The method according to claim 1, characterized in that, Based on standard coal samples, standard spectra of the CN characteristic peak in the first band and the C2 characteristic peak in the second band were obtained. The construction of the standard spectral characteristic matrix specifically includes: Laser-induced breakdown spectroscopy was performed on a standard coal sample to collect its original spectral data in the first and second bands; the spectral signals of the CN and C2 characteristic peaks were extracted from the original spectral data; and the extracted spectral signals were combined to obtain a standard spectral feature matrix.

4. The method according to claim 1, characterized in that, The spectra to be compared include at least one of rock spectra, metallic mineral spectra, overfocused spectra, defocused spectra, and obscured spectra; The spectral feature matrix to be compared has the same dimension as the standard spectral feature matrix.

5. A system for effective spectral screening of raw coal based on CN and C2 characteristic peaks, characterized in that, include: The standard spectrum construction module is used to obtain the standard spectra of the CN characteristic peak in the first band and the C2 characteristic peak in the second band based on the standard coal sample, and construct the standard spectrum feature matrix. The test spectrum processing module is used to collect the spectral data of the raw coal sample to be tested, extract the comparison spectrum in the same band as the standard spectrum, and construct the comparison spectrum feature matrix. The effective spectrum screening module is used to calculate the correlation coefficient between the standard spectral feature matrix and the spectral feature matrix to be compared, and compare the correlation coefficient with a preset threshold to screen effective spectra based on the comparison results.

6. The system according to claim 5, characterized in that, The system further includes: The evaluation module is used to evaluate the screened spectral data from multiple dimensions. The multidimensional evaluation includes at least one evaluation parameter among the relative standard deviation, sensitivity, specificity, or accuracy of the spectral line intensity.

7. The system according to claim 5, characterized in that, The standard spectrum construction module is specifically used for: Laser-induced breakdown spectroscopy was performed on a standard coal sample to collect its original spectral data in the first and second bands; the spectral signals of the CN and C2 characteristic peaks were extracted from the original spectral data; and the extracted spectral signals were combined to obtain a standard spectral feature matrix.

8. The system according to claim 5, characterized in that, The spectra to be compared include at least one of rock spectra, metallic mineral spectra, overfocused spectra, defocused spectra, and obscured spectra; The spectral feature matrix to be compared has the same dimension as the standard spectral feature matrix.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the effective spectral screening method for raw coal based on the CN and C2 characteristic peaks as described in any one of claims 1-4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an information transmission implementation program, which, when executed by a processor, implements the steps of the effective spectral screening method for raw coal based on the CN and C2 characteristic peaks as described in any one of claims 1-4.

Images