Coal mixing judgment method and storage medium

Abstract

The invention provides a coal mixing judgment method and a storage medium, and the method comprises the steps: carrying out the dynamic monitoring of the data of the same batch of coal at different time points, and the data comprises coal quality parameter data and coal briquette spectrum data; the coal quality parameter data comprises a plurality of parameter data, acquiring a difference value between each parameter data and each parameter standard value, and judging whether mixed coal exists or not according to the plurality of difference values; if coal blending is judged according to the coal quality parameter data, acquiring coal briquette spectral data, comparing the coal briquette spectral data with a single coal type standard spectral database, and further verifying whether coal blending exists or not according to the two indexes of the characteristic spectral peak and / or the spectral similarity of the spectral data. The technical problems that in the prior art, the period is long, the cost is high, and the representativeness is limited when whether coal is mixed or not is detected are solved, the mixed coal can be rapidly judged through the mixed coal judgment method, the mixed coal is judged from multiple angles, and a comprehensive mixed coal judgment system is established.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of coal, and in particular to a mixed coal judgment method and a storage medium. BACKGROUND

[0002] In the entire industrial chain of coal production, mining, washing, trading, and terminal use (such as power, metallurgy, and chemical industry), the phenomenon of mixed coal is widespread. Mixed coal refers to the physical mixing of two or more different types of coal from different origins, different coalification degrees, or different qualities. If the mixing process lacks precise quality control, it will result in drastic fluctuations and instability in the quality indicators (such as calorific value, ash content, volatile matter, and sulfur content) of the final coal product.

[0003] This quality instability can trigger a series of chain problems: in the combustion link, it can lead to decreased combustion efficiency and deteriorated boiler operating conditions; in terms of environmental protection, it can increase the emission of pollutants such as sulfur dioxide and smoke due to uncontrolled sulfur and ash content; in severe cases, it can even cause damage to combustion equipment, coal conveying systems, and other equipment, posing safety production hazards.

[0004] To identify and control mixed coal, existing technologies mainly rely on traditional manual sampling and laboratory chemical analysis methods. The Chinese national standard GB / T 15591-2013 "Commercial Coal Mixed Coal Judgment Method" represents the core of such traditional technologies, which mainly collects multiple coal samples at different positions, sends them to the laboratory for sample preparation and detailed industrial analysis, and finally determines the difference between the analysis results and the agreed coal quality through comparison.

[0005] However, such methods have inherent limitations: 1. Long detection cycle: From sampling, sample preparation to laboratory analysis, the entire process usually takes several hours or even days, which is severely lagging behind the real-time decision-making needs of production and trade.

[0006] 2. High cost: It relies on professional laboratories and personnel, with high labor, material, and time costs, making it difficult to conduct large-scale and high-frequency detection.

[0007] 3. Limited representativeness: Although the standard specifies the sampling method, manual sampling may still have representativeness errors, and cannot fully capture the overall non-uniformity of coal. SUMMARY

[0008] Based on the above problems, the present application provides a mixed coal judgment method and a storage medium, which solves the technical problems of long detection period, high cost, limited representation and the like in the prior art for detecting whether mixed coal exists, the mixed coal judgment method provided by the present application can obtain relevant information of coal in real time, realize rapid judgment of mixed coal, and meet the real-time monitoring demand in the process of coal production, trade and use; the accuracy of mixed coal judgment is improved by comprehensively considering various coal quality parameter indexes and spectrum information, adopting multi-index comprehensive judgment and spectrum information auxiliary judgment, combining dynamic monitoring and trend analysis; not only the abnormal situation of a single coal quality index is considered, but also mixed coal is judged from multiple angles through comprehensive scoring and spectrum information analysis, and a comprehensive mixed coal judgment system is established.

[0009] The present application provides a mixed coal judgment method, comprising: Dynamic monitoring of data of the same batch of coal at different time points, the data including coal quality parameter data and coal block spectrum data; The coal quality parameter data includes multiple parameter data, the difference between each parameter data and each parameter standard value is obtained, and whether there is mixed coal is judged according to the multiple differences; If it is judged that there is mixed coal according to the coal quality parameter data, the coal block spectrum data is obtained, the coal block spectrum data is compared with the single coal standard spectrum database, and whether there is mixed coal is further verified according to the two indexes of characteristic spectrum peak and / or spectrum similarity of the spectrum data.

[0010] In addition, the coal quality parameter data includes one or more of calorific value, ash content, volatile matter and sulfur content.

[0011] In addition, further verification of whether there is mixed coal according to the two indexes of characteristic spectrum peak and spectrum similarity of the spectrum data includes: When verification is performed according to the index of characteristic spectrum peak, the coal block spectrum data is compared with the single coal standard spectrum database, if the characteristic spectrum peak that does not exist in the single coal standard spectrum database appears in the coal block spectrum data or the characteristic spectrum peak appears disappearance, displacement or deformation, it is verified that there is mixed coal.

[0012] In addition, further verification of whether there is mixed coal according to the two indexes of characteristic spectrum peak and spectrum similarity of the spectrum data includes: When verification is performed according to the index of spectrum similarity, the similarity between the spectrum in the coal block spectrum data and the preset standard spectrum in the single coal standard spectrum database is calculated through a similarity algorithm, if the similarity is less than a threshold value, it is considered that there is mixed coal.

[0013] In addition, the single coal standard spectrum database stores standard spectrum data of single coal, denoted as , , …, }. Spectrum is set according to a linear mixing model is expressed as: = +…+ , wherein , , is the contribution degree of a single coal type; Based on non-negative constraints and summation constraints, the contribution degree of a single coal type is solved by an optimization algorithm under the constraint of minimizing the error between the calculated spectrum and the spectrum M in the coal block spectrum data; According to the contribution degree of a single coal type, each coal type and its proportion in the detected coal are output.

[0014] In addition, the non-negative constraints are , , all greater than 0.

[0015] In addition, the summation constraints are + +… =1.

[0016] In addition, the optimization algorithm is a non-negative least squares algorithm or a constrained least squares algorithm.

[0017] In addition, the error is the sum of squared residuals between the spectrum and the spectrum M in the coal block spectrum data.

[0018] The application also provides a storage medium storing computer instructions for executing the coal mixture judgment method described above when the computer executes the computer instructions.

[0019] The application solves the technical problems of long cycle, high cost, and limited representation in the prior art for detecting whether there is coal mixture in coal. The coal mixture judgment method provided by the application can obtain relevant information of coal in real time, realize rapid judgment of coal mixture, and meet the real-time monitoring needs in the process of coal production, trade and use. The application comprehensively considers various coal quality parameter indexes and spectrum information, adopts multi-index comprehensive judgment and spectrum information auxiliary judgment, combines dynamic monitoring and trend analysis, and improves the accuracy of coal mixture judgment. The application not only considers the abnormal situation of a single coal quality index, but also judges the coal mixture from multiple angles through comprehensive scoring and spectrum information analysis, and establishes a comprehensive coal mixture judgment system. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 The flowchart of the coal mixture judgment method provided by an embodiment of the application. DETAILED DESCRIPTION

[0021] The application will be described in further detail in connection with the specific embodiments and the accompanying drawings. It is intended to serve only to explain the specific embodiments and not to limit the application in any way. The scope of the application is defined by the claims.

[0022] With reference to Figure 1 The application proposes a mixed coal judgment method, comprising: Step S001, dynamically monitoring the data of the same batch of coal at different time points, including coal quality parameter data and coal block spectrum data; Step S002, the coal quality parameter data includes multiple parameter data, the difference between each parameter data and each parameter standard value is obtained, and whether there is mixed coal is judged according to the multiple differences; Step S003, if it is judged that there is mixed coal according to the coal quality parameter data, the coal block spectrum data is obtained, the coal block spectrum data is compared with the single coal standard spectrum database, and whether there is mixed coal is further verified according to the two indexes of characteristic spectrum peak and / or spectrum similarity of the spectrum data.

[0023] Coal quality parameter data collection: real-time acquisition of key coal quality parameters of coal by coal quality rapid detection equipment, i.e. multiple parameter data, including but not limited to calorific value, ash content, moisture content, volatile matter and sulfur content.

[0024] Coal block spectrum data collection: collecting the spectrum data (such as near-infrared spectrum, X-ray fluorescence spectrum) of the detected coal. Dynamic monitoring of the data of the same batch of coal at different time points means that the coal is detected, i.e. the detected coal.

[0025] When judging mixed coal by using coal quality parameter data index, the difference between each parameter data and each parameter standard value is obtained, which is as follows: Calorific value judgment: set the normal fluctuation range [Qmin, Qmax] of the calorific value of the detected coal, when the real-time detected calorific value Q exceeds the range, i.e. Q<Qmin or Q>Qmax, it is preliminarily determined that it may be mixed coal.

[0026] Ash content judgment: calculate the difference ΔA =|A - A0| between the ash content A of the real-time detected coal and the standard ash content A0 of the coal, if ΔA exceeds the set ash content threshold Ath, it is determined that it may be mixed coal.

[0027] Volatile matter judgment: calculate the difference ΔV =|V - V0| between the volatile matter V of the real-time detected coal and the standard volatile matter V0, when ΔV exceeds the set volatile matter threshold Vth, it is suspected to be mixed coal.

[0028] Sulfur content determination: Set the normal range of sulfur content [Smin, Smax], if the real-time detected sulfur content S of the coal is not within the range, i.e. S < Smin or S > Smax, there may be mixed coal.

[0029] Optionally, when judging whether there is mixed coal according to the plurality of difference values, weights are set for each coal quality parameter, the weights of the calorific value, ash content, volatile matter, and sulfur content are w1, w2, w3, and w4 respectively, and w1 + w2 + w3 + w4 = 1.

[0030] The scores S1, S2, S3, and S4 are determined according to the degree of deviation of each parameter from the normal range.

[0031] The comprehensive score S = w1S1 + w2S2 + w3S3 + w4S4 is calculated, and when S exceeds the set comprehensive threshold Sth, it is judged as mixed coal.

[0032] The principle of using spectral analysis to analyze mixed coal is as follows: the absorption, reflection, or emission characteristics of a specific band of light (such as near-infrared light, X-rays) of each pure, chemically fixed substance (here referring to a specific type of coal) are uniquely determined by its molecular structure (such as C-H, O-H, N-H chemical bonds), atomic composition, and crystal structure. This characteristic is manifested as a characteristic absorption peak, reflection valley, or fluorescence peak at a specific wavelength or energy. Coals from different origins and different coal-forming conditions differ in composition (such as fixed carbon, volatile matter, moisture, and minerals in ash). Therefore, a single coal type has its unique and stable standard spectrum. For example, the absorption of high-volatile coal at a specific wavelength in the near-infrared region will be significantly different from that of low-volatile coal.

[0033] In a mixture, if there is no strong chemical interaction between the components, the spectral signal (such as absorbance) of the mixture at a certain wavelength can be approximately considered as the linear superposition of the spectral signals of the pure components in proportion to their concentrations. If a batch of coal is physically mixed from two or more different types of coal (A, B, C...), the real-time spectrum measured by the spectrometer is actually the weighted sum of the standard spectra of each component coal (A, B, C...) in proportion to their mixing ratio.

[0034] Optionally, based on the above two core principles, mixed coal can be judged in the following ways: By matching characteristic spectral peaks: compare the real-time collected coal spectrum with the known single coal type standard spectrum database. If a characteristic peak that does not exist in a single coal type standard spectrum appears in the real-time spectrum, or a characteristic peak that should exist disappears, moves significantly, or deforms, it strongly suggests that other types of coal may exist in the sample, i.e. mixed coal has occurred.

[0035] Spectral similarity analysis: Since the spectrum of mixed coal is a mixture of multiple coal spectra, its overall shape and contour will inevitably differ from any single standard spectrum of a coal type. By calculating the similarity between the real-time spectrum and the preset standard spectrum, the difference can be quantified (such as calculating the similarity using correlation coefficient, Euclidean distance, or Mahalanobis distance). The lower the similarity, the greater the difference between the real-time spectrum and the standard spectrum, and the higher the likelihood of mixed coal.

[0036] The present application solves the technical problems of long detection period, high cost, and limited representation in the prior art for detecting whether mixed coal exists in coal. The mixed coal judgment method proposed by the present application can obtain real-time information of coal, realize rapid judgment of mixed coal, and meet the real-time monitoring needs in the process of coal production, trade, and use. The present application comprehensively considers multiple coal quality parameter indicators and spectral information, uses multi-index comprehensive judgment and spectral information auxiliary judgment, combines dynamic monitoring and trend analysis, and improves the accuracy of mixed coal judgment. The present application not only considers the abnormal situation of a single coal quality indicator, but also judges mixed coal from multiple angles through comprehensive scoring and spectral information analysis, and establishes a comprehensive mixed coal judgment system.

[0037] In one embodiment, the coal quality parameter data includes one or more of calorific value, ash content, volatile matter, and sulfur content.

[0038] Judgment is made by multiple coal quality parameters, making the judgment result of mixed coal more accurate.

[0039] When judging mixed coal using coal quality parameter data indicators, the difference between each parameter data and each parameter standard value is obtained, as follows: Calorific value judgment: Set the normal fluctuation range [Qmin, Qmax] of the calorific value of the detected coal. When the real-time detected calorific value Q exceeds this range, i.e., Q<Qmin or Q>Qmax, it is preliminarily determined that it may be mixed coal.

[0040] Ash content judgment: Calculate the difference ΔA =|A - A0| between the ash content A of the real-time detected coal and the standard ash content A0 of the coal type. If ΔA exceeds the set ash content threshold Ath, it is determined that it may be mixed coal.

[0041] Volatile matter judgment: Calculate the difference ΔV =|V - V0| between the volatile matter V of the real-time detected coal and the standard volatile matter V0. When ΔV exceeds the set volatile matter threshold Vth, it is suspected to be mixed coal.

[0042] Sulfur content determination: set the normal range of sulfur content [Smin, Smax], if the sulfur content S of the real-time detected coal is not in the range, i.e. S < Smin or S > Smax, there may be mixed coal.

[0043] Optionally, when judging whether there is mixed coal according to the plurality of difference values, weights are set for each coal quality parameter, the weights of the calorific value, ash content, volatile matter and sulfur content are w1, w2, w3 and w4 respectively, and w1 + w2 + w3 + w4 = 1.

[0044] The scores S1, S2, S3 and S4 are determined according to the degree of deviation of each parameter from the normal range.

[0045] The comprehensive score S = w1S1 + w2S2 + w3S3 + w4S4 is calculated, and when S exceeds the set comprehensive threshold Sth, it is judged that there is mixed coal.

[0046] In one of the embodiments, further verification of whether there is mixed coal according to the two indexes of characteristic spectral peak and spectral similarity of spectral data includes: When verification is performed according to the index of characteristic spectral peak, the coal block spectral data is compared with the single coal standard spectral database, and if the characteristic spectral peak that does not exist in the single coal standard spectral database appears in the coal block spectral data or the characteristic spectral peak appears, disappears, shifts or deforms, it is verified that there is mixed coal.

[0047] By characteristic spectral peak matching: the real-time collected coal spectrum is compared with the known single coal standard spectral database. If a characteristic peak that does not exist in a single coal standard spectrum appears in the real-time spectrum, or a characteristic peak disappears, shifts or deforms significantly, it strongly suggests that other types of coal may exist in the sample, i.e. mixed coal occurs.

[0048] Through this embodiment, when the characteristic spectral peak changes significantly, the mixed coal can be accurately judged according to this characteristic.

[0049] In one of the embodiments, further verification of whether there is mixed coal according to the two indexes of characteristic spectral peak and spectral similarity of spectral data includes: When verification is performed according to the index of spectral similarity, the similarity between the spectrum in the coal block spectral data and the pre-set standard spectrum in the single coal standard spectral database is calculated by a similarity algorithm, and if the similarity is less than a threshold value, it is considered that there is mixed coal.

[0050] Spectral similarity analysis: since the spectrum of the mixed coal is a mixture of the spectra of multiple coal types, the overall shape and contour of the spectrum will inevitably differ from the standard spectrum of any single coal type. The difference can be quantified by calculating the similarity between the real-time spectrum and the preset standard spectrum (e.g., using the correlation coefficient, Euclidean distance, or Mahalanobis distance to calculate the similarity). The lower the similarity, the greater the difference between the real-time spectrum and the standard spectrum, and the higher the possibility of mixed coal.

[0051] Through this embodiment, the mixed coal can be accurately determined according to the spectral similarity.

[0052] Optionally, the results of the determination of the characteristic spectral peak and the spectral similarity are weighted to obtain the final determination result.

[0053] In one of the embodiments, the single coal type standard spectrum database stores the standard spectrum data of the single coal type, denoted as: , ,… }; According to the linear mixing model, the spectrum is expressed as: = +…+ , wherein 、 ,… are the contribution degrees of the single coal types. Based on the non-negative constraint and the summation constraint, the contribution degrees of the single coal types are solved by an optimization algorithm under the constraint that the error between the calculated spectrum and the spectrum M in the coal block spectrum data is minimized; According to the contribution degrees of the single coal types, the types and proportions of the coal types in the detected coal are output.

[0054] In this embodiment, the types and proportions of the coal types in the mixed coal are obtained by the optimization algorithm, and more detailed results are given.

[0055] In one of the embodiments, the non-negative constraint is that 、 ,… are all greater than 0.

[0056] Since the proportions of the coal types cannot be negative at the physical level, the contribution degrees of the single coal types are subjected to the non-negative constraint, so that the mathematical formula is correctly solved.

[0057] In one of the embodiments, the summation constraint is that + +… = 1.

[0058] Since the sum of the proportions of all components is 100%, the contributions of single coal types are summed up to meet the real situation.

[0059] In one embodiment, the optimization algorithm is a non-negative least square algorithm or a constrained least square algorithm.

[0060] The non-negative least square algorithm has fast calculation speed and good numerical stability, and is not prone to errors due to calculation errors, and it can absolutely guarantee that all proportions solved are not negative.

[0061] The result obtained by the constrained least square algorithm is directly normalized.

[0062] In one embodiment, the error is the sum of squares of residuals between the spectrum M and the spectrum data of the coal block.

[0063] The residual sum of squares algorithm achieves the best balance point between calculation efficiency, theoretical rigor and engineering practicability, which guarantees that it is easy to solve mathematically and the solution is unique.

[0064] The present application also provides a storage medium, which stores computer instructions for executing the coal blending judgment method.

[0065] In the context of the present disclosure, the storage medium can be a tangible medium, which can contain or store programs for use by or in conjunction with an instruction execution system, device or apparatus. The storage medium can be a machine-readable signal medium or a machine-readable storage medium. Alternatively, the storage medium can be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a compact disc read-only memory (CD-ROM), a magnetic tape, a floppy disk and an optical data storage device, etc.

[0066] The above only describes the principles and preferred embodiments of the present application. It should be noted that, for those skilled in the art, on the basis of the principles of the present application, a number of other variants can also be made, which should also be considered as the protection scope of the present application.​

Claims

1. A method for determining mixed coal, characterized in that, include: Dynamic monitoring of data from the same batch of coal at different time points, including coal quality parameter data and coal block spectral data; Coal quality parameter data includes multiple parameters. The difference between each parameter and its standard value is obtained, and the presence of mixed coal is determined based on the multiple differences. If the coal quality parameter data indicates the presence of mixed coal, then the spectral data of the coal block is obtained. The spectral data of the coal block is compared with the standard spectral database for a single coal type. The presence of mixed coal is further verified based on the characteristic spectral peaks and / or spectral similarity of the spectral data.

2. The method for determining mixed coal according to claim 1, characterized in that, Coal quality parameters include one or more of the following: calorific value, ash content, volatile matter, and sulfur content.

3. The method for determining mixed coal according to claim 1, characterized in that, Further verification of the presence of mixed coal is based on two indicators: characteristic spectral peaks and spectral similarity of the spectral data. When verifying based on the characteristic spectral peak, the spectral data of the coal block is compared with the standard spectral database of a single coal type. If the spectral data of the coal block contains characteristic spectral peaks that are not found in the standard spectral database of a single coal type, or if the characteristic spectral peaks disappear, shift, or deform, then it is verified that there is mixed coal.

4. The method for determining mixed coal according to claim 1, characterized in that, Further verification of the presence of mixed coal is based on two indicators: characteristic spectral peaks and spectral similarity of the spectral data. When verifying based on the spectral similarity index, the similarity algorithm is used to calculate the similarity between the spectrum in the coal block spectral data and the preset standard spectrum in the standard spectral database of a single coal type. If the similarity is less than the threshold, it is considered to be mixed coal.

5. The method for determining mixed coal according to claim 1, characterized in that, The single-coal-type standard spectral database stores standard spectral data for a single coal type, denoted as { , , ..., }; The spectrum is set according to the linear mixing model. Represented as: = +…+ ,in , … Contribution of a single coal type; Based on nonnegativity constraints and summation constraints, the calculated spectrum is... Under the constraint of minimizing the error between the spectrum M in the coal block spectral data, the contribution of a single coal type is solved by an optimization algorithm; Output the types and proportions of each coal type in the tested coal based on the contribution of a single coal type.

6. The method for determining mixed coal according to claim 5, characterized in that, Nonnegativity constraint is , … All are greater than 0.

7. The method for determining mixed coal according to claim 5, characterized in that, Summation constraint is + +… =1.

8. The method for determining mixed coal according to claim 5, characterized in that, The optimization algorithm is either a non-negative least squares algorithm or a constrained least squares algorithm.

9. The method for determining mixed coal according to claim 5, characterized in that, Error is spectral The sum of squared residuals between the spectrum M in the coal block spectral data and the spectrum M in the coal block spectral data.

10. A storage medium, characterized in that, The storage medium stores computer instructions, which, when executed by the computer, are used to perform the coal mixing determination method as described in claim 1.

Images