Method and system for predicting content of elemental carbon in coal

Abstract

The invention discloses a method and system for predicting the content of elemental carbon in coal, and the method specifically comprises the following steps: (1) determining the coal type of the current coal, and obtaining a detection index of the corresponding coal type; and (2) inputting the related detection indexes into the elemental carbon content prediction model of the corresponding coal type, and measuring and calculating the elemental carbon content of the corresponding coal by the elemental carbon content prediction model. According to the method for predicting the content of the elemental carbon in the coal, the content of the elemental carbon in the current coal can be measured and calculated in real time, the carbon emission condition of coal combustion in the production process is reflected in real time, production planning is guided, and the method has the advantages of being low in cost, wide in application range and the like and has good application and popularization value.

Description

Technical Field

[0001] The present invention belongs to the technical field of carbon emission, and more specifically, the present invention relates to a method and system for predicting the elemental carbon content in coal. Background Art

[0002] China's steel industry carbon emissions account for more than 60% of the global steel industry's total carbon emissions and about 15% of the country's total carbon emissions. China's long-process smelting process uses coal as the main energy source, and the main carbon emissions in its production process come from the use (combustion) of coal resources.

[0003] At present, the steel industry is in the process of green and low-carbon transformation and development. Facing the realistic needs of the steel industry being included in the carbon trading market in 2025, the calculation of carbon emissions from coal in the domestic steel production process is mainly based on the LCA (Life Cycle Assessment) method set by the ISO14040 standard, using the carbon emission coefficient method. Compared with the large number of databases in multiple fields that have been formed abroad, the carbon emission coefficient of coal has regional and temporal characteristics, which does not conform to the actual situation of domestically produced coal.

[0004] According to IPCC2006 and the calculation method of coal carbon emission coefficient in the "Instructions for Filling in the 2023 Corporate Greenhouse Gas Emissions Accounting Report", elemental carbon content is one of the important parameters. Since the detection of elemental carbon content has high requirements on equipment, manpower, economy, etc., and the sample test results have a lag, it is impossible to predict the carbon emissions in the production process in real time. Summary of the invention

[0005] The present invention provides a method for predicting the elemental carbon content in coal, aiming to improve the above-mentioned problem.

[0006] The present invention is implemented as follows: a method for predicting the elemental carbon content in coal, the method specifically comprising the following steps:

[0007] (1) Determine the type of coal currently being produced and obtain the corresponding detection index of the coal type;

[0008] (2) The relevant detection indicators are input into the element carbon content prediction model of the corresponding coal type, and the element carbon content prediction model calculates the element carbon content of the corresponding coal.

[0009] Furthermore, the construction process of the element carbon content prediction model of each type of coal is as follows:

[0010] (21) constructing a coal sample, the coal sample including: coal type, coal detection index and elemental carbon content;

[0011] (22) Classify coal samples based on coal type and put coal samples of the same coal type into the same sample set;

[0012] (23) The coal samples in each sample set are analyzed to determine the correlation detection index with a strong correlation with the elemental carbon content, and a prediction model for the elemental carbon content of the corresponding coal type is constructed based on the correlation detection index.

[0013] Furthermore, the determination process of the correlation detection index is as follows:

[0014] The Pearson correlation coefficient between the carbon content of coal elements and each detection index is calculated, and the detection index with a Pearson correlation coefficient greater than the set correlation coefficient threshold is used as the correlation detection index.

[0015] Furthermore, when there are multiple correlation detection indicators, the process of constructing the element carbon content prediction model based on multiple correlation detection indicators is as follows:

[0016] (2351) Select the detection index with the largest Pearson correlation coefficient, extract the detection index and its element carbon content from all samples in the sample set, use the linear regression model to analyze the second element carbon content prediction model, and extract the fitting degree R of the second element carbon content prediction model. 2 ;

[0017] (2352) Extract multiple related detection indicators and their element carbon content from all samples in the sample set, use the multivariate linear regression model to analyze the third element carbon content prediction model, and extract the fitting degree R of the third element carbon content prediction model. 2 ;

[0018] (2353) Adjust the correlation coefficient threshold, return to step (2352), and output the third element carbon content prediction model with the highest fitting degree and its fitting degree R 2 ;

[0019] (2354) Detect the fitting degree R of the prediction model of the second element carbon content 2 Is it greater than the fitting degree R of the third element carbon content prediction model with the highest fitting degree? 2 If the test result is yes, the second element carbon content prediction model is used as the element carbon content prediction model. If the test result is no, the third element carbon content prediction model with the highest fitting degree is used as the element carbon content prediction model.

[0020] Furthermore, when there is only one correlation detection index, the construction process of the element carbon prediction model based on a single correlation detection index is as follows:

[0021] The selected correlation detection index and its element carbon content are extracted from all samples in the corresponding sample set, and the first element carbon content prediction model is analyzed using a linear regression model, and the fitting degree R of the first element carbon content prediction model is obtained. 2 , test the goodness of fit R 2 If the test result is greater than the set fitting threshold, the first element carbon content prediction model is used as the carbon prediction model.

[0022] Furthermore, the Pearson correlation coefficient r(X,Y i ) is calculated as follows:

[0023]

[0024] Among them, X is the data set composed of the elemental carbon content of all samples in the corresponding sample set, and Y i is a dataset consisting of the i-th detection index of all samples in the corresponding sample set, Cov(X,Y i ) represents the covariance of data set X and data set Y i The product of the covariances of Var|X| and Var|Y i | respectively represent the variance of data set X and data set Y i The variance of .

[0025] Furthermore, the testing indicators include: coal's bomb calorific value, total moisture, moisture, fixed carbon, and dry basis sulfur content.

[0026] The present invention is implemented as follows: a system for predicting the elemental carbon content in coal, the system comprising:

[0027] An input unit, a processing unit and an output unit are connected in sequence, wherein the processing unit integrates a prediction model of the elemental carbon content of each type of coal;

[0028] The input unit is used to input the current coal type and its detection indicators, and send them to the processing unit. The processing unit selects the corresponding element carbon content prediction model based on the current coal type, and inputs the relevant detection indicators into the selected element carbon content prediction model. The element carbon content prediction model calculates the element carbon content of the current coal and outputs it through the display unit.

[0029] Furthermore, the testing indicators include: coal's bomb calorific value, total moisture, moisture, fixed carbon, and dry basis sulfur content.

[0030] The method for predicting the elemental carbon content in coal provided by the present invention can calculate the elemental carbon content in the current coal in real time, reflect the carbon emissions of coal combustion in the production process in real time, and guide production planning. It has the advantages of low cost and wide application range, and has good promotion and application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 A flow chart of a method for predicting the elemental carbon content in coal provided by an embodiment of the present invention;

[0032] Figure 2 A schematic diagram of the structure of a system for predicting elemental carbon content in coal provided in an embodiment of the present invention. Implementation

[0034] The specific implementation modes of the present invention are further explained in detail below by describing the embodiments with reference to the accompanying drawings, so as to help those skilled in the art to have a more complete, accurate and in-depth understanding of the inventive concept and technical solution of the present invention.

[0035] Figure 1 A flow chart of a method for predicting the elemental carbon content in coal provided in an embodiment of the present invention, the method specifically comprises the following steps:

[0036] (1) Determine the type of coal currently being produced and obtain the corresponding detection index of the coal type;

[0037] In the embodiment of the present invention, the current coal is tested for testing indicators such as bomb calorific value, total moisture, moisture, fixed carbon, and dry basis sulfur content (the amount of total sulfur minus water).

[0038] (2) The relevant detection indicators are input into the element carbon content prediction model of the corresponding coal type. The element carbon content prediction model predicts the element carbon content of the corresponding coal. The element carbon content prediction can be used for coal CO 2 Calculation of emission factors.

[0039] In the embodiment of the present invention, the element carbon content prediction model of different coal types is different. The element carbon content prediction model is constructed based on the detection index that the corresponding coal type is strongly correlated with the element carbon content. Therefore, for different coal types, the detection index used to construct the element carbon content prediction model may be different, and the expression form of the element carbon content prediction model may also be different. The construction process of the element carbon content prediction model of each coal type is described below, and the construction process is specifically as follows:

[0040] (21) constructing a coal sample, the coal sample includes: coal type, coal detection index and element carbon content, and the element carbon content in the coal sample is the detection value;

[0041] (22) Classify coal samples based on coal type and put coal samples of the same coal type into the same sample set;

[0042] (23) The coal samples in each sample set are analyzed to determine the correlation detection index with a strong correlation with the elemental carbon content, and a prediction model for the elemental carbon content of the corresponding coal type is constructed based on the correlation detection index.

[0043] In the embodiment of the present invention, the coal sample analysis process in each sample set is specifically as follows:

[0044] (231) Calculate the Pearson correlation coefficient between elemental carbon content and each detection index;

[0045] The Pearson coefficient is used to analyze the correlation between element carbon content and detection index. The Pearson correlation coefficient between element carbon content and the i-th detection index is r(X,Y i ) is calculated as follows:

[0046]

[0047] Among them, X is the data set composed of the elemental carbon content of all samples in the corresponding sample set, and Y i is a dataset consisting of the i-th detection index of all samples in the corresponding sample set, Cov(X,Y i ) represents the covariance of data set X and data set Y i The product of the covariances of Var|X| and Var|Y i | respectively represent the variance of data set X and data set Y i The variance of .

[0048] (232) Selecting a detection indicator whose Pearson correlation coefficient is greater than a set correlation coefficient threshold as a correlation detection indicator;

[0049] In the embodiment of the present invention, r(X,Y i ) has a value range of -1≤r≤1. When |r(X,Y i )| is greater than 0.8, indicating that the correlation between the element carbon content and the i-th detection index is very strong. i )| is in the numerical range of 0.6 to 0.8, indicating that the correlation between the element carbon content and the i-th detection index is strong. Therefore, the correlation coefficient threshold is generally set around 0.6.

[0050] (233) Detect whether the correlation detection index is one, if the detection result is yes, execute step (234), if the detection result is no, execute step (235);

[0051] (234) Constructing an elemental carbon prediction model based on selected single correlation detection indicators;

[0052] In the embodiment of the present invention, the process of constructing the element carbon prediction model based on a single correlation detection index is as follows:

[0053] The selected correlation detection index and its element carbon content are extracted from all samples in the corresponding sample set, and the first element carbon content prediction model is analyzed using a linear regression model, and the fitting degree R of the first element carbon content prediction model is obtained. 2 , test the goodness of fit R 2 If the test result is yes, the first element carbon content prediction model is used as the coal element carbon prediction model, and the fitting degree R 2 If the value is less than the set fitting threshold, the Laida criterion is used to sort out the samples in the sample set, remove the deviated samples and re-fit the correlation coefficient to obtain the fitting equation. If the fitting degree R of the fitting equation is 2 If it is still not satisfied, the construction of the element carbon prediction model cannot be completed.

[0054] (235) A carbon prediction model is constructed based on multiple selected correlation detection indicators.

[0055] In an embodiment of the present invention, the process of constructing an element carbon prediction model based on multiple correlation detection indicators is as follows:

[0056] First, the correlation detection index with the largest Pearson correlation coefficient is selected, and the correlation detection index and its element carbon content are extracted from all samples in the sample set. The linear regression model is used to analyze the second element carbon content prediction model, and the fitting degree R of the second element carbon content prediction model is extracted. 2 ;

[0057] Extract multiple relevant detection indicators and their element carbon content from all samples in the sample set, use linear regression model to analyze the third element carbon content prediction model, and extract the fitting degree R of the third element carbon content prediction model. 2 ;

[0058] Adjust the correlation coefficient threshold, return to step (232), and output the third element carbon content prediction model with the highest fitting degree and its fitting degree R 2 ;

[0059] Detection of the fitting degree R of the prediction model of the second element carbon content 2 Is it greater than the fitting degree R of the third element carbon content prediction model with the highest fitting degree? 2 If the test result is yes, the second element carbon content prediction model is used as the element carbon content prediction model. If the test result is no, the third element carbon content prediction model with the highest fitting degree is used as the element carbon content prediction model for this type of coal.

[0060] The method for predicting the elemental carbon content in coal provided by the present invention can measure the elemental carbon content in the current coal in real time, reflect the carbon emissions of coal combustion in the production process in real time, and guide production planning. It has the advantages of low cost and wide application range, and has good promotion and application value.

[0061] Figure 2 The structural diagram of the element carbon content prediction system in coal provided by the embodiment of the present invention, for the convenience of explanation, only the part related to the embodiment of the present invention is shown, and the system includes:

[0062] An input unit, a processing unit and an output unit are connected in sequence, wherein the element carbon content prediction model of each type of coal is integrated in the processing unit, and the element carbon content prediction model process of the coal type is the same as the element carbon content prediction model construction process in the element carbon content prediction method, and the present invention will not be repeated here;

[0063] The input unit is used to input the coal type and its detection index of the current coal, and send them to the processing unit. The processing unit selects the corresponding element carbon content prediction model based on the coal type of the current coal, and inputs the relevant detection index into the selected element carbon content prediction model. The element carbon content prediction model outputs the element carbon content prediction value of the current coal, and the element carbon content prediction value can be used for coal CO 2 Calculation of emission coefficient and carbon emissions.

[0064] The present invention has been described exemplarily. Obviously, the specific implementation of the present invention is not limited to the above-mentioned method. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention, or the concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. A method for predicting the elemental carbon content in coal, characterized in that: The method specifically comprises the following steps: (1) Determine the type of coal currently being produced and obtain the corresponding detection index of the coal type; (2) The relevant detection indicators are input into the element carbon content prediction model of the corresponding coal type, and the element carbon content prediction model calculates the element carbon content of the corresponding coal.

2. The method for predicting the elemental carbon content in coal according to claim 2, characterized in that: The construction process of the element carbon content prediction model for each type of coal is as follows: (21) constructing a coal sample, the coal sample including: coal type, coal detection index and elemental carbon content; (22) Classify coal samples based on coal type and put coal samples of the same coal type into the same sample set; (23) The coal samples in each sample set are analyzed to determine the correlation detection index with a strong correlation with the elemental carbon content, and a prediction model for the elemental carbon content of the corresponding coal type is constructed based on the correlation detection index.

3. The method for predicting the elemental carbon content in coal according to claim 2, characterized in that: The specific process of determining the correlation detection index is as follows: The Pearson correlation coefficient between the carbon content of coal elements and each detection index is calculated, and the detection index with a Pearson correlation coefficient greater than the set correlation coefficient threshold is used as the correlation detection index.

4. The method for predicting the elemental carbon content in coal according to claim 2, characterized in that: When there are multiple correlation detection indicators, the process of constructing the element carbon prediction model based on multiple correlation detection indicators is as follows: (2351) Select the detection index with the largest Pearson correlation coefficient, extract the detection index and its element carbon content from all samples in the sample set, use the linear regression model to analyze the second element carbon content prediction model, and extract the fitting degree R of the second element carbon content prediction model. 2 ; (2352) Extract multiple related detection indicators and their element carbon content from all samples in the sample set, use the multivariate linear regression model to analyze the third element carbon content prediction model, and extract the fitting degree R of the third element carbon content prediction model. 2 ; (2353) Adjust the correlation coefficient threshold, return to step (2352), and output the third element carbon content prediction model with the highest fitting degree and its fitting degree R 2 ; (2354) Detect the fitting degree R of the prediction model of the second element carbon content 2 Is it greater than the fitting degree R of the third element carbon content prediction model with the highest fitting degree? 2 If the test result is yes, the second element carbon content prediction model is used as the element carbon content prediction model. If the test result is no, the third element carbon content prediction model with the highest fitting degree is used as the element carbon content prediction model.

5. The method for predicting the elemental carbon content in coal according to claim 2, characterized in that: When there is one correlation detection indicator, the construction process of the element carbon prediction model based on a single correlation detection indicator is as follows: The selected correlation detection index and its element carbon content are extracted from all samples in the corresponding sample set, and the first element carbon content prediction model is analyzed using a linear regression model, and the fitting degree R of the first element carbon content prediction model is obtained. 2 , test the goodness of fit R 2 If the test result is yes, the first element carbon content prediction model is used as the element carbon content prediction model.

6. The method for predicting the elemental carbon content in coal according to claim 3, characterized in that: The Pearson correlation coefficient between the element carbon content and the i-th detection index r(X,Y i ) is calculated as follows: Among them, X is the data set composed of the elemental carbon content of all samples in the corresponding sample set, and Y i is a dataset consisting of the i-th detection index of all samples in the corresponding sample set, Cov(X,Y i ) represents the covariance of data set X and data set Y i The product of the covariances of Var|X| and Var|Y i | respectively represent the variance of data set X and data set Y i The variance of .

7. The method for predicting the elemental carbon content in coal according to any one of claims 1 to 6, characterized in that: The testing indicators include: coal's bomb calorific value, total moisture, moisture, fixed carbon, and dry basis sulfur content.

8. A system for predicting the elemental carbon content in coal, characterized in that: The system comprises: An input unit, a processing unit and an output unit are connected in sequence, wherein the processing unit integrates a prediction model of the elemental carbon content of each type of coal; The input unit is used to input the current coal type and its detection indicators, and send them to the processing unit. The processing unit selects the corresponding element carbon content prediction model based on the current coal type, and inputs the relevant detection indicators into the selected element carbon content prediction model. The element carbon content prediction model calculates the element carbon content of the current coal and outputs it through the display unit.

9. The elemental carbon content prediction system in coal according to claim 8, characterized in that: The testing indicators include: coal's bomb calorific value, total moisture, moisture, fixed carbon, and dry basis sulfur content.

Images