CN113313394A - Coke quality evaluation method - Google Patents

Abstract

The invention discloses a coke quality evaluation method, which comprises the following steps: establishing an evaluation index system of coke quality; forming an original data matrix X by the evaluation indexes of each coke sample in the period, and carrying out standardized dimensionless processing on the original data matrix X to obtain a standardized matrix Z; calculating the correlation coefficient of each evaluation index in the standardized matrix Z, and constructing a correlation coefficient matrix R; calculating the eigenvalue and eigenvector of the correlation coefficient matrix R, and constructing a comprehensive characteristic index based on one eigenvector; and determining the previous p comprehensive characteristic indexes with large variance contribution rate, and calculating the weights of the previous p comprehensive characteristic indexes to obtain the comprehensive evaluation index of the coke quality. The defects that man-made subjective influence factors are large, the number of indexes is large, the indexes are overlapped and the like are overcome, a plurality of coke quality indexes are converted into a plurality of comprehensive characteristic indexes on the premise of losing little information by using a dimension reduction thought, and the evaluation method for comprehensively reflecting the coke quality is established.

Description

Coke quality evaluation method

Technical Field

The invention belongs to the technical field of coke for blast furnace iron making, and particularly relates to a coke quality evaluation method.

Background

Coke is a vital fuel for blast furnace ironmaking, and the quality of the coke directly influences various technical and economic indexes of the blast furnace. The coke mainly plays roles of a heating agent, a reducing agent, a carburizing agent and a material column framework in the blast furnace. With the continuous progress of iron-making technology, the coal injection quantity of a blast furnace is continuously improved, the action of a heating agent and a reducing agent of coke is partially replaced by the injected coal powder, meanwhile, the load of the coke in the blast furnace is increased due to the reduction of coke ratio, and the action of a stock column framework of the coke is more important in the smelting process. The coke is subjected to the influence of mechanical actions such as collision, extrusion, abrasion and the like in the blast furnace, and is simultaneously subjected to the influence of chemical actions such as dissolution loss reaction, alkali metal corrosion, slag iron corrosion and the like, and is continuously degraded from top to bottom. Under the condition of intensified smelting, the coke only has good enough quality, and can fully play the role of the coke in blast furnace iron making in the process of gradual degradation, so that the blast furnace is kept to be stable and smooth.

At present, no uniformly approved method for evaluating coke quality exists in the iron-making industry. According to researches such as 'technical research on a system for evaluating the quality of coke charged into a steel-bearing blast furnace', and the like, different weights are given to various indexes of the coke for scoring evaluation, so that artificial subjective influence factors are large, the quantity of the indexes is large, the indexes are overlapped, and the real quality of the coke cannot be objectively and comprehensively reflected.

Disclosure of Invention

The invention provides a coke quality evaluation method, aiming at improving the problems.

The invention is realized in such a way, and the coke quality evaluation method specifically comprises the following steps:

s1, establishing an evaluation index system of coke quality;

s2, forming the evaluation indexes of each coke sample in the period into an original data matrix X, and carrying out standardized dimensionless processing on the original data matrix X to obtain a standardized matrix Z;

s3, calculating the correlation coefficient of each evaluation index in the standardized matrix Z, and constructing a correlation coefficient matrix R;

s4, calculating the eigenvalue and the eigenvector of the correlation coefficient matrix R, and constructing a comprehensive characteristic index based on the eigenvector;

s5, determining the first p comprehensive characteristic indexes with large variance contribution rate, and calculating the weight of the first p comprehensive characteristic indexes to obtain the comprehensive evaluation index of coke quality.

Further, the evaluation index system of the coke quality comprises: cold strength index, hot performance index, physical and chemical index, wherein,

the cold strength indicators include: crushing strength M₄₀And abrasion resistance M₁₀；

The thermal state performance indexes include: reactive CRI and post-reaction intensity CSR;

the physical and chemical indexes comprise: ash content A_dMoisture content H₂O, S and V_dafFixed carbon C_adThe ratio L of the particle size of more than 80mm₈₀The particle size is 60-80 mm and the ratio L₆₀The particle size is 40-60 mm₄₀The particle size of 25-40 mm is L₂₅Coke breeze content L₀Average particle size D_AverageAnd powder-to-coke ratio C_{Fine coke}。

Further, the eigenvalues of the correlation coefficient matrix R are sorted from large to small as follows: lambda [ alpha ]₁≧λ₂…≧λ_j…≧λ_m≧ 0, and corresponding feature vector u₁、u₂…u_j…u_mWherein u is_j＝[u_1j,u_2j,u_3j,…,u_mj]^TThe overall characteristic index is expressed as follows:

F₁＝u₁₁z₁+u₂₁z₂+…+u_j1z_j+…+u_m1z_m

F₂＝u₁₂z₁+u₂₂z₂+…+u_j2z_j+…+u_m2z_m

…

F_j＝u_1jz₁+u_2jz₂+…+u_jjz_j+…+u_mjz_m

…

F_m＝u_1mz₁+u_2mz₂+…+u_jmz_j+…+u_mmz_m

wherein z is₁，z₂…z_j…z_mFor standardized evaluation index, F_mIs the m-th comprehensive characteristic index.

Further, the characteristic index coefficient u is integrated_ijThe expression (c) is specifically as follows:

K_ijis the value of the ith evaluation index in the jth comprehensive characteristic index in the comprehensive characteristic index load matrix, lambda_jAnd representing the characteristic value of the jth comprehensive characteristic index.

Further, the expression of the coke quality comprehensive evaluation index is specifically as follows:

F＝ω₁F₁+ω₂F₂+…ω_jF_j+…ω_pF_p

ω₁、ω₂…ω_j…ω_pthe weight of each comprehensive characteristic index of the coke is calculated by the following formula:

λ_jis the characteristic value, lambda, of the jth composite characteristic index_kThe characteristic value of the k-th comprehensive characteristic index is 1 to p

The invention overcomes the defects of large artificial subjective influence factor, large index quantity, overlapping of indexes and the like, converts a plurality of coke quality indexes into a plurality of comprehensive characteristic indexes by using a dimension reduction thought on the premise of losing little information, establishes an evaluation method for comprehensively reflecting the coke quality, provides a technical reference basis for purchasing coke, and can lead a blast furnace operator to prejudge the influence of the coke on the furnace condition in advance by dynamically tracking the quality fluctuation condition of the coke for the blast furnace, make operation countermeasures and reduce the blast furnace iron-making cost.

Drawings

FIG. 1 is a macadam map of coke command evaluation parameters provided by an embodiment of the invention;

FIG. 2 is a flow chart of a coke quality evaluation method according to an embodiment of the present invention.

Detailed Description

The following detailed description of the embodiments of the present invention will be given in order to provide those skilled in the art with a more complete, accurate and thorough understanding of the inventive concept and technical solutions of the present invention.

Fig. 2 is a flowchart of a coke quality evaluation method according to an embodiment of the present invention, where the method includes the following steps:

and S1, establishing an evaluation index system of coke quality.

And in a statistical period, selecting an evaluation index capable of reflecting the coke quality from the production detection system. The coke evaluation indexes are classified into 3 types including cold strength indexes, hot performance indexes and physical and chemical indexes, wherein the cold strength indexes comprise: crushing strength M₄₀And abrasion resistance M₁₀(ii) a The thermal state performance indexes include: reactive CRI and post-reaction intensity CSR; the physical and chemical indexes comprise: ash content A_dMoisture content H₂O, S and V_dafFixed carbon C_adThe ratio L of the particle size of more than 80mm₈₀The particle size is 60-80 mm and the ratio L₆₀The particle size is 40-60 mm₄₀The particle size of 25-40 mm is L₂₅Coke breeze content L₀Average particle size D_AverageAnd powder-to-coke ratio C_{Fine coke}。

S2, carrying out standardized dimensionless processing on the original data matrix X consisting of the m evaluation indexes of the n coke samples in a statistical period to obtain a standardized matrix Z. The expression of the original data matrix X is X ═ X (X)_ij)_n×mThe expression of the normalized matrix Z is Z ═ Z (Z)_ij)_n×m；

x_ijRepresenting the jth index value of the ith coke sample, wherein 1 is less than or equal to i less than or equal to n, and 1 is less than or equal to j less than or equal to m, namely the jth column element at the ith row in the original data matrix X; z is a radical of_ijThe j index value of the ith coke sample after the normalization treatment is represented, namely the i row and the j column of elements in the normalization matrix Z;

the arithmetic mean value of the jth coke original index, namely the arithmetic mean value of the jth column element in the original data matrix X; sigma_jIs the standard deviation of the j th coke original index, namely the standard deviation of the j th column element in the original data matrix X.

S3, calculating a correlation coefficient between the evaluation indexes in the normalized matrix Z, and constructing a correlation coefficient matrix R where R is (R)_jk)_m×m；

Is the arithmetic mean of the j column elements in the normalized matrix Z; r is_jkIs a correlation coefficient between the j index and the k index, r_jk＝r_kj，r_jj＝1，1≦j≦m，1≦k≦m。

S4, calculating a characteristic value and a characteristic vector of the correlation coefficient matrix R, and constructing a comprehensive characteristic index based on one characteristic vector;

solving the characteristic equation R-Lambda I to obtain the characteristic value Lambda which is ordered from large to small₁≧λ₂…≧λ_j…≧λ_m≧ 0, and corresponding feature vector u₁、u₂…u_j…u_mWherein u is_j＝[u_1j,u_2j,u_3j…u_mj]^T. And (3) establishing m comprehensive characteristic indexes according to the characteristic vectors:

z₁，z₂…z_j…z_mfor standardized evaluation index, F₁Weighing the first overall characteristic index, F₂Weighing the second overall characteristic index, F_jWeighing the jth comprehensive characteristic index, F_mThe m-th comprehensive characteristic index is called.

And S5, determining the number of the comprehensive characteristic indexes. Sorting the variance contribution rates of the comprehensive characteristic indexes in a descending order, calculating the variance contribution rate and the accumulated variance contribution rate of each comprehensive characteristic index, comprehensively determining the first p comprehensive characteristic indexes according to the principle that the accumulated variance contribution rate reaches a specific value (such as 75%, 80% and 85%) or the characteristic value is more than 1 and combining a lithotripsy (such as shown in figure 1), and using F to comprehensively determine the first p comprehensive characteristic indexes₁、F₂…F_j…F_pP comprehensive characteristic indexes are provided, namely the comprehensive characteristic indexes represent the original m indexes to comprehensively analyze the coke quality. To the aboveComprehensive characteristic index F of coke quality obtained by comprehensive characteristic index analysis₁、F₂…F_j…F_pAnalysis is performed and the physical significance represented by the overall characteristic index of coke quality is explained.

α_jThe variance contribution rate, S, of the composite characteristic index j_pP ≦ m, which is the cumulative contribution of the first p synthetic performance indicators.

In actual calculation, the characteristic value, the total variance interpretation table and the comprehensive characteristic index load matrix of the comprehensive characteristic index can be obtained through SPSS software, and the coefficient of each original evaluation index in different linear combinations of the comprehensive characteristic indexes, namely the comprehensive characteristic index coefficient u_ijAnd (4) solving the expression of the previous p comprehensive characteristic indexes.

K_ijIs the value of the ith evaluation index in the jth comprehensive characteristic index in the comprehensive characteristic index load matrix, lambda_jIs the characteristic value of the jth comprehensive characteristic index.

(6) And calculating the weights of the previous p comprehensive characteristic indexes to obtain the coke quality comprehensive evaluation index.

F＝ω₁F₁+ω₂F₂+…ω_jF_j+…ω_pF_p (10)

F is a comprehensive evaluation index of coke quality, F₁、F₂…F_j…F_pTo reflect the overall characteristic index of a certain property of the coke, omega₁、ω₂…ω_j…ω_pThe weight of each comprehensive characteristic index of the coke.

Note: f is determined according to the physical significance represented by each comprehensive characteristic index and the requirement of coke on each comprehensive characteristic index₁、F₂…F_j…F_pSign in the comprehensive evaluation index expression F.

By calculating the specific numerical value of the coke daily comprehensive evaluation index and tracking the change trend, a blast furnace operator can predict the influence of coke quality fluctuation on the furnace condition in advance and make positive countermeasures.

Compared with the existing coke quality evaluation method, the method overcomes the defects of large artificial subjective influence factor, large index quantity, overlapping of indexes and the like, converts a plurality of coke quality indexes into a plurality of comprehensive characteristic indexes by using the dimension reduction thought on the premise of losing little information, establishes an evaluation method for comprehensively reflecting the coke quality, provides a technical reference basis for purchasing the coke, and can lead a blast furnace operator to prejudge the influence of the coke on the furnace condition in advance by dynamically tracking the quality fluctuation condition of the coke for the blast furnace, make operation countermeasures and reduce the blast furnace iron-making cost.

The invention is further explained in detail with reference to the attached drawings, and the coke quality evaluation method comprises the following steps:

(1) and establishing a coke quality evaluation index system. In a statistical period, an evaluation index and a numerical value which can reflect the coke quality are selected from a production detection system. The coke evaluation index includes cold strength index, hot performance index and physical and chemical index, wherein the cold strength index includes crushing strength M₄₀And abrasion resistance M₁₀(ii) a The hot state performance index comprises reactive CRI and post-reaction strength CSR, and the physical and chemical index comprises ash A_dMoisture content H₂O, S, and L, the particle size is greater than 80mm₈₀The particle size is 60-80 mm and the ratio L₆₀The particle size is 40-60 mm₄₀The particle size of 25-40 mm is L₂₅Coke breeze content L₀Average particle size D_AverageAnd powder-to-coke ratio C_{Fine coke}Total 3 classes of 14 indices.

(2) And carrying out standardized dimensionless treatment on the collected and sorted coke quality index value of 3 months.

(3) Calculating the correlation coefficient among the indexes according to the normalized index values, and constructing a correlation coefficient matrix R, wherein R is (R)_jk)_14×14. As can be seen from Table 1, the overlap between some indexes is severe, such as CRI and CSR, which are highly linear.

TABLE 1 correlation coefficient of coke index

Parameter(s)	CRI	CSR	M40	M10	Ad	H2O	S	L80	L60	L40	L25	L0	DAverage	CFine coke
															CRI	1.000	-0.993	0.086	-0.049	-0.141	0.205	-0.061	0.147	0.062	0.004	0.065	-0.093	0.032	-0.139
CSR	-0.993	1.000	-0.096	0.062	0.129	-0.202	0.084	-0.139	-0.050	0.011	-0.048	0.109	-0.017	0.112
															M40	0.086	-0.096	1.000	-0.249	0.004	0.260	-0.164	-0.121	0.009	-0.092	-0.037	0.027	-0.063	0.122
M10	-0.049	0.062	-0.249	1.000	-0.159	-0.141	0.266	-0.056	0.042	0.112	0.086	0.084	0.086	0.179
															Ad	-0.141	0.129	0.004	-0.159	1.000	0.283	-0.257	-0.016	-0.032	-0.030	-0.131	-0.018	-0.047	0.059
H2O	0.205	-0.202	0.260	-0.141	0.283	1.000	-0.124	-0.039	-0.009	-0.004	0.024	-0.022	-0.004	-0.168
															S	-0.061	0.084	-0.164	0.266	-0.257	-0.124	1.000	0.147	0.125	0.216	0.197	0.192	0.199	-0.022
L80	0.147	-0.139	-0.121	-0.056	-0.016	-0.039	0.147	1.000	0.826	0.804	0.741	0.587	0.845	-0.077
															L60	0.062	-0.050	0.009	0.042	-0.032	-0.009	0.125	0.826	1.000	0.928	0.810	0.685	0.956	-0.031
L40	0.004	0.011	-0.092	0.112	-0.030	-0.004	0.216	0.804	0.928	1.000	0.886	0.693	0.992	-0.093
															L25	0.065	-0.048	-0.037	0.086	-0.131	0.024	0.197	0.741	0.810	0.886	1.000	0.522	0.909	-0.029
L0	-0.093	0.109	0.027	0.084	-0.018	-0.022	0.192	0.587	0.685	0.693	0.522	1.000	0.708	-0.055
															DAverage	0.032	-0.017	-0.063	0.086	-0.047	-0.004	0.199	0.845	0.956	0.992	0.909	0.708	1.000	-0.071
CFine coke	-0.139	0.112	0.122	0.179	0.059	-0.168	-0.022	-0.077	-0.031	-0.093	-0.029	-0.055	-0.071	1.000

(4) And calculating the eigenvalue and eigenvector of the correlation coefficient matrix R. If the 14 original indexes of coke are directly used for comprehensive evaluation, information is overlapped inevitably, and the objectivity of an evaluation result is influenced. SPSS software is used for carrying out comprehensive characteristic index analysis on the 14 indexes, and the characteristic value and the variance contribution rate of the correlation coefficient matrix R are shown in a table 2.

TABLE 2 extraction of main components of coke Performance index

(5) And determining the number of the comprehensive characteristic indexes. Comprehensively determining the first 5 comprehensive characteristics according to the principle that the cumulative variance contribution rate is more than 75% or the characteristic value is more than 1 and by combining the lithotripsy chart 1Index, using F₁，F₂，…F₅The total of 5 main components, namely the comprehensive characteristic index represents the original 14 indexes to carry out comprehensive analysis on the coke quality. The cumulative variance contribution rate of 5 comprehensive characteristic indexes is 79.029%, which exceeds 75%, and the method has strong explanatory power.

The characteristic value, the total variance interpretation table and the comprehensive characteristic index load matrix (table 3) of the comprehensive characteristic indexes can be obtained through SPSS software, and the coefficients of all the evaluation indexes in different linear combinations of the comprehensive characteristic indexes are determined, namely, the expressions of the first 5 comprehensive characteristic indexes are solved.

TABLE 3 comprehensive characteristic index load matrix K extracted from evaluation of coke properties_ij

The initial load matrix of the comprehensive characteristic index is combined with the characteristic value, so that the eigenvector of the comprehensive characteristic index can be obtained by solving, and the eigenvector is multiplied by the standardized coke quality index data, so that a comprehensive characteristic index expression is obtained:

F₁＝0.011*CRI-0.005*CSR-0.007*M₄₀+0.021*M₁₀-0.003*A_d+0.013*H₂O+0.083*S+0.390*L₈₀+0.427*L₆₀+0.434*L₄₀+0.398*L₂₅+0.339*L₀+0.442*D_average-0.022*C_{Fine coke}

F₂＝0.679*CRI-0.679*CSR+0.053*M₄₀+0.026*M₁₀-0.133*A_d+0.172*H₂O-0.039*S+0.085*L₈₀+0.027*L₆₀-0.010*L₄₀+0.047*L₂₅-0.106*L₀+0.009*D_Average-0.067*C_{Fine coke}

F₃＝0.017*CRI-0.029*CSR+0.058*M₄₀-0.211*M₁₀+0.713*A_d+0.477*H₂O-0.452*S-0.007*L₆₀-0.031*L₄₀-0.079*L₂₅-0.053*L₀-0.034*D_Average+0.012*C_{Fine coke}

F₄＝0.025*CRI-0.037*CSR+0.780*M₄₀-0.513*M₁₀-0.134*A_d+0.157*H₂O-0.244*S-0.055*L₈₀+0.026*L₆₀-0.081*L₄₀-0.015*L₂₅+0.039*L₀-0.044*D_Average+0.103*C_{Fine coke}

F₅＝-0.054*CRI+0.035*CSR+0.161*M₄₀+0.463*M₁₀+0.063*A_d-0.158*H₂O-0.017*S-0.077*L₈₀+0.024*L₆₀-0.022*L₄₀+0.021*L₂₅-0.006*L₀-0.009*D_Average+0.848*C_{Fine coke}

Note: the respective variables are data normalized by the original variables.

L₈₀，L₆₀，L₄₀，L₂₅，L₀And D_AverageWhen 6 indexes have higher load on the 1 st comprehensive characteristic index, namely the absolute value of the coefficient is more than 0.6, the 1 st comprehensive characteristic index basically reflects the information of the indexes, F is₁The method is defined as an index of comprehensive particle size characteristics of the coke, and the index accounts for 36.242% of the total information of the coke mass, and has strong explanatory power. F₁The coefficients of 6 granularity indexes in the expression are all positive, and F is required₁The larger the better.

The 2 nd comprehensive characteristic index basically reflects the information of CRI and CSR indexes, and F is₂Defined as the comprehensive high-temperature characteristic index of coke. The interpretation amount of the index in the total coke mass information is 15.965%, and the index has strong interpretation strength. F₂In the expression, the CRI coefficient is positive, the CSR coefficient is negative, and F is required₂The smaller the better.

The 3 rd comprehensive characteristic index basically reflects A_d，H₂Information of O, S index, F₃Is defined as the index of the comprehensive composition characteristic of the coke. The interpretation amount of the index in the total coke mass information is 11.777%, and the index has strong interpretation strength. F₃Water content in the expression H₂Coefficient of O is positive, ash A_dThe coefficient is positive and the absolute value of the coefficient is maximum, the coefficient of sulfur content S is negative, and F is required₃The smaller the better.

The 4 th comprehensive characteristic index basically reflects M₄₀，M₁₀Information of index, F₄Is defined as the index of the comprehensive cold strength characteristic of coke, F₄In the expression M₄₀Coefficient of (1) is positive, M₁₀Has a negative coefficient of (A), F₄The larger the better.

The 5 th comprehensive characteristic index basically reflects C_{Fine coke}Information of index, F₅Is defined as the comprehensive coke breeze characteristic index of coke, F₅The coefficient of the coke powder ratio in the expression is positive, F₅The smaller the better.

(6) And calculating the weights of the 5 comprehensive characteristic indexes to obtain a coke quality comprehensive evaluation index expression. Establishing a coke quality evaluation method by taking the ratio of the characteristic value corresponding to each comprehensive characteristic index to the sum of the total characteristic values of the extracted comprehensive characteristic indexes as weight, and F according to the requirements of the coke on each comprehensive characteristic index₁And F₄Taking a positive number, F₂，F₃And F₅Taking a negative number.

F＝0.454F₁-0.191F₂-0.135F₃+0.116F₄-0.105F₅

According to the comprehensive evaluation index expression of coke analysis, the daily comprehensive evaluation value F of the coke quality can be calculated, the quality fluctuation of the coke is objectively evaluated and analyzed, and human influence factors are eliminated.

The invention has been described above with reference to the accompanying drawings, it is obvious that the invention is not limited to the specific implementation in the above-described manner, and it is within the scope of the invention to apply the inventive concept and solution to other applications without substantial modification.

Claims (5)

1. The coke quality evaluation method is characterized by comprising the following steps:

s1, establishing an evaluation index system of coke quality;

s2, forming the evaluation indexes of each coke sample in the period into an original data matrix X, and carrying out standardized dimensionless processing on the original data matrix X to obtain a standardized matrix Z;

s3, calculating the correlation coefficient of each evaluation index in the standardized matrix Z, and constructing a correlation coefficient matrix R;

s4, calculating the eigenvalue and the eigenvector of the correlation coefficient matrix R, and constructing a comprehensive characteristic index based on the eigenvector;

s5, determining the first p comprehensive characteristic indexes with large variance contribution rate, and calculating the weight of the first p comprehensive characteristic indexes to obtain the comprehensive evaluation index of coke quality.

2. The coke quality evaluation method according to claim 1, wherein the evaluation index system for coke quality comprises: cold strength index, hot performance index, physical and chemical index, wherein,

the cold strength indicators include: crushing strength M₄₀And abrasion resistance M₁₀；

The thermal state performance indexes include: reactive CRI and post-reaction intensity CSR;

the physical and chemical indexes comprise: ash content A_dMoisture content H₂O, S and V_dafFixed carbon C_adThe ratio L of the particle size of more than 80mm₈₀The particle size is 60-80 mm and the ratio L₆₀The particle size is 40-60 mm₄₀The particle size of 25-40 mm is L₂₅Coke breeze content L₀Average particle size D_AverageAnd powder-to-coke ratio C_{Fine coke}。

3. The coke quality evaluation method of claim 1, wherein the eigenvalues of the correlation coefficient matrix R are sorted from large to small as: lambda [ alpha ]₁≧λ₂…≧λ_j…≧λ_m≧ 0, and corresponding feature vector u₁、u₂…u_j…u_mWherein u is_j＝[u_1j,u_2j,u_3j,…,u_mj]^TThe overall characteristic index is expressed as follows:

F₁＝u₁₁z₁+u₂₁z₂+…+u_j1z_j+…+u_m1z_m

F₂＝u₁₂z₁+u₂₂z₂+…+u_j2z_j+…+u_m2z_m

…

F_j＝u_1jz₁+u_2jz₂+…+u_jjz_j+…+u_mjz_m

…

F_m＝u_1mz₁+u_2mz₂+…+u_jmz_j+…+u_mmz_m

wherein z is₁，z₂…z_j…z_mFor standardized evaluation index, F_mIs the m-th comprehensive characteristic index.

4. The coke quality evaluation method according to claim 2, wherein the overall characteristic index coefficient u is_ijThe expression (c) is specifically as follows:

K_ijis the value of the ith evaluation index in the jth comprehensive characteristic index in the comprehensive characteristic index load matrix, lambda_jAnd representing the characteristic value of the jth comprehensive characteristic index.

5. The coke quality evaluation method according to claim 2, wherein the expression of the coke quality comprehensive evaluation index is as follows:

F＝ω₁F₁+ω₂F₂+…ω_jF_j+…ω_pF_p

ω₁、ω₂…ω_j…ω_pthe weight of each comprehensive characteristic index of the coke is calculated by the following formula:

λ_jis the characteristic value, lambda, of the jth composite characteristic index_kAnd the characteristic value of the kth comprehensive characteristic index is 1 to p.

Images