Method, system, equipment and medium for evaluating coking tendency of coal

Abstract

The invention relates to a method, system, equipment and medium for evaluating the coking tendency of coal types, and the method comprises the following steps: collecting a plurality of coal types with known coking tendency, and obtaining chemical components and physical properties of the coal types; chemical components and physical properties of coal types are combined into a data set for model training, and normalization processing is carried out on the data set; training a random forest model by using the normalized data set; inputting chemical components and physical properties of a coal type to be detected into the random forest model and the ARIMA model, and using an evaluation result of the ARIMA model as an evaluation weight; and adjusting the coking tendency obtained by the random forest model by using the evaluation weight, and taking the adjusted coking tendency as an evaluation result of the coking tendency of the coal type to be detected.

Description

Technical Field

[0001] The invention relates to a method, system, equipment and medium for evaluating the coking tendency of coal types, belonging to the technical field of coal analysis. Background Art

[0002] In the process of thermal power generation, boiler coking is a common problem. Coking refers to the deposition and accumulation of dust and smoke in the flue gas generated during the combustion process on the hot surface inside the boiler. This will cause carbon deposition on the heat exchange surface inside the boiler, thereby reducing the heat transfer efficiency, increasing energy consumption, leading to problems such as instability of the combustion system, equipment damage, and environmental pollution. Therefore, accurately predicting the coking tendency of coal types in advance is of great significance for optimizing coal utilization, reducing the risk of boiler coking, and improving the operating efficiency and economic benefits of power plants. However, there are many existing coal quality coking analysis data, and the analysis technology is single. If it is applied in practice, it requires complex and multiple steps to measure and analyze, resulting in the inability to conduct relevant analysis of coal types in a timely manner, affecting subsequent stacking, coal blending and other related work.

[0003] Random Forest is a classic ensemble learning method that combines the prediction results of multiple decision trees and integrates the prediction results by voting or averaging to improve the accuracy and stability of the model. The ARIMA model, full name Autoregressive Integrated Moving Average Model, is a classic time series forecasting and analysis method.

[0004] Prior art, such as the Chinese patent number "CN118553326A", discloses a domain knowledge-driven interpretable neural network coke hot quality prediction model. The establishment of the model includes the following steps: Step S1: Construction of a matching coal sample data set; Step S2: Construction and prediction of a neural network model, outputting the predicted coke hot quality value based on the numerical characteristics of the matching coal; Step S3: Identification of coal parameter groups, identifying coal parameter groups that do not conform to the coking domain knowledge; Step S4: Calculation of coal parameter group weight scores; Step S5: Optimization of coal parameter group importance, using a weight loss function to optimize the weight scores of the identified coal parameter groups.

[0005] The problem with the above-mentioned prior art is that it is highly dependent on the quality and integrity of the matching coal sample data set. If the data set is biased or incomplete, the prediction accuracy of the neural network model may be reduced. The weight loss function optimization step may be difficult because it is necessary to find a suitable optimization algorithm to minimize the loss function and ensure that the weight scores of the identified coal parameter groups are reasonably adjusted. If the optimization algorithm is not properly selected or the parameters are not set reasonably, the weight scores may not accurately reflect the importance of the coal parameter groups, thereby affecting the accuracy of the prediction results. Summary of the invention

[0006] In order to solve the above problems existing in the prior art, the present invention proposes a method, system, equipment and medium for evaluating the coking tendency of coal.

[0007] The technical solution of the present invention is as follows:

[0008] In one aspect, the present invention provides a method for evaluating the coking tendency of a type of coal, comprising the following steps:

[0009] Collect several types of coal with known coking tendency to obtain the chemical composition and physical properties of the coal; the chemical composition of the coal includes sulfur, ash, iron oxide, aluminum oxide, calcium oxide, magnesium oxide, water, sulfur trioxide, potassium oxide, sodium oxide, silicon oxide and titanium oxide, and the physical property is the melting and softening temperature of coal ash;

[0010] Combining the chemical composition and physical properties of the coal types into a data set for model training, and normalizing the data set;

[0011] Use the normalized dataset to train a random forest model;

[0012] The chemical composition and physical properties of the coal to be tested are input into the random forest model and the ARIMA model, and the evaluation results of the ARIMA model are used as evaluation weights; the coking tendency obtained by the random forest model is adjusted using the evaluation weights, and the adjusted coking tendency is used as the evaluation result of the coking tendency of the coal to be tested.

[0013] As a preferred implementation, the normalization method is:

[0014]

[0015] Among them, x new represents the dataset after normalization, x represents the dataset before normalization, and x min Represents the minimum value of the data set, x max Indicates the maximum value of the data set.

[0016] As a preferred implementation, the random forest model training method is:

[0017] Extract the features of the data set to form a feature set {d1...di...dn}, and use a random sampling method with replacement to select several subsets {x1...xi...xn} from the data set;

[0018] Randomly select several features from the feature set {d1...di...dn} to construct a decision tree Tr;

[0019] Classify the features according to the reduced Gini impurity of the i-th subset xi when the decision tree Tr splits;

[0020] According to the classification results of each decision tree, the category with the most votes is taken as the prediction result of random forest.

[0021] As a preferred implementation, the Gini impurity calculation method is:

[0022]

[0023] Among them, G represents the Gini impurity, p(i) represents the probability that the i-th subset xi belongs to this feature;

[0024] Calculate the Gini impurity reduction of the i-th subset xi when splitting as the Gini impurity reduction, and normalize the Gini impurity reduction using the following formula:

[0025]

[0026] Among them, G k represents the reduction of Gini impurity after normalization, G s represents the reduction in Gini impurity, represents the reduction in total Gini impurity;

[0027] The normalized Gini impurity reduction is used as the feature score of the i-th subset xi.

[0028] As a preferred embodiment, the method of using the evaluation weight to adjust the coking tendency obtained by the random forest model is:

[0029]

[0030] Among them, W represents the final evaluation result, U represents the evaluation result of the random forest model, Z represents the evaluation weight of the ARIMA model, and F represents a random number in the interval (0, 1).

[0031] On the other hand, the present invention also provides a system for evaluating the coking tendency of coal, comprising:

[0032] Data acquisition module: collects several types of coal with known coking tendency to obtain the chemical composition and physical properties of the coal; the chemical composition of the coal includes sulfur, ash, iron oxide, aluminum oxide, calcium oxide, magnesium oxide, moisture, sulfur trioxide, potassium oxide, sodium oxide, silicon oxide and titanium oxide, and the physical property is the melting and softening temperature of coal ash;

[0033] Data processing module: combining the chemical composition and physical properties of coal types into a data set for model training, and normalizing the data set;

[0034] Model training module: Use the normalized data set to train the random forest model;

[0035] Result evaluation module: the chemical composition and physical properties of the coal to be tested are input into the random forest model and the ARIMA model, and the evaluation results of the ARIMA model are used as evaluation weights; the coking tendency obtained by the random forest model is adjusted using the evaluation weights, and the adjusted coking tendency is used as the evaluation result of the coking tendency of the coal to be tested.

[0036] On the other hand, the present invention further provides an electronic device having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for evaluating the coking tendency of a type of coal as described in any embodiment of the present invention.

[0037] On the other hand, the present invention also provides a computer-readable storage medium for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method for evaluating the coking tendency of coal types as described in any embodiment of the present invention.

[0038] The present invention has the following beneficial effects:

[0039] The present invention not only takes into account the chemical composition of coal (such as sulfur, ash, various oxides, etc.), but also combines physical properties (such as coal ash melting softening temperature). This comprehensive consideration makes the evaluation of the coking tendency of coal types more comprehensive and accurate. By normalizing the data set, the dimensional differences between different features are eliminated, and the efficiency and accuracy of model training are improved. As an integrated learning method, the random forest model has strong learning and generalization capabilities, can handle high-dimensional data and complex nonlinear relationships, thereby improving the prediction accuracy of the coking tendency of coal types. By introducing the ARIMA model, the prediction results of the random forest model are auxiliary evaluated, and the evaluation results of the ARIMA model are used as adjustment weights to further improve the reliability and stability of the prediction results. By combining the prediction results of the random forest model and the ARIMA model, the deviation that may be caused by a single model is avoided, making the evaluation results more objective and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 The present invention is a flowchart of the method for implementing the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] It should be understood that the step numbers used in this document are only for convenience of description and are not intended to limit the order in which the steps are executed.

[0043] It should be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0044] The terms “include” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0045] The term "and / or" means and includes any and all possible combinations of one or more of the associated listed items.

[0046] Embodiment 1:

[0047] See also Figure 1 The present invention provides a method for evaluating the coking tendency of coal, comprising the following steps:

[0048] Collect several types of coal with known coking tendency to obtain the chemical composition and physical properties of the coal; the chemical composition of the coal includes sulfur, ash, iron oxide, aluminum oxide, calcium oxide, magnesium oxide, water, sulfur trioxide, potassium oxide, sodium oxide, silicon oxide and titanium oxide, and the physical property is the melting and softening temperature of coal ash;

[0049] Combining the chemical composition and physical properties of the coal types into a data set for model training, and normalizing the data set;

[0050] Use the normalized dataset to train a random forest model;

[0051] The chemical composition and physical properties of the coal to be tested are input into the random forest model and the ARIMA model, and the evaluation results of the ARIMA model are used as evaluation weights; the coking tendency obtained by the random forest model is adjusted using the evaluation weights, and the adjusted coking tendency is used as the evaluation result of the coking tendency of the coal to be tested.

[0052] As a preferred implementation, the normalization method is:

[0053]

[0054] Among them, x new represents the dataset after normalization, x represents the dataset before normalization, and x min Represents the minimum value of the data set, x max Indicates the maximum value of the data set.

[0055] By normalizing the data set, the dimensional differences between different features are eliminated, and the efficiency and accuracy of model training are improved.

[0056] As a preferred implementation, the random forest model training method is:

[0057] Extract the features of the data set to form a feature set {d1...di...dn}, and use a random sampling method with replacement to select several subsets {x1...xi...xn} from the data set;

[0058] Randomly select several features from the feature set {d1...di...dn} to construct a decision tree Tr;

[0059] Classify the features according to the reduced Gini impurity of the i-th subset xi when the decision tree Tr splits;

[0060] According to the classification results of each decision tree, the category with the most votes is taken as the prediction result of random forest.

[0061] As an integrated learning method, the random forest model has strong learning and generalization capabilities and can handle high-dimensional data and complex nonlinear relationships, thereby improving the prediction accuracy of the coking tendency of coal types.

[0062] As a preferred implementation, the Gini impurity calculation method is:

[0063]

[0064] Among them, G represents the Gini impurity, p(i) represents the probability that the i-th subset xi belongs to this feature;

[0065] Calculate the Gini impurity reduction of the i-th subset xi when splitting as the Gini impurity reduction, and normalize the Gini impurity reduction using the following formula:

[0066]

[0067] Among them, G k represents the reduction of Gini impurity after normalization, G s represents the reduction in Gini impurity, represents the reduction in total Gini impurity;

[0068] The normalized Gini impurity reduction is used as the feature score of the i-th subset xi.

[0069] Although the random forest model itself may be complex, by looking at the split nodes of each decision tree, we can still understand which features have an important impact on the prediction results, thereby enhancing the interpretability of the model.

[0070] As a preferred embodiment, the method of using the evaluation weight to adjust the coking tendency obtained by the random forest model is:

[0071]

[0072] Among them, W represents the final evaluation result, U represents the evaluation result of the random forest model, Z represents the evaluation weight of the ARIMA model, and F represents a random number in the interval (0, 1).

[0073] By combining the prediction results of the random forest model and the ARIMA model, the bias that may be caused by a single model is avoided, making the evaluation results more objective and accurate.

[0074] Embodiment 2:

[0075] The present invention also provides a system for evaluating the coking tendency of coal types, comprising:

[0076] Data acquisition module: collects several types of coal with known coking tendency to obtain the chemical composition and physical properties of the coal; the chemical composition of the coal includes sulfur, ash, iron oxide, aluminum oxide, calcium oxide, magnesium oxide, moisture, sulfur trioxide, potassium oxide, sodium oxide, silicon oxide and titanium oxide, and the physical property is the melting and softening temperature of coal ash;

[0077] Data processing module: combining the chemical composition and physical properties of coal types into a data set for model training, and normalizing the data set;

[0078] Model training module: Use the normalized data set to train the random forest model;

[0079] Result evaluation module: the chemical composition and physical properties of the coal to be tested are input into the random forest model and the ARIMA model, and the evaluation results of the ARIMA model are used as evaluation weights; the coking tendency obtained by the random forest model is adjusted using the evaluation weights, and the adjusted coking tendency is used as the evaluation result of the coking tendency of the coal to be tested.

[0080] Embodiment three:

[0081] This embodiment provides an electronic device having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for evaluating the coking tendency of a type of coal as described in any embodiment of the present invention is implemented.

[0082] Embodiment 4:

[0083] This embodiment provides a computer-readable storage medium for storing one or more programs. When the one or more programs are executed by the one or more processors, the one or more processors implement the method for evaluating the coking tendency of coal types as described in any embodiment of the present invention.

[0084] Embodiment five:

[0085] The melting points of various components in coal ash are different, and their effects on coking are also different. Among them, iron and calcium will enhance coking. In a reducing atmosphere, Fe2O3 in the ash will be reduced to FeO, which will lower the ash melting point and enhance coking. Calcium can significantly reduce the viscosity of silicate glass, potassium is a co-solvent that promotes the formation of glass, and silicon can generally reduce coking, but too high a silicon content will produce amorphous glass, which will increase coking. AL2O3 in the ash can reduce coking. At present, the alkali-acid ratio can be used to determine the coking tendency, the silicon-aluminum ratio can be used to determine the coking, and the iron-calcium ratio can be used as the coking characteristic of coal.

[0086] (1) Acid-base ratio

[0087] In the process of boiler coking, the acid-base ratio is an important indicator used to describe the balance ratio of acidic elements (such as sulfur, chlorine, fluorine, etc.) and alkaline elements (such as calcium, magnesium, potassium, etc.) in coal ash. The acid-base ratio has an important influence on the degree of boiler coking and the properties of coking materials.

[0088] The ash produced during boiler combustion contains acidic elements and alkaline elements. Acidic elements act as coking aids and easily form liquid compounds at high temperatures, promoting the formation of coking materials. Alkaline elements, on the other hand, have an anti-coking effect and can neutralize acidic elements and reduce the formation of coking materials.

[0089] An appropriate acid-base ratio can reduce the formation of coking materials, reduce the thickness and adhesion of the coking layer, improve the thermal conductivity of the boiler, and reduce the risk of damage and blockage to equipment caused by coking.

[0090] In addition, the acid-base ratio can also affect the properties and composition of coking materials. For high-acid ash, high-temperature coking materials are generally formed, which have a high melting point and adhesion; while for high-alkalinity ash, low-temperature coking materials are generally formed, which have a low melting point and are easy to fall off and clean. Therefore, by adjusting the acid-base ratio in the coal quality, the properties of the coking materials can be controlled, thereby affecting the characteristics of the coking layer.

[0091]

[0092] When the alkali-acid ratio is >0.4, this type of coal has a greater tendency to coke.

[0093] (2) Silicon to aluminum ratio

[0094] In the process of boiler coking, the silicon-aluminum ratio is an important indicator used to describe the relative content of silicon and aluminum in coal. It has an important impact on boiler operation and combustion effect. They will form coal ash during the combustion process and accumulate in the boiler. When the silicon-aluminum ratio is high, it means that there is relatively more silicon in the coal, which is easy to form high-silicon ash; conversely, when the silicon-aluminum ratio is low, there is relatively more aluminum in the coal, which is easy to form high-aluminum ash.

[0095] During the coking process, silicon and aluminum in the ash will form a liquid compound at high temperature, also known as a molten material. This molten material will adhere to the surface of the heat exchange tubes and components when the boiler is cooled, forming a coking layer. The formation of the coking layer will reduce the heat conduction of the boiler, hinder the circulation, affect the thermal efficiency and flue gas flow, and even cause equipment damage and unstable operation.

[0096]

[0097] When the silicon-aluminum ratio is greater than 1.18, the coal will coke. At present, the silicon-aluminum ratio of coal supplied to power plants is greater than 1.18, which will inevitably cause coking.

[0098] (3) Iron-calcium ratio

[0099] The iron-calcium ratio (Fe / Ca ratio) refers to the ratio of the iron (Fe) to calcium (Ca) content in the flue gas during the combustion process. Under normal circumstances, a reasonable iron-calcium ratio helps prevent the occurrence of coking. A high iron-calcium ratio can promote the discharge of ash after combustion in solid form, reducing the tendency to coke. However, an excessively high iron-calcium ratio may increase the softening temperature of the slag, making it more difficult to discharge, thereby promoting the occurrence of coking.

[0100]

[0101] When the iron-calcium ratio is >3.0, the coal is prone to coking.

[0102] (4) Fusion characteristics of coal ash

[0103] The melting characteristics of coal ash are used to determine the tendency of coking. The initial deformation temperature IT, softening temperature ST and flow temperature FT are usually used to indicate the melting characteristics of coal ash. The general rule is that the lower these characteristic temperatures are, the greater the risk of coking.

[0104] The IT temperature is used to judge the coking property of coal, and the limits are:

[0105]

[0106] It is expressed by coke residue index:

[0107]

[0108] Among them: Fe 2 O 3 (iron oxide), etc. – the mass share of each component in the fuel ash;

[0109] S d ——Sulfur content of fuel on a dry basis.

[0110] Coke slag index R jz The relationship with the coking degree is shown in Table 1.

[0111] Table 1 Coke slag index and coking degree

[0112]

[0113] (5) Analysis of coking mechanism during coal combustion

[0114] Coal quality factors: Coal is a complex organic matter, mainly composed of carbon, hydrogen, oxygen, nitrogen, sulfur and other elements. The inorganic matter (ash) contained in it will be converted into the main source of coke residue during the combustion process. Different types and varieties of coal have different coking tendencies. High-coking coal and lignite contain more unstable carbon residue and are easy to produce coke; while anthracite and stone coal contain less unstable carbon residue and are not easy to coke.

[0115] Heating rate: The rate at which the temperature of coal particles rises during heating has a great influence on the tendency to coke. The faster the heating rate, the higher the degree of coking.

[0116] Temperature and time: The longer the coal particles stay at high temperature, the higher the coking degree. The higher the temperature, the faster the coking rate.

[0117] Combustion conditions: The temperature, oxygen supply and combustion rate during the combustion process have an important influence on the formation of coke slag. The organic matter in the coal decomposes at high temperature to produce solid residues, which contain a large amount of carbon. When the temperature of the coal surface rises to about

[0118] At temperatures above 1000°C, organic matter undergoes pyrolysis, gasification, and re-aggregation to form droplets called tar. Tar tends to aggregate between coal particles and in pores due to surface tension, forming a viscous, gelatinous substance called coke.

[0119] Combustion process: During the combustion process, the organic matter in the coal first undergoes pyrolysis to produce gas and coke. The coke further burns at high temperature to produce carbon dioxide and water vapor. When iron-containing coal burns in the absence of oxygen, FeO will appear as an intermediate product. FeO and FeS form a very easy to melt eutectic, and its flow temperature is much lower than the flow temperature of single FeO and FeS. Coking caused by calcium mainly occurs in lignite furnaces, and other alkaline components also promote coking. According to experience, the result is that coal contains more Na 2 SO 4 When NaCl is present, there is a serious tendency to coking.

[0120] Embodiment six:

[0121] Table 2

[0122]

[0123]

[0124] Referring to Table 2, the chemical composition and physical properties of the coal sample are analyzed. This embodiment provides an algorithm for the coking tendency of coal types. By inputting data such as the sulfur content of 1.74, ash content of 30.86, Fe2O3 (iron oxide) of 6.52, AL2O3 (aluminum oxide) of 28.10, CaO (calcium oxide) of 3.02, MgO (magnesium oxide) of 0.75, moisture content of 1.00, SO3 (sulfur trioxide) of 0.85, K2O (potassium oxide) of 1.50, NA2O (sodium oxide) of 0.32, SiO2 (silicon oxide) of 54.82, TiO2 (titanium oxide) of 1.04, and ST (ash melting and softening temperature) of 1470 of a coal sample into the algorithm model, the model can quickly and accurately predict in advance that the coal type has a coking tendency and is a type of coal that is easy to coke based on the input chemical composition and physical property data of the coal sample.

[0125] In the embodiments of the present application, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. Among them, A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, c can be single or multiple.

[0126] Those of ordinary skill in the art will appreciate that the various units and algorithm steps described in the embodiments disclosed herein can be implemented in a combination of electronic hardware, computer software, and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0127] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0128] In several embodiments provided in the present application, if any function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory; hereinafter referred to as: ROM), random access memory (Random Access Memory; hereinafter referred to as: RAM), disk or optical disk, and other media that can store program codes.

[0129] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for evaluating the coking tendency of coal, characterized in that: The following steps are involved: Collect several types of coal with known coking tendency to obtain the chemical composition and physical properties of the coal; the chemical composition of the coal includes sulfur, ash, iron oxide, aluminum oxide, calcium oxide, magnesium oxide, water, sulfur trioxide, potassium oxide, sodium oxide, silicon oxide and titanium oxide, and the physical property is the melting and softening temperature of coal ash; Combining the chemical composition and physical properties of the coal types into a data set for model training, and normalizing the data set; Use the normalized dataset to train a random forest model; The chemical composition and physical properties of the coal to be tested are input into the random forest model and the ARIMA model, and the evaluation results of the ARIMA model are used as evaluation weights; the coking tendency obtained by the random forest model is adjusted using the evaluation weights, and the adjusted coking tendency is used as the evaluation result of the coking tendency of the coal to be tested.

2. The method for evaluating the coking tendency of coal according to claim 1, characterized in that: The normalization method is: Among them, x new represents the dataset after normalization, x represents the dataset before normalization, and x min Represents the minimum value of the data set, x max Indicates the maximum value of the data set.

3. The method for evaluating the coking tendency of coal according to claim 1, characterized in that: The random forest model training method is: Extract the features of the data set to form a feature set {d1...di...dn}, and use a random sampling method with replacement to select several subsets {x1...xi...xn} from the data set; Randomly select several features from the feature set {d1…di…dn} to construct a decision tree Tr; Classify the features according to the reduced Gini impurity of the i-th subset xi when the decision tree Tr splits; According to the classification results of each decision tree, the category with the most votes is taken as the prediction result of random forest.

4. The method for evaluating the coking tendency of coal according to claim 3, characterized in that: The Gini impurity calculation method is: Among them, G represents the Gini impurity, p(i) represents the probability that the i-th subset xi belongs to this feature; Calculate the Gini impurity reduction of the i-th subset xi when splitting as the Gini impurity reduction, and normalize the Gini impurity reduction using the following formula: Among them, G k represents the reduction of Gini impurity after normalization, G s represents the reduction in Gini impurity, represents the reduction in total Gini impurity; The normalized Gini impurity reduction is used as the feature score of the i-th subset xi.

5. The method for evaluating the coking tendency of coal according to claim 1, characterized in that: The method for adjusting the coking tendency obtained by the random forest model using the evaluation weights is: Among them, W represents the final evaluation result, U represents the evaluation result of the random forest model, Z represents the evaluation weight of the ARIMA model, and F represents a random number in the interval (0, 1).

6. A system for evaluating the coking tendency of coal, characterized in that: include: Data acquisition module: collects several types of coal with known coking tendency to obtain the chemical composition and physical properties of the coal; the chemical composition of the coal includes sulfur, ash, iron oxide, aluminum oxide, calcium oxide, magnesium oxide, moisture, sulfur trioxide, potassium oxide, sodium oxide, silicon oxide and titanium oxide, and the physical property is the melting and softening temperature of coal ash; Data processing module: combining the chemical composition and physical properties of coal types into a data set for model training, and normalizing the data set; Model training module: Use the normalized data set to train the random forest model; Result evaluation module: the chemical composition and physical properties of the coal to be tested are input into the random forest model and the ARIMA model, and the evaluation results of the ARIMA model are used as evaluation weights; the coking tendency obtained by the random forest model is adjusted using the evaluation weights, and the adjusted coking tendency is used as the evaluation result of the coking tendency of the coal to be tested.

7. The system for evaluating the coking tendency of coal according to claim 6, characterized in that: The model training module, the random forest model training method is: Extract the features of the data set to form a feature set {d1...di...dn}, and use a random sampling method with replacement to select several subsets {x1...xi...xn} from the data set; Randomly select several features from the feature set {d1...di...dn} to construct a decision tree Tr; Classify the features according to the reduced Gini impurity of the i-th subset xi when the decision tree Tr splits; According to the classification results of each decision tree, the category with the most votes is taken as the prediction result of random forest.

8. The system for evaluating the coking tendency of coal according to claim 6, characterized in that: The method for the result evaluation module to adjust the coking tendency obtained by the random forest model using the evaluation weight is: Among them, W represents the final evaluation result, U represents the evaluation result of the random forest model, Z represents the evaluation weight of the ARIMA model, and F represents a random number in the interval (0, 1).

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method for evaluating the coking tendency of a type of coal as described in any one of claims 1 to 5 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for evaluating the coking tendency of a type of coal as described in any one of claims 1 to 5 is implemented.

Images