A Carbon Sink Data Analysis Method and System for Coal Mining Subsidence Areas

Abstract

The present invention relates to a carbon sink data analysis method and system for coal mining subsidence areas, belonging to the field of carbon sink data processing, including: analyzing the carbon sink influence trend according to the subsidence characteristic information to obtain the carbon sink influence coefficients of soil, vegetation and water body; collecting the main carbon sink body characteristic information to conduct an auxiliary carbon sink trend influence analysis on the main carbon sink method to obtain the auxiliary carbon sink influence coefficient, and combining the carbon sink influence coefficients of soil, vegetation and water body to correct the influence on the basic carbon sink parameters of the main carbon sink method and two secondary carbon sink methods to obtain the carbon sink parameters of soil, vegetation and water body. Through the present invention, the technical problem that the traditional method fails to fully consider the special geographical environment of coal mining subsidence areas and the mutual influence between carbon sink methods, resulting in insufficient accuracy and comprehensiveness of carbon sink data analysis can be solved; the scientificity, accuracy and comprehensiveness of carbon sink data analysis can be improved, and the effect of accurately evaluating the carbon sink capacity of coal mining subsidence areas under different geographical conditions can be achieved.

Description

Technical Field

[0001] The present invention relates to the field of carbon sink data processing, and particularly to a carbon sink data analysis method and system for coal mining subsidence areas. Background Art

[0002] A carbon sink refers to a system in nature that can absorb and store carbon dioxide, including soil, vegetation, water bodies, etc. As a typical type of land degradation, coal mining subsidence areas usually have the characteristics of strong destructiveness and difficult ecological restoration. The phenomena such as surface subsidence, waterlogging, and soil degradation during coal mining seriously affect the ecological functions of the land, especially the carbon sink function.

[0003] Traditional carbon sink assessment methods are mainly based on standardized climate and soil conditions, and fail to fully consider the influence of the special environment of coal mining subsidence areas. The geographical environment of coal mining subsidence areas is complex, and factors such as the depth, slope, and waterlogging degree of surface subsidence directly affect the carbon sink capabilities of soil, vegetation, and water bodies. In addition, carbon sink methods do not exist in isolation, but are intertwined and interact with each other. For example, changes in soil composition will directly affect the growth of vegetation, and the growth of vegetation will in turn affect the carbon storage capacity of the soil. There is also a correlation between the organic carbon concentration in water bodies and the carbon storage in soil. Traditional carbon sink data analysis methods often ignore these interactions and dynamic changes, resulting in insufficient accuracy and comprehensiveness in the carbon sink data analysis of coal mining subsidence areas. Summary of the Invention

[0004] In view of the technical problem that traditional carbon sink data analysis methods fail to fully consider the special geographical environment of coal mining subsidence areas and the mutual influence between carbon sink methods, resulting in insufficient accuracy and comprehensiveness in the carbon sink data analysis of coal mining subsidence areas, the present invention provides a carbon sink data analysis method and system for coal mining subsidence areas to solve this problem.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a method for analyzing carbon sink data of coal mining subsidence areas, including: collecting subsidence characteristic information of coal mining subsidence areas, classifying to obtain soil carbon sink contribution coefficients, vegetation carbon sink contribution coefficients, and water body carbon sink contribution coefficients, and determining the main carbon sink method and the other two secondary carbon sink methods; collecting soil carbon sink data, vegetation carbon sink data, and water body carbon sink data within the coal mining subsidence areas to obtain basic soil carbon sink parameters, basic vegetation carbon sink parameters, and basic water body carbon sink parameters; according to the subsidence characteristic information, conducting carbon sink influence trend analysis to obtain soil carbon sink influence coefficients, vegetation carbon sink influence coefficients, and water body carbon sink influence coefficients; collecting carbon sink body characteristic information of the two secondary carbon sink methods and the main carbon sink method, conducting auxiliary carbon sink trend influence analysis on the main carbon sink method to obtain auxiliary carbon sink influence coefficients, and combining the soil carbon sink influence coefficients, vegetation carbon sink influence coefficients, and water body carbon sink influence coefficients to correct the influence on the basic carbon sink parameters of the main carbon sink method and the two secondary carbon sink methods to obtain soil carbon sink parameters, vegetation carbon sink parameters, and water body carbon sink parameters as the results of carbon sink data analysis.

[0007] Optionally, the method for analyzing carbon sink data of coal mining subsidence areas further includes: collecting the subsidence slope and subsidence depth of the coal mining subsidence area as subsidence characteristic information; constructing a carbon sink contribution classifier according to the sample carbon sink data of the sample coal mining subsidence area; inputting the subsidence characteristic information into the carbon sink contribution classifier, and outputting to obtain the soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient, and water body carbon sink contribution coefficient of soil carbon sink, vegetation carbon sink, and water body carbon sink; selecting the carbon sink method corresponding to the largest contribution coefficient as the main carbon sink method, and taking the other two carbon sink methods as secondary carbon sink methods.

[0008] Optionally, the method for analyzing carbon sink data of coal mining subsidence areas further includes: collecting the subsidence characteristic information of multiple sample coal mining subsidence areas to obtain a subsidence characteristic information set; collecting the proportion of the carbon sink amounts of soil carbon sink, vegetation carbon sink, and water body carbon sink within each sample coal mining subsidence area, and labeling to obtain a sample soil carbon sink contribution coefficient set, a sample vegetation carbon sink contribution coefficient set, and a sample water body carbon sink contribution coefficient set; using the subsidence characteristic information set as input features, and using the sample soil carbon sink contribution coefficient set, the sample vegetation carbon sink contribution coefficient set, and the sample water body carbon sink contribution coefficient set as output features, and based on machine learning, constructing and training a carbon sink contribution classifier until convergence.

[0009] Optionally, the method for analyzing carbon sink data of coal mining subsidence areas further includes: testing the soil organic carbon content within the coal mining subsidence area to obtain soil carbon sink data; collecting the quantities of different vegetation within the coal mining subsidence area and calculating to obtain vegetation carbon sink data; testing the dissolved organic carbon concentration of the water body within the coal mining subsidence area to obtain water body carbon sink data.

[0010] Optionally, the carbon sink data analysis method for coal mining subsidence areas further includes: collecting the subsidence characteristic information of multiple sample coal mining subsidence areas to obtain a subsidence characteristic information set; collecting the change ratios of the soil carbon sink parameters, vegetation carbon sink parameters, and water body carbon sink parameters under different sample coal mining subsidence areas within a preset time length to obtain a sample soil carbon sink influence coefficient set, a sample vegetation carbon sink influence coefficient set, and a sample water body carbon sink influence coefficient set; using the subsidence characteristic information set as input features, using the sample soil carbon sink influence coefficient set, the sample vegetation carbon sink influence coefficient set, and the sample water body carbon sink influence coefficient set as output features, and based on machine learning, constructing and training a carbon sink influence trend analyzer until convergence; inputting the subsidence characteristic information into the carbon sink influence trend analyzer, and outputting to obtain a soil carbon sink influence coefficient, a vegetation carbon sink influence coefficient, and a water body carbon sink influence coefficient.

[0011] Optionally, the carbon sink data analysis method for coal mining subsidence areas further includes: collecting the carbon sink body characteristic information of the two secondary carbon sink methods and the primary carbon sink method, wherein the carbon sink body characteristic information of soil carbon sink, vegetation carbon sink, and water body carbon sink includes soil component information, vegetation type information, and water body component information; collecting a sample carbon sink body characteristic information set, and collecting the change ratios of the carbon sink parameters of other carbon sink methods under each sample carbon sink body characteristic information within a preset time length, and labeling them as a sample auxiliary carbon sink influence coefficient set; using the sample carbon sink body characteristic information set as input features, using the sample auxiliary carbon sink influence coefficient set as output features, constructing an auxiliary carbon sink influence analyzer and training it until convergence; inputting the carbon sink body characteristic information of the two secondary carbon sink methods into the auxiliary carbon sink influence analyzer respectively, and outputting to obtain two auxiliary carbon sink influence coefficients.

[0012] Optionally, the carbon sink data analysis method for coal mining subsidence areas further includes: calculating to obtain three total carbon sink influence coefficients according to the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient, in combination with the two auxiliary carbon sink influence coefficients; using the three total carbon sink influence coefficients to perform influence correction on the basic carbon sink parameters of the primary carbon sink method and the two secondary carbon sink methods to obtain soil carbon sink parameters, vegetation carbon sink parameters, and water body carbon sink parameters as the carbon sink data analysis results.

[0013] In a second aspect, the present invention provides a carbon sink data analysis system for coal mining subsidence areas, including: a carbon sink contribution coefficient obtaining module, configured to collect the subsidence characteristic information of the coal mining subsidence area, classify and obtain the soil carbon sink contribution coefficient, the vegetation carbon sink contribution coefficient, and the water body carbon sink contribution coefficient, and determine the main carbon sink mode and the other two secondary carbon sink modes; a basic carbon sink parameter obtaining module, configured to collect the soil carbon sink data, the vegetation carbon sink data, and the water body carbon sink data in the coal mining subsidence area, and obtain the basic soil carbon sink parameter, the basic vegetation carbon sink parameter, and the basic water body carbon sink parameter; a carbon sink influence trend analysis module, configured to perform a carbon sink influence trend analysis according to the subsidence characteristic information, and obtain the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient; a basic carbon sink parameter correction module, configured to collect the carbon sink main body characteristic information of the two secondary carbon sink modes and the main carbon sink mode, perform an auxiliary carbon sink trend influence analysis on the main carbon sink mode, obtain the auxiliary carbon sink influence coefficient, and combine the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient to perform an influence correction on the basic carbon sink parameters of the main carbon sink mode and the two secondary carbon sink modes, and obtain the soil carbon sink parameter, the vegetation carbon sink parameter, and the water body carbon sink parameter as the carbon sink data analysis result.

[0014] The beneficial effects of the present invention are as follows: by collecting the subsidence characteristic information of the coal mining subsidence area, classifying and obtaining the soil carbon sink contribution coefficient, the vegetation carbon sink contribution coefficient, and the water body carbon sink contribution coefficient, and determining the main carbon sink mode and the other two secondary carbon sink modes; then collecting the soil carbon sink data, the vegetation carbon sink data, and the water body carbon sink data in the coal mining subsidence area, and obtaining the basic soil carbon sink parameter, the basic vegetation carbon sink parameter, and the basic water body carbon sink parameter; then performing a carbon sink influence trend analysis according to the subsidence characteristic information, and obtaining the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient; on the other hand, collecting the carbon sink main body characteristic information of the two secondary carbon sink modes and the main carbon sink mode, performing an auxiliary carbon sink trend influence analysis on the main carbon sink mode, and obtaining the auxiliary carbon sink influence coefficient; finally, combining the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient to perform an influence correction on the basic carbon sink parameters of the main carbon sink mode and the two secondary carbon sink modes, and obtaining the soil carbon sink parameter, the vegetation carbon sink parameter, and the water body carbon sink parameter as the carbon sink data analysis result; that is to say, by using machine learning algorithms to accurately analyze the subsidence characteristics of coal mining subsidence areas and the mutual influence between different carbon sink modes, the scientificity, accuracy, and comprehensiveness of carbon sink data analysis in coal mining subsidence areas can be improved, thereby achieving the technical effect of accurately evaluating the carbon sink capacity of coal mining subsidence areas under different geographical conditions. Description of the Drawings

[0015] Figure 1 is a schematic flow chart of a carbon sink data analysis method for coal mining subsidence areas provided by the present invention;

[0016] Figure 2 This is a schematic structural diagram of a carbon sink data analysis system for coal mining subsidence areas provided by the present invention.

[0017] In the accompanying drawings, the components represented by each reference numeral are described as follows:

[0018] Carbon sink contribution coefficient acquisition module 11, basic carbon sink parameter acquisition module 12, carbon sink influence trend analysis module 13, basic carbon sink parameter correction module 14. Specific implementation manners

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts shall fall within the protection scope of the present invention.

[0020] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0021] In the description of the present invention, the term "for example" is used to mean "serving as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for purposes of explanation. It should be understood that those skilled in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present invention.

[0022] Embodiment 1, as Figure 1 shown, the embodiment of the present invention provides a carbon sink data analysis method for coal mining subsidence areas, including:

[0023] S100: Collect the subsidence characteristic information of the coal mining subsidence area, classify and obtain the soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient, and water body carbon sink contribution coefficient, and determine the main carbon sink mode and the other two secondary carbon sink modes.

[0024] Furthermore, step S100 of the present invention further includes:

[0025] S110: Collect the subsidence slope and subsidence depth of the coal mining subsidence area as subsidence characteristic information.

[0026] Specifically, collect characteristic parameters such as the phreatic level, subsidence slope and subsidence depth of the coal mining subsidence area as subsidence characteristic information. Preferably, collect the subsidence slope and subsidence depth of the coal mining subsidence area as subsidence characteristic information.

[0027] Among them, the subsidence slope refers to the inclination angle of the subsidence ground surface, which reflects the severity of ground settlement. A larger slope (such as a subsidence slope of 35° - 40°) usually means more serious surface subsidence, and soil erosion is likely to occur on the soil surface, making it difficult for vegetation to grow. An overly large slope may also exacerbate the water accumulation problem, thereby affecting the water body carbon sequestration capacity. The subsidence depth refers to the surface settlement depth caused by coal mining activities, which reflects the degree and scope of land settlement. In areas with a relatively large subsidence depth (such as a subsidence depth of 8m - 12m), the physical properties of the soil may change, affecting the soil carbon storage capacity; in deep subsidence areas, water accumulation may occur, leading to an increase in the impact on water body carbon sequestration, but the persistence and stability of water body carbon sequestration still need to be considered; at the same time, the vegetation restoration in deep subsidence areas faces great challenges, and the vegetation carbon sequestration capacity is limited.

[0028] By collecting the characteristic information of the subsidence slope and depth, it will help to accurately analyze the carbon sequestration characteristics of different coal mining subsidence areas, understand the mutual influence between various carbon sequestration methods (such as soil carbon sequestration, vegetation carbon sequestration, water body carbon sequestration), and thus provide key data support for the assessment of the carbon sequestration capacity of coal mining subsidence areas.

[0029] S120: Construct a carbon sequestration contribution classifier according to the sample carbon sequestration data of the sample coal mining subsidence area.

[0030] Furthermore, step S120 of the present invention further includes:

[0031] S121: Collect the subsidence characteristic information of multiple sample coal mining subsidence areas to obtain a subsidence characteristic information set; S122: Collect the proportion of the carbon sequestration amounts of soil carbon sequestration, vegetation carbon sequestration, and water body carbon sequestration in each sample coal mining subsidence area, and label to obtain a sample soil carbon sequestration contribution coefficient set, a sample vegetation carbon sequestration contribution coefficient set, and a sample water body carbon sequestration contribution coefficient set; S123: Use the subsidence characteristic information set as input features, and use the sample soil carbon sequestration contribution coefficient set, the sample vegetation carbon sequestration contribution coefficient set, and the sample water body carbon sequestration contribution coefficient set as output features, and construct and train a carbon sequestration contribution classifier based on machine learning until convergence.

[0032] Specifically, first, collect the subsidence characteristic information (phreatic level, subsidence slope, and subsidence depth) of multiple sample coal mining subsidence areas to obtain a set of subsidence characteristic information. Then, collect the proportion of the carbon sink amounts of soil carbon sink, vegetation carbon sink, and water body carbon sink in each sample coal mining subsidence area, that is, the ratio of the carbon sink amount of a single carbon sink method to the total carbon sink amount, and label the proportion of the carbon sink amount as the carbon sink contribution coefficient. For example, in a certain coal mining subsidence area, the soil carbon sink accounts for 60% of the total carbon sink amount, the vegetation carbon sink accounts for 30%, and the water body carbon sink accounts for 10%. Then, the soil carbon sink contribution coefficient is 0.6, the vegetation carbon sink contribution coefficient is 0.3, and the water body carbon sink contribution coefficient is 0.1. Obtain a set of sample soil carbon sink contribution coefficients, a set of sample vegetation carbon sink contribution coefficients, and a set of sample water body carbon sink contribution coefficients. Among them, there is a corresponding relationship between the subsidence characteristic information and the sample soil carbon sink contribution coefficient, the sample vegetation carbon sink contribution coefficient, and the sample water body carbon sink contribution coefficient.

[0033] Build a carbon sink contribution classifier based on a machine learning algorithm. In the process of building the carbon sink contribution classifier, the task of machine learning is to analyze the relationship between the collected subsidence characteristic information and the corresponding carbon sink contribution coefficients, extract patterns from them, and use them to predict the carbon sink contribution coefficients of new data. For example, build a carbon sink contribution classifier based on a BP neural network. The carbon sink contribution classifier is a BP neural network model that can be iteratively optimized in machine learning, including an input layer (the input data is the subsidence characteristic information), multiple hidden layers, and an output layer (the output data is the soil carbon sink contribution coefficient, the vegetation carbon sink contribution coefficient, and the water body carbon sink contribution coefficient). Further, use the set of subsidence characteristic information as input features, and use the set of sample soil carbon sink contribution coefficients, the set of sample vegetation carbon sink contribution coefficients, and the set of sample water body carbon sink contribution coefficients as output features to perform supervised training on the carbon sink contribution classifier. First, pass the input data (subsidence characteristic information) layer by layer into the network. After the non-linear transformation of the activation function, the calculation result of the output layer is the predicted value of the input features, and the predicted soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient, and water body carbon sink contribution coefficient are obtained. Among them, the ReLU function is used as the activation function, as shown in the following formula: f(x)=max(0,x). Enhance the non-linear output ability of the carbon sink contribution classifier through the activation function.

[0034] Next, calculate the error between the predicted output and the true output of the network through the mean square error loss function. The error represents the gap between the carbon sink contribution coefficient predicted by the model and the true value. Among them, the mean square error loss function is as follows:

[0035] ;

[0036] where LOSS is the error, m is the number of sample data, The i-th group of soil carbon sink contribution coefficients, vegetation carbon sink contribution coefficients, and water body carbon sink contribution coefficients predicted by the carbon sink contribution classifier, The true i-th group of sample soil carbon sink contribution coefficients, sample vegetation carbon sink contribution coefficients, and sample water body carbon sink contribution coefficients. By calculating the differences between the three respectively, then calculating the square of the sum of the differences, and further calculating the obtained error. Then, for each output node, calculate the gradient of the output error with respect to each weight and bias. During the backpropagation process, the output error is propagated back to the network, and the gradients of each layer can be calculated through the chain rule, as shown in the following formula:

[0037] ;

[0038] where LOSS is the error, is the gradient of the error with respect to the weights of the output layer, is the weight matrix from the hidden layer to the output layer, is the activation value of the i-th group of soil carbon sink contribution coefficients, vegetation carbon sink contribution coefficients, and water body carbon sink contribution coefficients in the activation function; and use the gradient descent method to update the weights and biases of each connection. By continuously adjusting the learning rate and the gradient descent method, the model gradually approaches the optimal solution; repeat the above process until the model converges, that is, the error converges to a smaller value after multiple training iterations. For example, if the error is less than 0.001, the training is completed and converges. Among them, during each training, the weights and biases in the network are adjusted, so that the prediction results are getting closer and closer to the true values, and a trained carbon sink contribution classifier is obtained.

[0039] S130: Input the collapse feature information into the carbon sink contribution classifier, and output the soil carbon sink contribution coefficients, vegetation carbon sink contribution coefficients, and water body carbon sink contribution coefficients of soil carbon sink, vegetation carbon sink, and water body carbon sink; S140: Select the carbon sink method corresponding to the largest contribution coefficient as the main carbon sink method, and use the other two carbon sink methods as the secondary carbon sink methods.

[0040] Specifically, input the collapse feature information into the trained carbon sink contribution classifier, and output the soil carbon sink contribution coefficients, vegetation carbon sink contribution coefficients, and water body carbon sink contribution coefficients of soil carbon sink, vegetation carbon sink, and water body carbon sink. Finally, select the carbon sink method corresponding to the largest contribution coefficient as the main carbon sink method, and use the other two carbon sink methods as the secondary carbon sink methods. For example, assume that the soil carbon sink contribution coefficient is 0.6, the vegetation carbon sink contribution coefficient is 0.3, and the water body carbon sink contribution coefficient is 0.1. Then the main carbon sink method is soil carbon sink, and the other two secondary carbon sink methods are vegetation carbon sink and water body carbon sink respectively.

[0041] By constructing a carbon sink contribution classifier based on machine learning, the intelligence and scientific nature of the classification of carbon sink contribution coefficients can be improved, and further, the scientific nature, accuracy, and efficiency of the classification of carbon sink contribution coefficients can be enhanced, providing more accurate and comprehensive data support for the assessment of the carbon sink capacity of coal mining subsidence areas.

[0042] S200: Collect the soil carbon sink data, vegetation carbon sink data, and water body carbon sink data within the coal mining subsidence area to obtain basic soil carbon sink parameters, basic vegetation carbon sink parameters, and basic water body carbon sink parameters.

[0043] Furthermore, step S200 of the present invention further includes:

[0044] S210: Test the soil organic carbon content within the coal mining subsidence area to obtain soil carbon sink data; S220: Collect the quantities of different vegetation within the coal mining subsidence area and calculate to obtain vegetation carbon sink data; S230: Test the dissolved organic carbon concentration of the water body within the coal mining subsidence area to obtain water body carbon sink data.

[0045] Specifically, through soil sample analysis, test the soil organic carbon content within the coal mining subsidence area, such as the dry oxidation method, element analysis method, etc., to determine the organic carbon content in the soil sample. According to the collected soil organic carbon content data, calculate the total carbon storage in the soil in this area, designated as soil carbon sink data. In the coal mining subsidence area, collect information on different types and quantities of vegetation, such as using drones or satellite images to obtain vegetation coverage and distribution. Different types of vegetation have different carbon sink capacities (the amount of carbon contained in unit biomass), which can be determined by referring to the literature or conducting field experiments. Finally, according to the vegetation quantity and the unit vegetation carbon sink amount corresponding to the vegetation type, calculate the total vegetation carbon sink, designated as vegetation carbon sink data.

[0046] Through water body sample collection and analysis, test the dissolved organic carbon (DOC) concentration of the water body within the coal mining subsidence area, such as using high-performance liquid chromatography (HPLC) or spectroscopy (such as ultraviolet-visible spectroscopy) to measure the organic carbon concentration in the water body. Multiply the dissolved organic carbon concentration by the water body volume to calculate the carbon sink amount of the water body, designated as water body carbon sink data. By collecting data such as soil organic carbon content, vegetation quantity and type, and water body dissolved organic carbon concentration, the carbon sink capacity of the coal mining subsidence area can be comprehensively and accurately evaluated.

[0047] S300: According to the subsidence characteristic information, conduct an analysis of the carbon sink influence trend to obtain a soil carbon sink influence coefficient, a vegetation carbon sink influence coefficient, and a water body carbon sink influence coefficient.

[0048] Furthermore, step S300 of the present invention further includes:

[0049] S310: Collect the subsidence feature information of multiple sample coal mining subsidence areas to obtain a set of subsidence feature information; S320: Collect the change ratios of soil carbon sink parameters, vegetation carbon sink parameters, and water body carbon sink parameters under different sample coal mining subsidence areas within a preset time length to obtain a set of sample soil carbon sink impact coefficients, a set of sample vegetation carbon sink impact coefficients, and a set of sample water body carbon sink impact coefficients; S330: Use the set of subsidence feature information as input features, and use the set of sample soil carbon sink impact coefficients, the set of sample vegetation carbon sink impact coefficients, and the set of sample water body carbon sink impact coefficients as output features, and based on machine learning, construct and train a carbon sink impact trend analyzer until convergence; S340: Input the subsidence feature information into the carbon sink impact trend analyzer, and output to obtain a soil carbon sink impact coefficient, a vegetation carbon sink impact coefficient, and a water body carbon sink impact coefficient.

[0050] Specifically, collect the subsidence feature information (phreatic level, subsidence slope, and subsidence depth) of multiple sample coal mining subsidence areas to obtain a set of subsidence feature information. Then, collect the change ratios of soil carbon sink parameters, vegetation carbon sink parameters, and water body carbon sink parameters under different sample coal mining subsidence areas within a preset time length. The preset time length can be set according to actual situations, such as three months or half a year. Subsidence features will affect different carbon sink methods. For example, a larger slope (such as a subsidence slope of 35° - 40°) may cause the loss of soil organic carbon, making it impossible for vegetation to grow, resulting in a decrease in the carbon sink amounts of soil and vegetation. An increase in water accumulation may lead to an increase in the carbon sink amount of the water body. Obtain a set of sample soil carbon sink impact coefficients, a set of sample vegetation carbon sink impact coefficients, and a set of sample water body carbon sink impact coefficients. Among them, the carbon sink impact coefficient reflects the impact of subsidence features on carbon sink methods. The carbon sink impact coefficient is (current carbon sink amount - initial carbon sink amount) / initial carbon sink amount. This ratio can be positive (increasing the carbon sink amount) or negative (decreasing the carbon sink amount). The soil carbon sink impact coefficient represents the change ratio of the soil carbon sink amount. A larger slope may cause soil carbon loss, and a deeper depth (such as a subsidence depth of 8m - 12m) may increase carbon storage. The vegetation carbon sink impact coefficient represents the change ratio of the vegetation carbon sink amount. For example, when the growth of vegetation in the subsidence area is restricted, the carbon sink capacity decreases. The water body carbon sink impact coefficient represents the impact of subsidence features on the carbon sink amount of the water body, especially the impact of changes in water accumulation and water body area on the carbon sink capacity. For example, a larger slope may cause water to converge to low-lying areas, thereby increasing the water body carbon storage amount.

[0051] Further, a carbon sink impact trend analyzer is constructed based on machine learning. For example, a carbon sink impact trend analyzer is constructed based on a BP neural network. The carbon sink impact trend analyzer includes an input layer (the input data is subsidence feature information), multiple hidden layers, and an output layer (the output data is the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient). Then, the set of subsidence feature information is used as the input feature, and the set of sample soil carbon sink impact coefficients, the set of sample vegetation carbon sink impact coefficients, and the set of sample water body carbon sink impact coefficients are used as the output features to perform supervised training on the carbon sink impact trend analyzer. First, the input feature is transmitted through the input layer to each hidden layer. In each layer, the activation value of each node is calculated through weighted summation, and then the activation value is passed to the next layer through an activation function until the output layer. The output result is the predicted soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient. Next, the error between the predicted output and the actual output is calculated through the mean square error loss function. Then, through the backpropagation algorithm, the error of each layer of nodes is calculated, and the error is transmitted layer by layer from the output layer to the input layer. By calculating the gradient of the weights of each layer, an optimization algorithm (such as the gradient descent method) is used to update the weights in the network. Among them, the activation function, the loss function, and the method of calculating the gradient of the carbon sink impact trend analyzer are the same as those of the carbon sink contribution classifier in the foregoing content, but the specific training data is different. Therefore, after the training converges, a carbon sink impact trend analyzer with different network parameters such as weights can be obtained, and the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient are predicted through the subsidence feature information.

[0052] Through multiple iterations, the processes of forward propagation, loss calculation, backpropagation, and weight update are repeatedly executed. Each iteration will adjust the weights of the neural network, making the loss function gradually decrease. Eventually, the model can accurately predict the carbon sink impact coefficient. When the loss function converges to a small value (such as 0.001) or reaches a predetermined number of training times (such as iteratively training for 100 epochs), the training is stopped, and a trained carbon sink impact trend analyzer is obtained.

[0053] Finally, the subsidence feature information is input into the carbon sink impact trend analyzer for analysis, and the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient are output. By constructing a carbon sink impact trend analyzer based on machine learning, the intelligence and scientific nature of carbon sink impact trend analysis can be improved. Furthermore, the scientific nature, accuracy, and efficiency of obtaining the carbon sink impact coefficient can be improved, providing more accurate and comprehensive data support for the carbon sink capacity assessment of coal mining subsidence areas.

[0054] S400: Collect the carbon sink body characteristic information of the two secondary carbon sink methods and the primary carbon sink method, conduct an analysis on the influence of the secondary carbon sink trend on the primary carbon sink method to obtain the secondary carbon sink influence coefficient, and combine the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient to perform influence correction on the basic carbon sink parameters of the primary carbon sink method and the two secondary carbon sink methods, so as to obtain the soil carbon sink parameter, the vegetation carbon sink parameter, and the water body carbon sink parameter as the carbon sink data analysis results.

[0055] Further, step S400 of the present invention further includes:

[0056] S410: Collect the carbon sink body characteristic information of the two secondary carbon sink methods and the primary carbon sink method. Among them, the carbon sink body characteristic information of soil carbon sink, vegetation carbon sink, and water body carbon sink includes soil component information, vegetation type information, and water body component information; S420: Collect the sample carbon sink body characteristic information set, and collect the change ratio of the carbon sink parameters of other carbon sink methods within a preset time length under each sample carbon sink body characteristic information, and label it as the sample secondary carbon sink influence coefficient set; S430: Use the sample carbon sink body characteristic information set as the input feature, use the sample secondary carbon sink influence coefficient set as the output feature, construct a secondary carbon sink influence analyzer and train it until convergence; S440: Input the carbon sink body characteristic information of the two secondary carbon sink methods into the secondary carbon sink influence analyzer respectively, and output to obtain two secondary carbon sink influence coefficients.

[0057] Specifically, first, collect the carbon sink body characteristic information of the two secondary carbon sink methods and the primary carbon sink method. Among them, the carbon sink body characteristic information of soil carbon sink, vegetation carbon sink, and water body carbon sink includes soil component information (including key factors affecting soil carbon sink capacity such as soil organic carbon content, mineral composition, acidity and alkalinity, etc.), vegetation type information (such as grasses, shrubs, trees, etc., and there are significant differences in carbon sink capacity among different vegetation types), and water body component information (information such as dissolved organic carbon concentration, pH value, temperature of the water body, which affects the water body carbon sink capacity). Then, collect the sample carbon sink body characteristic information set, and collect the change ratio of the carbon sink parameters of other carbon sink methods within a preset time length under each sample carbon sink body characteristic information. The carbon sink ratio is (current carbon sink amount - initial carbon sink amount) / initial carbon sink amount, and this ratio can be a positive value (increasing carbon sink amount) or a negative value (decreasing carbon sink amount), and label it as the sample secondary carbon sink influence coefficient to obtain the sample secondary carbon sink influence coefficient set.

[0058] Next, an auxiliary carbon sink impact analyzer is constructed based on the BP neural network. The auxiliary carbon sink impact analyzer includes an input layer (the input data is the carbon sink main body feature information), multiple hidden layers, and an output layer (the output data is the auxiliary carbon sink impact coefficient). Further, the sample carbon sink main body feature information set is used as the input feature, and the sample auxiliary carbon sink impact coefficient set is used as the output feature to perform supervised training on the auxiliary carbon sink impact analyzer. First, the carbon sink main body feature information is input into the neural network through the input layer, and a predicted value, that is, the auxiliary carbon sink impact coefficient predicted for the currently input carbon sink main body feature information, is obtained through the output layer. Then, for each group of samples, the difference between the predicted value and the true value (that is, the auxiliary carbon sink impact coefficient in the sample) is calculated. The commonly used error calculation method is the mean square error. Then, according to the calculated error, the backpropagation algorithm is used to calculate the gradient of each layer, that is, the adjustment direction of the weights and biases. Through the chain rule, the error is propagated backward layer by layer to the input layer, and the contribution of each weight and bias to the total error is calculated. The gradient value indicates how to adjust each weight and bias to reduce the prediction error. Then, according to the gradient calculated by the backpropagation, an optimization algorithm (such as gradient descent) is used to update the weights and biases in the network. The optimization algorithm makes the network gradually approach the true auxiliary carbon sink impact coefficient by reducing the error. The steps of forward propagation, error calculation, backpropagation, and weight update are repeated until the error of the network converges (that is, reaches the preset error threshold or the maximum number of training rounds), and the trained auxiliary carbon sink impact analyzer is obtained.

[0059] Finally, the carbon sink main body feature information of the two secondary carbon sink methods is respectively input into the auxiliary carbon sink impact analyzer, and two auxiliary carbon sink impact coefficients of the two secondary carbon sink methods are output.

[0060] Further, step S400 of the present invention further includes:

[0061] S450: According to the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient, combined with the two auxiliary carbon sink impact coefficients, three total carbon sink impact coefficients are calculated; S460: Using the three total carbon sink impact coefficients, the basic carbon sink parameters of the main carbon sink method and the two secondary carbon sink methods are corrected for influence to obtain the soil carbon sink parameters, the vegetation carbon sink parameters, and the water body carbon sink parameters as the carbon sink data analysis results.

[0062] Specifically, first, according to the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient, combined with the two auxiliary carbon sink impact coefficients, calculate the total carbon sink impact coefficient. Among them, the total carbon sink impact coefficient of the main carbon sink method is the sum of 1 and the main carbon sink impact coefficient corresponding to the main carbon sink method and the other two auxiliary carbon sink impact coefficients; the total carbon sink impact coefficients of the other two secondary carbon sink methods are the sum of 1 and the main carbon sink impact coefficient corresponding to the main carbon sink method. For example, assuming that the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient, and the water body carbon sink impact coefficient are 0.3, 0.2, and 0.1 respectively, the main carbon sink method is soil carbon sink, and the two auxiliary carbon sink impact coefficients are 0.05 and 0.06 respectively, then the total carbon sink impact coefficient corresponding to the main carbon sink method is 1 + 0.3 + 0.05 + 0.06 = 1.41; the total carbon sink impact coefficients of the two secondary carbon sink methods are 1 + 0.3 = 1.3.

[0063] Finally, use the three total carbon sink impact coefficients to perform impact correction on the basic carbon sink parameters of the main carbon sink method and the two secondary carbon sink methods, that is, multiply the total carbon sink impact coefficient of the main carbon sink method by the basic carbon sink parameter of the main carbon sink method, and multiply the total carbon sink impact coefficient of the secondary carbon sink method by the basic carbon sink parameters of the two secondary carbon sink methods to obtain the soil carbon sink parameter, the vegetation carbon sink parameter, and the water body carbon sink parameter as the carbon sink data analysis result. By combining the total carbon sink impact coefficients of soil, vegetation, and water body to perform impact correction on the basic carbon sink parameters of each carbon sink method, the true carbon sink capacity of each carbon sink method in the coal mining subsidence area can be more accurately reflected. This process considers the interaction between carbon sink methods, thus providing a more scientific and comprehensive basis for the evaluation of the carbon sink capacity of coal mining subsidence areas.

[0064] The carbon sink data analysis method for coal mining subsidence areas provided by the embodiments of the present invention has at least the following technical effects:

[0065] By collecting the subsidence characteristic information of coal mining subsidence areas, classifying to obtain the soil carbon sink contribution coefficient, the vegetation carbon sink contribution coefficient, and the water body carbon sink contribution coefficient, and determining the main carbon sink method and the other two secondary carbon sink methods; then collecting the soil carbon sink data, the vegetation carbon sink data, and the water body carbon sink data within the coal mining subsidence area to obtain the basic soil carbon sink parameters, the basic vegetation carbon sink parameters, and the basic water body carbon sink parameters; then analyzing the carbon sink influence trend according to the subsidence characteristic information to obtain the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient; on the other hand, collecting the carbon sink body characteristic information of the two secondary carbon sink methods and the main carbon sink method, and conducting an auxiliary carbon sink trend influence analysis on the main carbon sink method to obtain the auxiliary carbon sink influence coefficient; finally, combining the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient, and performing influence correction on the basic carbon sink parameters of the main carbon sink method and the two secondary carbon sink methods to obtain the soil carbon sink parameters, the vegetation carbon sink parameters, and the water body carbon sink parameters as the carbon sink data analysis results; that is to say, by using machine learning algorithms to accurately analyze the subsidence characteristics of coal mining subsidence areas and the mutual influences between different carbon sink methods, the scientificity, accuracy, and comprehensiveness of carbon sink data analysis for coal mining subsidence areas can be improved, thereby achieving the technical effect of accurately evaluating the carbon sink capacity of coal mining subsidence areas under different geographical conditions.

[0066] Embodiment 2, as Figure 2 shown, based on the same inventive concept as the carbon sink data analysis method for coal mining subsidence areas provided in Embodiment 1, the embodiment of the present invention further provides a carbon sink data analysis system for coal mining subsidence areas, including: a carbon sink contribution coefficient obtaining module 11, configured to collect the subsidence characteristic information of coal mining subsidence areas, classify to obtain the soil carbon sink contribution coefficient, the vegetation carbon sink contribution coefficient, and the water body carbon sink contribution coefficient, and determine the main carbon sink method and the other two secondary carbon sink methods; a basic carbon sink parameter obtaining module 12, configured to collect the soil carbon sink data, the vegetation carbon sink data, and the water body carbon sink data within the coal mining subsidence area to obtain the basic soil carbon sink parameters, the basic vegetation carbon sink parameters, and the basic water body carbon sink parameters; a carbon sink influence trend analysis module 13, configured to analyze the carbon sink influence trend according to the subsidence characteristic information to obtain the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient; a basic carbon sink parameter correction module 14, configured to collect the carbon sink body characteristic information of the two secondary carbon sink methods and the main carbon sink method, conduct an auxiliary carbon sink trend influence analysis on the main carbon sink method to obtain the auxiliary carbon sink influence coefficient, combine the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient, and the water body carbon sink influence coefficient, and perform influence correction on the basic carbon sink parameters of the main carbon sink method and the two secondary carbon sink methods to obtain the soil carbon sink parameters, the vegetation carbon sink parameters, and the water body carbon sink parameters as the carbon sink data analysis results.

[0067] Furthermore, the carbon sink data analysis system for coal mining subsidence areas further includes: collecting the subsidence slope and subsidence depth of the coal mining subsidence area as subsidence characteristic information; constructing a carbon sink contribution classifier according to the sample carbon sink data of the sample coal mining subsidence area; inputting the subsidence characteristic information into the carbon sink contribution classifier, and outputting the soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient, and water body carbon sink contribution coefficient of the soil carbon sink, vegetation carbon sink, and water body carbon sink; selecting the carbon sink method corresponding to the largest contribution coefficient as the main carbon sink method, and taking the other two carbon sink methods as the secondary carbon sink methods.

[0068] Furthermore, the carbon sink data analysis system for coal mining subsidence areas further includes: collecting the subsidence characteristic information of multiple sample coal mining subsidence areas to obtain a set of subsidence characteristic information; collecting the proportion of the carbon sink amounts of the soil carbon sink, vegetation carbon sink, and water body carbon sink in each sample coal mining subsidence area, and marking to obtain a set of sample soil carbon sink contribution coefficients, a set of sample vegetation carbon sink contribution coefficients, and a set of sample water body carbon sink contribution coefficients; using the set of subsidence characteristic information as input features, and using the set of sample soil carbon sink contribution coefficients, the set of sample vegetation carbon sink contribution coefficients, and the set of sample water body carbon sink contribution coefficients as output features, and based on machine learning, constructing and training the carbon sink contribution classifier until convergence.

[0069] Furthermore, the carbon sink data analysis system for coal mining subsidence areas further includes: testing the soil organic carbon content in the coal mining subsidence area to obtain soil carbon sink data; collecting the quantity of different vegetation in the coal mining subsidence area and calculating to obtain vegetation carbon sink data; testing the dissolved organic carbon concentration in the water body in the coal mining subsidence area to obtain water body carbon sink data.

[0070] Furthermore, the carbon sink data analysis system for coal mining subsidence areas further includes: collecting the subsidence characteristic information of multiple sample coal mining subsidence areas to obtain a set of subsidence characteristic information; collecting the change ratios of the soil carbon sink parameters, vegetation carbon sink parameters, and water body carbon sink parameters under different sample coal mining subsidence areas within a preset time length to obtain a set of sample soil carbon sink influence coefficients, a set of sample vegetation carbon sink influence coefficients, and a set of sample water body carbon sink influence coefficients; using the set of subsidence characteristic information as input features, and using the set of sample soil carbon sink influence coefficients, the set of sample vegetation carbon sink influence coefficients, and the set of sample water body carbon sink influence coefficients as output features, and based on machine learning, constructing and training the carbon sink influence trend analyzer until convergence; inputting the subsidence characteristic information into the carbon sink influence trend analyzer, and outputting the soil carbon sink influence coefficient, vegetation carbon sink influence coefficient, and water body carbon sink influence coefficient.

[0071] Further, the carbon sink data analysis system for coal mining subsidence areas further includes: collecting the carbon sink main body characteristic information of the two secondary carbon sink methods and the main carbon sink method, wherein the carbon sink main body characteristic information of soil carbon sink, vegetation carbon sink and water body carbon sink includes soil component information, vegetation type information and water body component information; collecting a set of sample carbon sink main body characteristic information, and collecting the change ratio of the carbon sink parameters of other carbon sink methods within a preset time length under each sample carbon sink main body characteristic information, which is marked as a set of sample auxiliary carbon sink influence coefficients; using the set of sample carbon sink main body characteristic information as input features and using the set of sample auxiliary carbon sink influence coefficients as output features to construct an auxiliary carbon sink influence analyzer and training it until convergence; inputting the carbon sink main body characteristic information of the two secondary carbon sink methods into the auxiliary carbon sink influence analyzer respectively, and outputting two auxiliary carbon sink influence coefficients.

[0072] Further, the carbon sink data analysis system for coal mining subsidence areas further includes: calculating three total carbon sink influence coefficients according to the soil carbon sink influence coefficient, vegetation carbon sink influence coefficient and water body carbon sink influence coefficient, and combining the two auxiliary carbon sink influence coefficients; using the three total carbon sink influence coefficients to perform influence correction on the basic carbon sink parameters of the main carbon sink method and the two secondary carbon sink methods to obtain soil carbon sink parameters, vegetation carbon sink parameters and water body carbon sink parameters as the carbon sink data analysis results.

[0073] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they learn the basic creative concept.

[0074] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. A method for analyzing carbon sink data in coal mining subsidence areas, characterized in that: Methods include: Collect the subsidence characteristic information of coal mining subsidence areas, classify and obtain the soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient and water carbon sink contribution coefficient, and determine the main carbon sink mode and the other two secondary carbon sink modes, including: Collect the collapse slope and collapse depth of the coal mining collapse area as collapse characteristic information; Based on the sample carbon sink data of sample coal mining subsidence areas, a carbon sink contribution classifier is constructed; Input the collapse characteristic information into the carbon sink contribution classifier, and output the soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient and water body carbon sink contribution coefficient of the soil carbon sink, vegetation carbon sink and water body carbon sink; The carbon sink method corresponding to the largest contribution coefficient is selected as the primary carbon sink method, and the other two carbon sink methods are selected as secondary carbon sink methods; Collecting soil carbon sink data, vegetation carbon sink data and water body carbon sink data in the coal mining subsidence area to obtain basic soil carbon sink parameters, basic vegetation carbon sink parameters and basic water body carbon sink parameters; According to the collapse characteristic information, a carbon sink impact trend analysis is performed to obtain the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient and the water body carbon sink impact coefficient, including: Collecting collapse characteristic information of multiple sample coal mining subsidence areas to obtain a collapse characteristic information set; Collect the change ratios of soil carbon sink parameters, vegetation carbon sink parameters and water body carbon sink parameters of different samples of coal mining subsidence underground within a preset time length, and obtain a set of sample soil carbon sink influence coefficients, a set of sample vegetation carbon sink influence coefficients and a set of sample water body carbon sink influence coefficients; Using the collapse feature information set as input features, using the sample soil carbon sink influence coefficient set, the sample vegetation carbon sink influence coefficient set and the sample water body carbon sink influence coefficient set as output features, based on machine learning, constructing and training a carbon sink impact trend analyzer until convergence; Inputting the collapse characteristic information into the carbon sink impact trend analyzer, and outputting the soil carbon sink impact coefficient, the vegetation carbon sink impact coefficient and the water body carbon sink impact coefficient; Collecting the carbon sink main characteristic information of the two secondary carbon sink modes and the main carbon sink mode, performing auxiliary carbon sink trend impact analysis on the main carbon sink mode, and obtaining the auxiliary carbon sink impact coefficient, including: Collecting the carbon sink main characteristic information of the two secondary carbon sink modes and the main carbon sink mode, wherein the carbon sink main characteristic information of soil carbon sink, vegetation carbon sink and water body carbon sink includes soil composition information, vegetation type information and water body composition information; Collect the main characteristic information set of sample carbon sinks, and under each main characteristic information of sample carbon sinks, the change ratio of carbon sink parameters of other carbon sink methods within a preset time length, marked as the sample auxiliary carbon sink influence coefficient set; Using the sample carbon sink main feature information set as input features, using the sample auxiliary carbon sink influence coefficient set as output features, constructing an auxiliary carbon sink influence analyzer and training it until convergence; Inputting the carbon sink main characteristic information of the two secondary carbon sink modes into the auxiliary carbon sink impact analyzer respectively, and outputting two auxiliary carbon sink impact coefficients; Combined with the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient and the water body carbon sink influence coefficient, the basic carbon sink parameters of the main carbon sink mode and the two secondary carbon sink modes are corrected for impact, and soil carbon sink parameters, vegetation carbon sink parameters and water body carbon sink parameters are obtained as carbon sink data analysis results.

2. The carbon sink data analysis method for coal mining subsidence according to claim 1 is characterized in that: According to the sample carbon sink data of sample coal mining subsidence areas, a carbon sink contribution classifier is constructed, including: Collecting collapse characteristic information of multiple sample coal mining subsidence areas to obtain a collapse characteristic information set; Collect the carbon sink proportions of soil carbon sink, vegetation carbon sink, and water body carbon sink in each sample coal mining subsidence area, and annotate to obtain the sample soil carbon sink contribution coefficient set, sample vegetation carbon sink contribution coefficient set, and sample water body carbon sink contribution coefficient set; The collapse feature information set is used as input features, and the sample soil carbon sink contribution coefficient set, the sample vegetation carbon sink contribution coefficient set and the sample water body carbon sink contribution coefficient set are used as output features. Based on machine learning, a carbon sink contribution classifier is constructed and trained until convergence.

3. The carbon sink data analysis method for coal mining subsidence according to claim 1, characterized in that: The soil carbon sink data, vegetation carbon sink data and water body carbon sink data in the coal mining subsidence area are collected to obtain basic soil carbon sink parameters, basic vegetation carbon sink parameters and basic water body carbon sink parameters, including: Testing the soil organic carbon content in the coal mining subsidence area to obtain soil carbon sink data; Collecting the number of different vegetation in the coal mining subsidence area and calculating and obtaining vegetation carbon sink data; The dissolved organic carbon concentration of the water in the coal mining subsidence area is tested to obtain water carbon sink data.

4. The carbon sink data analysis method for coal mining subsidence according to claim 1, characterized in that: Combined with the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient and the water body carbon sink influence coefficient, the basic carbon sink parameters of the main carbon sink mode and the two secondary carbon sink modes are corrected for impact, and soil carbon sink parameters, vegetation carbon sink parameters and water body carbon sink parameters are obtained as carbon sink data analysis results, including: According to the soil carbon sink influence coefficient, the vegetation carbon sink influence coefficient and the water body carbon sink influence coefficient, combined with the two auxiliary carbon sink influence coefficients, three total carbon sink influence coefficients are calculated; The three total carbon sink impact coefficients are used to perform impact correction on the basic carbon sink parameters of the main carbon sink mode and the two secondary carbon sink modes, and soil carbon sink parameters, vegetation carbon sink parameters and water body carbon sink parameters are obtained as carbon sink data analysis results.

5. A carbon sink data analysis system for coal mining subsidence, characterized in that: The steps for implementing the carbon sink data analysis method for coal mining subsidence area according to any one of claims 1 to 4 include: The carbon sink contribution coefficient acquisition module is used to collect the collapse characteristic information of coal mining subsidence areas, classify and obtain the soil carbon sink contribution coefficient, vegetation carbon sink contribution coefficient and water body carbon sink contribution coefficient, and determine the main carbon sink mode and the other two secondary carbon sink modes; A basic carbon sink parameter acquisition module is used to collect soil carbon sink data, vegetation carbon sink data and water body carbon sink data in the coal mining subsidence area to obtain basic soil carbon sink parameters, basic vegetation carbon sink parameters and basic water body carbon sink parameters; A carbon sink impact trend analysis module is used to perform carbon sink impact trend analysis based on the collapse characteristic information to obtain soil carbon sink impact coefficient, vegetation carbon sink impact coefficient and water body carbon sink impact coefficient; The basic carbon sink parameter correction module is used to collect the main carbon sink characteristic information of the two secondary carbon sink modes and the main carbon sink mode, perform auxiliary carbon sink trend impact analysis on the main carbon sink mode, obtain the auxiliary carbon sink impact coefficient, and combine the soil carbon sink impact coefficient, vegetation carbon sink impact coefficient and water carbon sink impact coefficient to perform impact correction on the basic carbon sink parameters of the main carbon sink mode and the two secondary carbon sink modes, and obtain soil carbon sink parameters, vegetation carbon sink parameters and water carbon sink parameters as the carbon sink data analysis results.

Images