Carbon emission factor determination method and device, equipment and storage medium

Abstract

The invention discloses a carbon emission factor determination method and device, equipment and a storage medium. The method comprises the following steps: acquiring first carbon emission activity associated data of a to-be-determined carbon emission factor and second carbon emission activity associated data of a disclosed carbon emission factor; performing dimension reduction processing on the second carbon emission activity associated data by using an adaptive Gaussian nucleation principal component analysis algorithm to obtain first dimension reduction data; and determining a target data set matched with the first carbon emission activity associated data from the carbon emission activity feature data set by using a preset classification model, and determining a value of the to-be-determined carbon emission factor according to a carbon emission factor corresponding to the target data set. According to the technical scheme provided by the embodiment of the invention, the problems of nonlinearity and high dimensionality of the carbon emission activity associated data of the unknown carbon emission factor are solved, and the matching speed is considered while the matching precision of the unknown carbon emission factor is improved.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, equipment and storage medium for determining a carbon emission factor. Background Art

[0002] In the accounting of corporate carbon emissions, the emission factor method has become the mainstream way to estimate greenhouse gas emissions because of its wide application scenarios and low implementation difficulty. In the accounting process, selecting a reasonable carbon emission factor will have a significant impact on the final result.

[0003] At present, the number of authoritative carbon emission factors disclosed is limited, which is insufficient to cover all scenarios encountered in actual business. As a result, when measuring undisclosed carbon emission activities, it is often necessary to manually select similar carbon emission factors based on subjective judgment.

[0004] However, the existing methods of determining carbon emission factors are mostly based on expert experience, which is often limited by the completeness of empirical rules and cannot be well applied to undisclosed carbon emission activities. The accuracy of estimating new carbon emission factors that will be disclosed in the future is poor, which affects the results of carbon emission accounting. Summary of the invention

[0005] The present disclosure provides a carbon emission factor determination method, device, equipment and storage medium to solve the problem of poor accuracy in estimating a newly added carbon emission factor.

[0006] In a first aspect, the present disclosure provides a method for determining a carbon emission factor, comprising:

[0007] Obtaining data associated with a first carbon emission activity for which a carbon emission factor is to be determined and data associated with a second carbon emission activity for which a carbon emission factor has been disclosed;

[0008] Using an adaptive Gaussian kernelized principal component analysis algorithm to perform dimensionality reduction processing on the second carbon emission activity associated data to obtain first dimensionality reduction data, wherein the first dimensionality reduction data includes a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor;

[0009] A preset classification model is used to determine a target data group matching the first carbon emission activity associated data from the carbon emission activity characteristic data group, and a value of the to-be-determined carbon emission factor is determined according to the carbon emission factor corresponding to the target data group.

[0010] Furthermore, before obtaining the second carbon emission activity associated data of the disclosed carbon emission factor, it also includes: obtaining the associated data to be counted of carbon emission activities corresponding to multiple disclosed carbon emission factors, wherein each of the associated data to be counted of the disclosed carbon emission factors contains multiple types of associated data; for each type of the associated data, determining the database modeling method according to the logical correlation between the current type of associated data, and using the database modeling method to model the associated data to be counted to obtain a target database; using the target database to generate the second carbon emission activity associated data, wherein the second carbon emission activity associated data includes a matrix containing the associated data to be counted.

[0011] Furthermore, the second carbon emission activity associated data includes the total amount of main raw materials input, the total amount of auxiliary raw materials input, the total amount of packaging of input raw materials, the total amount of energy supply required, the total amount of main products output, the total amount of by-products output, the total amount of waste output and the type of production process in the carbon emission activity corresponding to the disclosed carbon emission factor.

[0012] Furthermore, the use of an adaptive Gaussian kernelized principal component analysis algorithm to reduce the dimension of the second carbon emission activity associated data to obtain first reduced dimension data includes: using an adaptive Gaussian kernelized principal component analysis algorithm to map the second carbon emission activity associated data to a space of preset dimensions to obtain a reduced dimension matrix to be determined, wherein the preset dimension of the reduced dimension matrix to be determined is smaller than the dimension of the second carbon emission activity associated data; determining the ratio of the sum of the sample variances of the eigenvectors in the reduced dimension matrix to be determined to the total variance of the eigenvectors in the reduced dimension matrix to be determined, and updating the hyperfunction of the Gaussian kernel function according to the ratio to obtain a target hyperfunction; generating the first reduced dimension data according to the reduced dimension matrix to be determined and the target hyperfunction.

[0013] Further, the updating of the hyperfunction of the Gaussian kernel function according to the proportion to obtain the target hyperfunction includes: if the proportion is greater than a preset value, determining the current hyperfunction of the Gaussian kernel function as the target hyperfunction; wherein, generating the first dimensionality reduction data according to the to-be-determined dimensionality reduction matrix and the target hyperfunction includes: determining the to-be-determined dimensionality reduction matrix corresponding to the target hyperfunction as the first dimensionality reduction data.

[0014] Furthermore, before using the preset classification model to determine the target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group, it also includes: using a random forest algorithm to generate an initial classification model, and using the second carbon emission activity associated data and the newly added third carbon emission activity associated data of the disclosed carbon emission factor to train the initial classification model to obtain the preset classification model.

[0015] Furthermore, the use of a preset classification model to determine, from the carbon emission activity characteristic data group, a target data group that matches the first carbon emission activity associated data, and determining the value of the carbon emission factor to be determined based on the carbon emission factor corresponding to the target data group, includes: using the preset classification model to perform multiple rounds of processing on the first dimensionality reduction data and the first carbon emission activity associated data to obtain a preset number of target data groups that match the first carbon emission activity associated data, wherein the first dimensionality reduction data processed in the current round by the preset classification model does not include the target data group determined in the previous round; determining the carbon emission factor corresponding to each of the target data groups as a matching carbon emission factor, and determining the value of the carbon emission factor to be determined based on the matching carbon emission factor.

[0016] In a second aspect, the present disclosure provides a carbon emission factor determination device, comprising:

[0017] A data acquisition module, used to acquire first carbon emission activity-related data of a to-be-determined carbon emission factor and second carbon emission activity-related data of a disclosed carbon emission factor;

[0018] a data dimension reduction module, configured to perform dimension reduction processing on the second carbon emission activity associated data using an adaptive Gaussian kernelized principal component analysis algorithm to obtain first dimension reduction data, wherein the first dimension reduction data includes a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor;

[0019] A carbon emission factor matching module is used to determine a target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group using a preset classification model, and determine the value of the carbon emission factor to be determined based on the carbon emission factor corresponding to the target data group.

[0020] In a third aspect, the present disclosure provides an electronic device, the electronic device comprising:

[0021] at least one processor;

[0022] and a memory communicatively coupled to the at least one processor;

[0023] The memory stores a computer program that can be executed by at least one processor, and the computer program is executed by at least one processor so that the at least one processor can execute the carbon emission factor determination method of the first aspect mentioned above.

[0024] In a fourth aspect, the present disclosure provides a computer-readable storage medium storing computer instructions, which are used to enable a processor to implement the carbon emission factor determination method of the first aspect when executed.

[0025] The carbon emission factor determination scheme provided by the present invention utilizes an adaptive Gaussian kernelization principal component analysis algorithm to reduce the dimension of the carbon emission activity associated data of the disclosed carbon emission factors, thereby solving the problems of nonlinearity and high dimensionality of the carbon emission activity associated data. A classification model is then used to identify, from the carbon emission activity associated data of the disclosed carbon emission factors, a data group that matches the carbon emission activity associated data of the carbon emission factor to be determined. The carbon emission factor corresponding to the data group is the carbon emission factor that matches the carbon emission factor to be determined. The carbon emission factor can be used to determine the value of the carbon emission factor to be determined, thereby improving the matching accuracy while taking into account the matching speed.

[0026] It should be understood that the content described in this section is not intended to identify the key or important features of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0028] Figure 1 is a flow chart of a method for determining a carbon emission factor according to Embodiment 1 of the present disclosure;

[0029] Figure 2 is a flow chart of generating a classification model according to the first embodiment of the present disclosure;

[0030] Figure 3 is a flow chart of a method for determining a carbon emission factor provided according to Embodiment 2 of the present disclosure;

[0031] Figure 4 is a flow chart for generating first dimensionality reduction data provided according to the second embodiment of the present disclosure;

[0032] Figure 5 is a structural schematic diagram of a carbon emission factor determination device provided according to Embodiment 3 of the present disclosure;

[0033] Figure 6 It is a structural diagram of an electronic device provided according to the fourth embodiment of the present disclosure. DETAILED DESCRIPTION

[0034] In order to enable those skilled in the art to better understand the scheme of the present disclosure, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in the field without creative work should fall within the scope of protection of the present disclosure.

[0035] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present disclosure described here can be implemented in an order other than those illustrated or described here. In the description of the present disclosure, unless otherwise specified, "multiple" 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: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0036] Embodiment 1

[0037] Figure 1 A flowchart of a method for determining a carbon emission factor is provided for the first embodiment of the present disclosure. This embodiment is applicable to the case of determining a newly added unknown carbon emission factor. The method can be executed by a carbon emission factor determination device. The carbon emission factor determination device can be implemented in the form of hardware and / or software. The carbon emission factor determination device can be configured in an electronic device. The electronic device can be composed of two or more physical entities or one physical entity.

[0038] like Figure 1 As shown, a method for determining a carbon emission factor provided in the first embodiment of the present disclosure specifically includes the following steps:

[0039] S101. Obtain first carbon emission activity-related data of a to-be-determined carbon emission factor and second carbon emission activity-related data of a disclosed carbon emission factor.

[0040] In this embodiment, due to the large differences between different carbon emission activities, the screening of carbon emission activity related data should be relatively independent, versatile, easy to obtain, and comprehensive in dimension. The first carbon emission activity related data and the second carbon emission activity related data may include input raw material information, output raw material information, and production process information of the carbon emission activity.

[0041] S102. Use an adaptive Gaussian kernelized principal component analysis algorithm to perform dimensionality reduction processing on the second carbon emission activity associated data to obtain first dimensionality reduction data, wherein the first dimensionality reduction data contains a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor.

[0042] In this embodiment, the first carbon emission activity associated data and the second carbon emission activity associated data may be expressed in the form of a matrix containing the corresponding carbon emission activity associated data. Since the number of elements and dimensions in the second carbon emission activity associated data matrix are usually large, this brings great complexity to the subsequent matching. At the same time, since there may be redundancy in the second carbon emission activity associated data, it will cause the preset classification model to overfit, resulting in the result output by the preset classification model being non-globally optimal. Therefore, it is necessary to reduce the dimension of the second carbon emission activity associated data. The dimensionality reduction processing of the adaptive Gaussian kernelized principal component analysis algorithm can filter out the carbon emission activity characteristic data from the second carbon emission activity associated data to obtain the first reduced dimensionality data, wherein the carbon emission activity characteristic data group in the first reduced dimensionality data may include multiple carbon emission activity characteristic data.

[0043] S103: Determine a target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group using a preset classification model, and determine a value of the to-be-determined carbon emission factor according to the carbon emission factor corresponding to the target data group.

[0044] In this embodiment, the first dimension reduction data and the first carbon emission activity associated data can be input into a preset classification model to obtain a carbon emission activity characteristic data group that matches the first carbon emission activity associated data, that is, a target data group. If the number of target data groups is 1, the value of the carbon emission factor corresponding to the target data group can be determined as the value of the carbon emission factor to be determined.

[0045] The carbon emission factor determination method provided by the embodiment of the present disclosure utilizes an adaptive Gaussian kernelization principal component analysis algorithm to reduce the dimension of the carbon emission activity associated data of the disclosed carbon emission factors, thereby solving the problems of nonlinearity and high dimensionality of the carbon emission activity associated data. The classification model is then used to identify, from the carbon emission activity associated data of the disclosed carbon emission factors, a data group that matches the carbon emission activity associated data of the carbon emission factor to be determined. The carbon emission factor corresponding to the data group is the carbon emission factor that matches the carbon emission factor to be determined. The carbon emission factor can be used to determine the value of the carbon emission factor to be determined, thereby improving the matching accuracy while taking into account the matching speed.

[0046] Optionally, before obtaining the second carbon emission activity associated data of the disclosed carbon emission factor, it also includes: determining the associated data to be counted of carbon emission activities corresponding to multiple disclosed carbon emission factors, wherein each of the associated data to be counted of the disclosed carbon emission factors contains multiple types of associated data; for each type of the associated data, determining the database modeling method according to the logical correlation between the current type of associated data, and using the database modeling method to model the associated data to be counted to obtain a target database; using the target database to generate the second carbon emission activity associated data, wherein the second carbon emission activity associated data includes a matrix containing the associated data to be counted.

[0047] Optionally, the second carbon emission activity associated data includes the total amount of main raw materials input, the total amount of auxiliary raw materials input, the total amount of packaging of input raw materials, the total amount of energy supply required, the total amount of main products output, the total amount of by-products output, the total amount of waste output and the type of production process in the carbon emission activity corresponding to the disclosed carbon emission factor.

[0048] Specifically, the types of associated data included in the associated data to be counted include: the total amount of main raw materials input into carbon emission activities, the total amount of auxiliary raw materials input, the total amount of packaging of input raw materials, the total amount of energy supply required, the total amount of main products output, the total amount of by-products output, the total amount of waste output and the type of production process.

[0049] For example, the disclosed carbon emission factors and the associated data to be counted can be modeled, and the modeling information is shown in the modeling information table in Table 1 below:

[0050] Table 1 Modeling information table

[0051]

[0052]

[0053] The information in the above table can be used to generate multiple (target) databases of the associated data to be counted, and the target database can be used to generate the second carbon emission activity associated data. The second carbon emission activity associated data can be represented as the following matrix:

[0054] A i =[X i ,Y i ,Z i ,U i ,N i ,M i ,W i ,E i ]

[0055] in, A i is the second carbon emission activity related data matrix corresponding to the disclosed carbon emission factors, X i Y is the main raw material vector input in the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the quantity or content of each type of main raw material in the main raw material target database. i is the auxiliary raw material vector input in the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the quantity or content of each type of auxiliary raw material in the auxiliary raw material target database. i The packaging vector of the input raw material in the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the number of each type of packaging in the raw material packaging target database. i is the energy supply vector required in the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the quantity or content of each type of energy supply required in the energy supply target database. i M is the main product vector output from the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the quantity or content of each type of main product in the main product target database. i W is the by-product vector output from the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the quantity or content of each type of by-product in the by-product target database. i E is the waste vector output in the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the quantity or content of each type of waste in the waste target database. i is the production process vector in the carbon emission activity corresponding to the carbon emission factor. The elements contained in this vector are the types of production processes in the production process target database. If this type is involved, the value is 1, and if this type is not involved, the value is 0. In summary, under normal circumstances, A iThe dimension of is large and contains many elements, which brings great complexity to the subsequent matching calculation, so further dimensionality reduction processing is needed.

[0056] Optionally, before determining the target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group using the preset classification model, it also includes: generating an initial classification model using a random forest algorithm, and training the initial classification model using the second carbon emission activity associated data and the newly added third carbon emission activity associated data of the disclosed carbon emission factor to obtain a preset classification model. The advantage of this setting is that by using the decision tree in the classification model generated by the random forest algorithm, the identification problem of the newly disclosed carbon emission factors in the future is effectively solved, and by training the classification model using the newly added carbon emission activity associated data of the disclosed carbon emission factors, the classification model has better algorithm robustness.

[0057] Specifically, random forest can be understood as a classifier containing multiple decision trees. The random forest algorithm can be used to generate an initial classification model. Figure 2 is a flowchart for generating a classification model. Figure 2 The feature vector in is the feature vector containing the associated data of carbon emission activities. Figure 2 As shown, it is possible to regularly check whether there is any newly added third carbon emission activity-related data with disclosed carbon emission factors. If there is, it can be included in the classification model training set that contains the second carbon emission activity-related data. By randomly extracting training samples from the classification model training set, the features of the newly added carbon emission activity-related data can be included in the decision tree of the classification model. Among them, the parameters that need to be set in the random forest algorithm can be determined based on experience or preliminary experiments, such as:

[0058] 1) The number of forest trees N can be determined based on the number of major categories of disclosed carbon emission factors, such as 28;

[0059] 2) The number of features M in the training set will affect the convergence speed of the algorithm, which can be determined through experiments, such as taking a value of 3;

[0060] 3) The depth of the decision tree can be determined based on the sub-category depth of the major categories of disclosed carbon emission factors, such as a value of 12.

[0061] Embodiment 2

[0062] Figure 3 This is a flow chart of a method for determining a carbon emission factor provided in the second embodiment of the present disclosure. The technical solution of the embodiment of the present disclosure is further optimized on the basis of the above-mentioned optional technical solutions, and a specific method for determining a newly added unknown carbon emission factor is given.

[0063] Optionally, the method of reducing the dimension of the second carbon emission activity associated data by using an adaptive Gaussian kernelized principal component analysis algorithm to obtain the first reduced dimension data includes: using an adaptive Gaussian kernelized principal component analysis algorithm to map the second carbon emission activity associated data to a space of preset dimensions to obtain a to-be-determined reduced dimension matrix, wherein the preset dimension of the to-be-determined reduced dimension matrix is ​​smaller than the dimension of the second carbon emission activity associated data; determining the proportion of the sum of the sample variances of the eigenvectors in the to-be-determined reduced dimension matrix to the overall variance of the eigenvectors in the to-be-determined reduced dimension matrix, and updating the hyperfunction of the Gaussian kernel function according to the proportion to obtain the target hyperfunction; generating the first reduced dimension data according to the to-be-determined reduced dimension matrix and the target hyperfunction. The advantage of such a setting is that, by using the adaptive Gaussian kernelized principal component analysis algorithm, the data complexity of the carbon emission activity associated data is effectively reduced, and the difficulty of engineering implementation of the algorithm application is reduced by adaptive parameter adjustment, and the efficiency and accuracy of the algorithm are improved without losing data information.

[0064] Optionally, the method of using a preset classification model to determine a target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group, and determining the value of the carbon emission factor to be determined based on the carbon emission factor corresponding to the target data group, includes: using a preset classification model to perform multiple rounds of processing on the first dimensionality reduction data and the first carbon emission activity associated data to obtain a preset number of target data groups that match the first carbon emission activity associated data, wherein the first dimensionality reduction data processed by the preset classification model in the current round does not include the target data group determined in the previous round; determining the carbon emission factor corresponding to each of the target data groups as a matching carbon emission factor, and determining the value of the carbon emission factor to be determined based on the matching carbon emission factor. The advantage of such a setting is that by eliminating the previous round target data group in the first dimensionality reduction data, it is possible to determine multiple carbon emission factors with a high degree of matching, and then through subsequent analysis of these carbon emission factors, the accuracy of determining unknown carbon emission factors can be further improved.

[0065] like Figure 3 As shown, a method for determining a carbon emission factor provided in Embodiment 2 of the present disclosure specifically includes the following steps:

[0066] S201. Obtain first carbon emission activity associated data of a carbon emission factor to be determined; determine associated data to be counted of carbon emission activities corresponding to multiple disclosed carbon emission factors; for each type of associated data, determine a database modeling method according to the logical correlation between the current type of associated data, and use the database modeling method to model the associated data to be counted to obtain a target database; and use the target database to generate second carbon emission activity associated data.

[0067] S202: using an adaptive Gaussian kernelized principal component analysis algorithm, mapping the second carbon emission activity related data to a space of preset dimensions to obtain a dimension reduction matrix to be determined.

[0068] The preset dimension of the to-be-determined dimensionality reduction matrix is ​​smaller than the dimension of the second carbon emission activity associated data.

[0069] Specifically, Figure 4 is a flow chart for generating the first dimension reduction data. Figure 4 As shown, the hyperparameter γ of the Gaussian function can be initialized first, such as taking it as 0.001. The mathematical expression of the Gaussian kernel function is as follows:

[0070] K(x,y)=exp(-γ||xy|| 2 )

[0071] Among them, the factor that affects the shape of the Gaussian kernel function is the hyperfunction γ, γ>0. When γ is larger, the Gaussian kernel function will make the data points more concentrated in the high-dimensional space, but it will also lead to the loss of data feature information due to excessive concentration. The Gaussian kernel function K(x,y) is a function that calculates the kernel matrix K between two sample sets x and y. The kernel matrix K is a symmetric matrix, in which the elements represent the similarity or correlation between the two samples.

[0072] The dimension reduction matrix A′ to be determined i Methods of determination include:

[0073] A′ i =[α1,α2,α3,…,α p ]·[K(x1,A i ),K(x2,A i ),K(x3,A i ),…,K(x p ,A i )] T

[0074] Among them, p(p <L A ) is the principal component variable, indicating that A i The dimension of is reduced to p dimensions. Figure 4 As shown, by calculating the eigenvalue of the Gaussian kernel function K(x,y), the eigenvalue α of the Gaussian kernel matrix K can be obtained. P And the eigenvector K(x,A) of the Gaussian kernel matrix i ). If A iIt contains 12,000 elements, and the preset dimension can be set to 6, which can not only meet the need to distinguish carbon emission factors, but also reduce the calculation cycle of the following steps. That is, any second carbon emission activity related data (matrix or vector) containing 12,000 elements can be represented by a 6-dimensional low-dimensional vector.

[0075] S203, determining the ratio of the sum of the sample variances of the eigenvectors in the to-be-determined reduced dimension matrix to the overall variance of the eigenvectors in the to-be-determined reduced dimension matrix, and updating the hyperfunction of the Gaussian kernel function according to the ratio to obtain a target hyperfunction.

[0076] Specifically, from a mathematical point of view, if the sample variance of the reduced dimension matrix is ​​to be determined The sum of Account for the total variance If the ratio is larger than the preset value, it means that the vector in the dimension reduction matrix contributes more information to distinguish the differences between the eigenvalues, so the vector in the dimension reduction matrix to be determined after dimension reduction is called the principal component. Otherwise, it means that the nonlinear data set A i It is still impossible to completely separate them in high-dimensional space, and it is necessary to update the hyperfunction of the Gaussian kernel function, such as increasing the hyperparameters, to obtain the target hyperfunction.

[0077] Optionally, updating the hyperfunction of the Gaussian kernel function according to the proportion to obtain a target hyperfunction includes: if the proportion is greater than a preset value, determining the current hyperfunction of the Gaussian kernel function as the target hyperfunction; wherein, generating the first dimensionality reduction data according to the to-be-determined dimensionality reduction matrix and the target hyperfunction includes: determining the to-be-determined dimensionality reduction matrix corresponding to the target hyperfunction as the first dimensionality reduction data.

[0078] Specifically, Figure 4 As shown, it can be determined whether the percentage of the sum of the variances of the eigenvectors in the low-dimensional space (i.e., the sum of the sample variances of the eigenvectors in the to-be-determined reduced-dimensional matrix) to the total variance (i.e., the overall variance of the eigenvectors in the to-be-determined reduced-dimensional matrix) is greater than 99%. If so, the current hyperfunction of the Gaussian kernel function is determined as the target hyperfunction, and the to-be-determined reduced-dimensional matrix corresponding to the target hyperfunction is Figure 4 If not, the current hyperfunction is increased by a specified step size, such as 0.01, and a new round of calculations is performed using the hyperparameters obtained after the increase to determine the new ratio until the ratio is greater than the preset value.

[0079] S204: Generate first dimensionality reduction data according to the to-be-determined dimensionality reduction matrix and the target hyperfunction.

[0080] Specifically, by determining the target hyperfunction, the dimension reduction matrix to be determined can be converted into the first dimension reduction data using an adaptive Gaussian kernelized principal component analysis algorithm.

[0081] S205: Generate an initial classification model using a random forest algorithm, and train the initial classification model using the second carbon emission activity associated data and the newly added third carbon emission activity associated data with disclosed carbon emission factors to obtain a preset classification model.

[0082] S206: Perform multiple rounds of processing on the first dimensionality reduction data and the first carbon emission activity associated data using a preset classification model to obtain a preset number of target data groups that match the first carbon emission activity associated data.

[0083] Among them, the first dimensionality reduction data processed by the preset classification model in the current round does not include the target data group determined in the previous round.

[0084] Specifically, the preset classification model can usually output a target data group with the highest matching degree with the first carbon emission activity associated data in each round. When the preset number is greater than 1, the target data group output in the previous round can be identified by removing the first dimensionality reduction data from the target data group output in each round of processing, thereby obtaining multiple target data groups with matching degrees from high to low. Among them, the variance of the vector containing the first carbon emission activity associated data and the vector in the carbon emission activity characteristic data group corresponding to multiple disclosed carbon emission factors (i.e., the vector in the matrix containing the first dimensionality reduction data) can be determined as an indicator of matching degree. The smaller the variance, the higher the similarity and the higher the matching degree. Conversely, the lower the similarity, the lower the matching degree.

[0085] S207: Determine the carbon emission factor corresponding to each of the target data groups as a matching carbon emission factor, and determine the value of the to-be-determined carbon emission factor according to the matching carbon emission factor.

[0086] Specifically, the determined multiple matching carbon emission factors can be determined as matching carbon emission factors of the carbon emission factor to be determined. The specific value of the carbon emission factor to be determined can be finally determined by analyzing the matching degree of these matching carbon emission factors.

[0087] The carbon emission factor determination method provided in the embodiment of the present disclosure effectively reduces the data complexity of the data associated with carbon emission activities by utilizing an adaptive Gaussian kernelized principal component analysis algorithm, and reduces the engineering implementation difficulty of the algorithm application by adaptive parameter adjustment. Without losing data information, the efficiency and accuracy of the algorithm are improved, and by eliminating the front wheel target data group in the first dimensionality reduction data, it is possible to determine multiple carbon emission factors with a high degree of matching. Through subsequent analysis of these carbon emission factors, the accuracy of determining unknown carbon emission factors can be further improved.

[0088] Embodiment 3

[0089] Figure 5 This is a schematic diagram of the structure of a device for determining a carbon emission factor provided in Embodiment 3 of the present disclosure. Figure 5 As shown, the device includes: a data acquisition module 301, a data dimension reduction module 302 and a carbon emission factor matching module 303, wherein:

[0090] A data acquisition module, used to acquire first carbon emission activity-related data of a to-be-determined carbon emission factor and second carbon emission activity-related data of a disclosed carbon emission factor;

[0091] a data dimension reduction module, configured to perform dimension reduction processing on the second carbon emission activity associated data using an adaptive Gaussian kernelized principal component analysis algorithm to obtain first dimension reduction data, wherein the first dimension reduction data includes a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor;

[0092] A carbon emission factor matching module is used to determine a target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group using a preset classification model, and determine the value of the carbon emission factor to be determined based on the carbon emission factor corresponding to the target data group.

[0093] The carbon emission factor determination device provided by the embodiment of the present disclosure uses an adaptive Gaussian kernelization principal component analysis algorithm to reduce the dimension of the carbon emission activity associated data of the disclosed carbon emission factors, thereby solving the problems of nonlinearity and high dimensionality of the carbon emission activity associated data. The classification model is then used to identify, from the carbon emission activity associated data of the disclosed carbon emission factors, a data group that matches the carbon emission activity associated data of the carbon emission factor to be determined. The carbon emission factor corresponding to the data group is the carbon emission factor that matches the carbon emission factor to be determined. The carbon emission factor can be used to determine the value of the carbon emission factor to be determined, thereby improving the matching accuracy while taking into account the matching speed.

[0094] Optionally, the device further comprises:

[0095] A module for obtaining data to be counted, used for obtaining the associated data to be counted of carbon emission activities corresponding to a plurality of disclosed carbon emission factors before obtaining the associated data of the second carbon emission activity of the disclosed carbon emission factor, wherein each associated data to be counted of the disclosed carbon emission factor includes multiple types of associated data;

[0096] A modeling module, for determining a database modeling method for each type of the associated data according to the logical association between the associated data of the current type, and using the database modeling method to model the associated data to be counted, so as to obtain a target database;

[0097] The associated data generating module is used to generate second carbon emission activity associated data using the target database, wherein the second carbon emission activity associated data includes a matrix including the associated data to be counted.

[0098] Optionally, the second carbon emission activity associated data includes the total amount of main raw materials input, the total amount of auxiliary raw materials input, the total amount of packaging of input raw materials, the total amount of energy supply required, the total amount of main products output, the total amount of by-products output, the total amount of waste output and the type of production process in the carbon emission activity corresponding to the disclosed carbon emission factor.

[0099] Optionally, the data dimension reduction module includes:

[0100] a dimensionality reduction unit, configured to map the second carbon emission activity associated data to a space of preset dimensions using an adaptive Gaussian kernelized principal component analysis algorithm to obtain a to-be-determined reduced dimension matrix, wherein the preset dimension of the to-be-determined reduced dimension matrix is ​​smaller than the dimension of the second carbon emission activity associated data;

[0101] A target hyperfunction determination unit, used to determine the ratio of the sum of the sample variances of the eigenvectors in the to-be-determined reduced-dimensionality matrix to the overall variance of the eigenvectors in the to-be-determined reduced-dimensionality matrix, and to update the hyperfunction of the Gaussian kernel function according to the ratio to obtain the target hyperfunction;

[0102] The dimension reduction data generating unit is used to generate first dimension reduction data according to the dimension reduction matrix to be determined and the target hyperfunction.

[0103] Optionally, updating the hyperfunction of the Gaussian kernel function according to the proportion to obtain a target hyperfunction includes: if the proportion is greater than a preset value, determining the current hyperfunction of the Gaussian kernel function as the target hyperfunction; wherein, generating the first dimensionality reduction data according to the to-be-determined dimensionality reduction matrix and the target hyperfunction includes: determining the to-be-determined dimensionality reduction matrix corresponding to the target hyperfunction as the first dimensionality reduction data.

[0104] Optionally, the device further comprises:

[0105] A classification model determination module is used to generate an initial classification model using a random forest algorithm before determining a target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group using a preset classification model, and to train the initial classification model using the second carbon emission activity associated data and the newly added third carbon emission activity associated data of the disclosed carbon emission factor to obtain a preset classification model.

[0106] Optional, carbon emission factor matching module includes:

[0107] a target data group determination unit, configured to perform multiple rounds of processing on the first dimensionality reduction data and the first carbon emission activity associated data using a preset classification model to obtain a preset number of target data groups that match the first carbon emission activity associated data, wherein the first dimensionality reduction data processed in the current round by the preset classification model does not include the target data group determined in the previous round;

[0108] The carbon emission factor determination unit is used to determine the carbon emission factor corresponding to each of the target data groups as a matching carbon emission factor, and determine the value of the to-be-determined carbon emission factor according to the matching carbon emission factor.

[0109] The carbon emission factor determination device provided in the embodiments of the present disclosure can execute the carbon emission factor determination method provided in any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

[0110] Embodiment 4

[0111] Figure 6 A schematic diagram of an electronic device 40 that can be used to implement an embodiment of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0112] like Figure 6 As shown, the electronic device 40 includes at least one processor 41, and a memory connected to the at least one processor 41, such as a read-only memory (ROM) 42, a random access memory (RAM) 43, etc., wherein the memory stores a computer program that can be executed by at least one processor, and the processor 41 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 42 or the computer program loaded from the storage unit 48 to the random access memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 can also be stored. The processor 41, the ROM 42, and the RAM 43 are connected to each other through a bus 44. An input / output (I / O) interface 45 is also connected to the bus 44.

[0113] A number of components in the electronic device 40 are connected to the I / O interface 45, including: an input unit 46, such as a keyboard, a mouse, etc.; an output unit 47, such as various types of displays, speakers, etc.; a storage unit 48, such as a disk, an optical disk, etc.; and a communication unit 49, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0114] The processor 41 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the processor 41 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 41 executes the various methods and processes described above, such as a carbon emission factor determination method.

[0115] In some embodiments, the carbon emission factor determination method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 48. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 40 via the ROM 42 and / or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the carbon emission factor determination method described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform the carbon emission factor determination method in any other appropriate manner (e.g., by means of firmware).

[0116] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0117] Computer programs for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0118] The computer device provided above can be used to execute the carbon emission factor determination method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0119] Embodiment 5

[0120] In the context of the present disclosure, a computer-readable storage medium may be a tangible medium, wherein the computer-executable instructions, when executed by a computer processor, are used to perform a carbon emission factor determination method, the method comprising:

[0121] Obtaining data associated with a first carbon emission activity for which a carbon emission factor is to be determined and data associated with a second carbon emission activity for which a carbon emission factor has been disclosed;

[0122] Using an adaptive Gaussian kernelized principal component analysis algorithm to perform dimensionality reduction processing on the second carbon emission activity associated data to obtain first dimensionality reduction data, wherein the first dimensionality reduction data includes a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor;

[0123] A preset classification model is used to determine a target data group matching the first carbon emission activity associated data from the carbon emission activity characteristic data group, and a value of the to-be-determined carbon emission factor is determined according to the carbon emission factor corresponding to the target data group.

[0124] In the context of the present disclosure, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by an instruction execution system, device, or equipment or used in conjunction with an instruction execution system, device, or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0125] The computer device provided above can be used to execute the carbon emission factor determination method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0126] It is worth noting that in the embodiment of the above-mentioned carbon emission factor determination device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this disclosure.

[0127] Note that the above are only preferred embodiments of the present disclosure and the technical principles used. Those skilled in the art will understand that the present disclosure is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present disclosure. Therefore, although the present disclosure is described in more detail through the above embodiments, the present disclosure is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present disclosure, and the scope of the present disclosure is determined by the scope of the attached claims.

Claims

1. A method for determining a carbon emission factor, characterized in that: include: Obtaining data associated with a first carbon emission activity for which a carbon emission factor is to be determined and data associated with a second carbon emission activity for which a carbon emission factor has been disclosed; Using an adaptive Gaussian kernelized principal component analysis algorithm to perform dimensionality reduction processing on the second carbon emission activity associated data to obtain first dimensionality reduction data, wherein the first dimensionality reduction data includes a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor; A preset classification model is used to determine a target data group matching the first carbon emission activity associated data from the carbon emission activity characteristic data group, and a value of the to-be-determined carbon emission factor is determined according to the carbon emission factor corresponding to the target data group.

2. The method according to claim 1, characterized in that Before obtaining the second carbon emission activity-related data for which the carbon emission factor has been disclosed, the method further includes: Obtaining the to-be-counted associated data of carbon emission activities corresponding to a plurality of disclosed carbon emission factors, wherein each of the to-be-counted associated data of the disclosed carbon emission factors includes multiple types of associated data; For each type of the associated data, determine a database modeling method according to the logical association between the current type of associated data, and use the database modeling method to model the associated data to be counted, so as to obtain a target database; The target database is used to generate second carbon emission activity related data, wherein the second carbon emission activity related data includes a matrix including the related data to be counted.

3. The method according to claim 1 or 2, characterized in that: The second carbon emission activity associated data includes the total amount of main raw materials input, the total amount of auxiliary raw materials input, the total amount of packaging of input raw materials, the total amount of energy supply required, the total amount of main products output, the total amount of by-products output, the total amount of waste output and the type of production process in the carbon emission activity corresponding to the disclosed carbon emission factor.

4. The method according to claim 1, characterized in that The step of reducing the dimension of the second carbon emission activity associated data by using an adaptive Gaussian kernelized principal component analysis algorithm to obtain first reduced dimension data includes: Using an adaptive Gaussian kernelized principal component analysis algorithm, mapping the second carbon emission activity associated data to a space of preset dimensions to obtain a to-be-determined reduced dimension matrix, wherein the preset dimension of the to-be-determined reduced dimension matrix is ​​smaller than the dimension of the second carbon emission activity associated data; Determine the ratio of the sum of the sample variances of the eigenvectors in the to-be-determined reduced-dimensionality matrix to the overall variance of the eigenvectors in the to-be-determined reduced-dimensionality matrix, and update the hyperfunction of the Gaussian kernel function according to the ratio to obtain a target hyperfunction; First dimensionality reduction data is generated according to the to-be-determined dimensionality reduction matrix and the target hyperfunction.

5. The method according to claim 4, characterized in that The step of updating the hyperfunction of the Gaussian kernel function according to the ratio to obtain a target hyperfunction includes: If the ratio is greater than a preset value, the current hyperfunction of the Gaussian kernel function is determined as the target hyperfunction; The step of generating first dimension reduction data according to the dimension reduction matrix to be determined and the target hyperfunction includes: The to-be-determined dimension reduction matrix corresponding to the target hyperfunction is determined as the first dimension reduction data.

6. The method according to claim 1, characterized in that Before determining the target data group matching the first carbon emission activity associated data from the carbon emission activity characteristic data group using the preset classification model, the method further includes: An initial classification model is generated using a random forest algorithm, and the initial classification model is trained using the second carbon emission activity associated data and the newly added third carbon emission activity associated data with disclosed carbon emission factors to obtain a preset classification model.

7. The method according to claim 1 or 6, characterized in that: The using a preset classification model to determine a target data group matching the first carbon emission activity associated data from the carbon emission activity characteristic data group, and determining the value of the to-be-determined carbon emission factor according to the carbon emission factor corresponding to the target data group, includes: Using a preset classification model to perform multiple rounds of processing on the first dimensionality reduction data and the first carbon emission activity associated data to obtain a preset number of target data groups that match the first carbon emission activity associated data, wherein the first dimensionality reduction data processed by the preset classification model in the current round does not include the target data group determined in the previous round; The carbon emission factor corresponding to each of the target data groups is determined as a matching carbon emission factor, and the value of the to-be-determined carbon emission factor is determined according to the matching carbon emission factor.

8. A device for determining a carbon emission factor, characterized in that: include: A data acquisition module, used to acquire first carbon emission activity-related data of a to-be-determined carbon emission factor and second carbon emission activity-related data of a disclosed carbon emission factor; a data dimension reduction module, configured to perform dimension reduction processing on the second carbon emission activity associated data using an adaptive Gaussian kernelized principal component analysis algorithm to obtain first dimension reduction data, wherein the first dimension reduction data includes a carbon emission activity characteristic data group corresponding to the disclosed carbon emission factor; A carbon emission factor matching module is used to determine a target data group that matches the first carbon emission activity associated data from the carbon emission activity characteristic data group using a preset classification model, and determine the value of the carbon emission factor to be determined based on the carbon emission factor corresponding to the target data group.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the carbon emission factor determination method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the carbon emission factor determination method according to any one of claims 1 to 7 when executed.

Images