A method, device, medium and equipment for attributing air temperature change driving factors

Abstract

The application discloses a kind of air temperature change driving factor attribution method, device, medium and equipment.Therein, method includes: the historical data obtained is re-sampled according to preset grid scale, and pretreatment data is obtained, wherein pretreatment data includes pretreatment air temperature data and pretreatment driving factor data;Respectively on pretreatment air temperature data and pretreatment driving factor data are carried out empirical orthogonal function decomposition, and air temperature decomposition data and driving factor decomposition data are obtained;XGBoot regression fitting is carried out to air temperature sequence of air temperature decomposition data and driving factor sequence of driving factor decomposition data, and XGBoost model is obtained;Based on air temperature spatial pattern of air temperature decomposition data, XGBoost model and the SHAP explainability decomposition of each driving factor to air temperature change for sample data for explanation, and the attribution result of each driving factor affecting air temperature change is obtained.

Description

Technical Field

[0001] This invention relates to the field of temperature change analysis technology, and more specifically, to a method, apparatus, medium, and equipment for attributing temperature change driving factors. Background Technology

[0002] Against the backdrop of global warming and anomalous atmospheric circulation, temperature change has become one of the most direct and critical response variables in the climate system. Whether it's surface temperature or temperature stratification at different troposphere levels, the variations in its annual amplitude, seasonal transitions, diurnal temperature range, and frequency of extreme events are all influenced by external forcing factors (such as greenhouse gas concentrations (CO2), volcanic eruptions, and solar activity), internal climate modes (such as ENSO, PDO, AO, and NAO), and human activities (such as land-use change, urban sprawl, and industrial emissions). Traditional temperature attribution methods for these multi-source drivers mainly include: multiple linear regression or stepwise regression; quantitative attribution based on CMIP model comparisons; and simple empirical mode decomposition (such as EOF / SSA) or sensitivity experiments. However, these methods have the following significant limitations in real complex climate systems: they cannot take into account both the spatial structure and temporal nonstationarity of temperature: traditional regression analysis is mainly based on time series and ignores the spatially coordinated change patterns of global or regional temperature fields; they have weak modeling capabilities for nonlinear and interaction effects: most methods are based on linear assumptions and are difficult to reveal the interaction relationships between complex variables; and the attribution results are difficult to interpret and revert to the actual physical space: even if they can provide overall contribution values, it is difficult to construct interpretable maps down to each grid point. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a method, apparatus, medium, and device for attributing driving factors of temperature change.

[0004] According to one aspect of the present invention, a method for attributing drivers of temperature change by integrating empirical orthogonal decomposition and interpretable machine learning is provided, comprising: The acquired historical data is resampled according to a preset grid scale to obtain preprocessed data, which includes preprocessed temperature data and preprocessed driving factor data. Empirical orthogonal function decomposition was performed on the preprocessed temperature data and the preprocessed driving factor data respectively to obtain temperature decomposition data and driving factor decomposition data. XGBoot regression fitting was performed on the temperature series of the temperature decomposition data and the driving factor series of the driving factor decomposition data to obtain the XGBoost model. Based on the temperature spatial model, XGBoost model, and sample data used for interpretation, SHAP interpretability decomposition was performed on each driving factor of temperature change to obtain the attribution results of each driving factor affecting temperature change.

[0005] According to another aspect of the present invention, a device for attributing temperature change driving factors by fusing empirical orthogonal decomposition and interpretable machine learning is provided, comprising: The sampling module is used to resample the acquired historical data according to a preset grid scale to obtain preprocessed data, which includes preprocessed temperature data and preprocessed driving factor data. The decomposition module is used to perform empirical orthogonal function decomposition on the preprocessed temperature data and the preprocessed driving factor data respectively to obtain temperature decomposition data and driving factor decomposition data. The fitting module is used to perform XGBoot regression fitting on the temperature sequence of the temperature decomposition data and the driving factor sequence of the driving factor decomposition data to obtain the XGBoost model. The interpretation decomposition module is used to perform SHAP interpretability decomposition on the various driving factors of temperature change based on the temperature spatial pattern, XGBoost model and sample data used for interpretation of temperature decomposition data, and to obtain the attribution results of each driving factor affecting temperature change.

[0006] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing a computer program for performing the methods described in any of the above aspects of the present invention.

[0007] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: a processor; a memory for storing executable instructions of the processor; the processor being configured to read the executable instructions from the memory and execute the instructions to implement the method described in any of the preceding aspects of the present invention.

[0008] Therefore, this invention proposes a multi-source temperature attribution analysis method that integrates empirical orthogonal decomposition, XGBoost nonlinear modeling, and SHAP interpretability mechanism. Its core innovations include: (1) a spatiotemporally unified mode decomposition mechanism (EOF): using EOF technology to decompose the multidimensional temperature field and driving factor field into finite spatial modes and temporal principal components (PC), while compressing dimensions, retaining key climate mode features and establishing physically interpretable factor expressions; (2) nonlinear multi-output regression modeling (XGBoost): using XGBoost in the principal component space to establish a nonlinear mapping relationship between temperature and multi-source driving factors, automatically capturing the interaction enhancement and saturation effects between variables, and improving the accuracy of attribution fitting; (3) contribution decomposition and grid reconstruction (SHAP+). (4) Support for multi-scale attribution analysis: By selecting different time resolutions (month, season, year) and the number of EOF modes, attribution analysis of temperature (not limited to temperature) under multi-scale backgrounds such as medium- and long-term changes and extreme events can be flexibly realized. Attached Figure Description

[0009] Exemplary embodiments of the present invention can be more fully understood by referring to the following figures: Figure 1 This is a flowchart illustrating an exemplary embodiment of the present invention for a method of attributing temperature change driving factors by fusing empirical orthogonal decomposition and interpretable machine learning. Figure 2 This is a schematic diagram of the structure of an attribution device for temperature change driving factors that integrates empirical orthogonal decomposition and interpretable machine learning, provided in an exemplary embodiment of the present invention. Figure 3 This is the structure of an electronic device provided in an exemplary embodiment of the present invention. Detailed Implementation

[0010] Hereinafter, exemplary embodiments according to the present invention will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments of the present invention. It should be understood that the present invention is not limited to the exemplary embodiments described herein.

[0011] It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of the invention.

[0012] Those skilled in the art will understand that the terms "first," "second," etc., in the embodiments of the present invention are only used to distinguish different steps, devices, or modules, and do not represent any specific technical meaning, nor do they indicate a necessary logical order between them.

[0013] It should also be understood that in the embodiments of the present invention, "multiple" can refer to two or more, and "at least one" can refer to one, two or more.

[0014] It should also be understood that any component, data or structure mentioned in the embodiments of the present invention can generally be understood as one or more unless explicitly defined or given contrary instructions in the context.

[0015] Furthermore, the term "and / or" in this invention is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this invention generally indicates that the preceding and following related objects have an "or" relationship.

[0016] It should also be understood that the description of the various embodiments in this invention emphasizes the differences between the various embodiments, and the similarities or similarities can be referred to each other. For the sake of brevity, they will not be described in detail.

[0017] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0018] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0019] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, they should be considered part of the specification.

[0020] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0021] The embodiments of this invention can be applied to electronic devices such as terminal devices, computer systems, and servers, and can operate together with a wide range of other general-purpose or special-purpose computing system environments or configurations. Well-known examples of terminal devices, computing systems, environments, and / or configurations suitable for use with electronic devices such as terminal devices, computer systems, and servers include, but are not limited to: personal computer systems, server computer systems, thin clients, thick clients, handheld or laptop devices, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputer systems, mainframe computer systems, and distributed cloud computing environments including any of the above systems, etc.

[0022] Electronic devices such as terminal devices, computer systems, and servers can be described in the general context of computer system executable instructions (such as program modules) executed by a computer system. Typically, program modules can include routines, programs, object programs, components, logic, data structures, etc., which perform specific tasks or implement specific abstract data types. Computer systems / servers can be implemented in distributed cloud computing environments, where tasks are executed by remote processing devices linked through communication networks. In distributed cloud computing environments, program modules can reside on local or remote computing system storage media, including storage devices.

[0023] Exemplary methods Figure 1 This is a flowchart illustrating an exemplary embodiment of the present invention regarding a method for attributing temperature change drivers by fusing empirical orthogonal decomposition and interpretable machine learning. This embodiment can be applied to electronic devices, such as… Figure 1 As shown, the temperature change driver attribution method 100, which integrates empirical orthogonal decomposition and interpretable machine learning, includes the following steps: Step 101: Resample the acquired historical data according to a preset grid scale to obtain preprocessed data, which includes preprocessed temperature data and preprocessed driving factor data. Step 102: Perform empirical orthogonal function decomposition on the preprocessed temperature data and the preprocessed driving factor data respectively to obtain temperature decomposition data and driving factor decomposition data. Step 103: Perform XGBoot regression fitting on the temperature sequence of the temperature decomposition data and the driving factor sequence of the driving factor decomposition data to obtain the XGBoost model. Step 104: Based on the temperature spatial pattern of the temperature decomposition data, the XGBoost model, and the sample data used for interpretation, perform SHAP interpretability decomposition on each driving factor of temperature change to obtain the attribution results of each driving factor affecting temperature change.

[0024] Specifically, addressing the technical problems existing in the background technology, this invention proposes a multi-source temperature attribution analysis method that integrates empirical orthogonal decomposition (EOF), XGBoost nonlinear modeling, and SHAP interpretability mechanism. Its core innovations include: (1) a spatiotemporally unified mode decomposition mechanism (EOF): using EOF technology to decompose the multidimensional temperature field and driving factor field into finite spatial modes and temporal principal components (PC), while compressing dimensions, retaining key climate mode features and establishing physically interpretable factor expressions; (2) nonlinear multi-output regression modeling (XGBoost): using XGBoost in the principal component space to establish a nonlinear mapping relationship between temperature and multi-source driving factors, automatically capturing the interaction enhancement and saturation effects between variables, and improving the accuracy of attribution fitting; (3) contribution decomposition and grid reconstruction (SHAP+ (EOF inverse mapping): Combining the SHAP framework, the local contribution of each factor principal component to temperature change at different times is quantified, and the contribution is mapped back to the original grid space based on the EOF spatial model to generate a high-resolution, visualized attribution map of driving factors; (4) Support for multi-scale attribution analysis: By selecting different time resolutions (month, season, year) and the number of EOF modes, attribution analysis of temperature (not limited to temperature) under multi-scale backgrounds such as medium- and long-term changes and extreme events can be flexibly realized. The specific implementation is as follows: 1. Preprocess all raw data (including but not limited to: CMIP6 carbon dioxide concentration (CO2), stratospheric aerosol optical thickness (VAOD), ERA5 temperature (TEM), precipitation (TP), NOAA AVHRR normalized difference vegetation index (NDVI), DMSP / VIIRS stable light (NDB)) by: (1) uniformly cropping the data to a global grid; (2) synthesizing all non-monthly scale products into monthly scale products; (3) resampling the grid data to a 0.25°×0.25° grid. All preprocessed factors (divided into temperature and driving factors) are then subjected to the next step of empirical orthogonal function (EOF) decomposition.

[0025] 2. Empirical Orthogonal Function (EOF) decomposition is a classic dimensionality reduction and mode extraction technique that can decompose high-dimensional lattice fields into a few physically meaningful "spatial modes" and their corresponding "temporal principal components (PCs)". In multi-source attribution analysis, this scheme first performs EOF decomposition on temperature and each driving factor separately, compressing the original multidimensional space-time data into several PC sequences and spatial modes, so that subsequent XGBoost fitting and SHAP interpretation can quantify the contribution of each driving factor to temperature change in a low-dimensional feature space.

[0026] Temperature after pretreatment Passing the exam l seed driving factors Its EOF decomposition can be uniformly written as:

[0027]

[0028] Parameter description: i , j For spatial grid indexing; For time indexing; m The number of EOF models used for temperature; k l For the first l The number of EOF modes taken by each driving factor; For the temperature p Temporal principal components of the pattern; For the temperature p The spatial EOF mode of the pattern; For the first l The driving factor is the first q Temporal principal components of the pattern; For the first l Driving factor number p The spatial EOF mode of the pattern; For the temperature residual field; for l Individual residual fields.

[0029] 3. XGBoost Regression Fitting: XGBoost is based on the Gradient Boosting Decision Tree (GBDT) framework. It iteratively generates multiple weak learners (decision trees) to gradually correct the prediction errors of the preceding model. The PC time series of all driving factors are concatenated into a feature vector. Temperature PCs as multiple output targets The training objective is to minimize the squared error loss.

[0030] In the formula, L The number of driving factor types; The first driving factor is the first mode of the time principal component; For the first factor k Temporal principal components of mode 1; For the first l The time principal components of the first mode of the driving factors; For the first l The factor of the first k Temporal principal components of mode 1; For the first L The time principal components of the first mode of the driving factors; For the first L The factor of the firstk L Temporal principal components of the pattern; The first principal component of the model predicts the temperature. The second principal component of the model predicts the temperature. The temperature predicted by the model is the first m Principal components of the model; For XGBoost multi-output regression functions; This is a set of model hyperparameters, such as the number of trees (n_estimators), maximum depth (max_depth), step size (learning_rate), row sampling rate (subsample), column sampling rate (colsample_bytree), etc.

[0031] 4. SHAP interpretability decomposition, mapping SHAP contributions back to spatial patterns, grid-level attribution reconstruction: for the trained XGBoost model At every moment For each output mode and input factors Perform SHAP value decomposition; map SHAP contributions back to spatial patterns to obtain driving factors. No. Pattern to grid At any moment Contributions, summarizing driving factors At grid points Total contribution.

[0032]

[0033] Parameter description: As driving factor l No. q Pattern at grid ,time t Attribution contribution; : driving factor l No. q The index of the pattern in the feature vector The obtained SHAP value; For the temperature p The spatial EOF mode of the pattern; m The number of EOF models extracted for temperature.

[0034] Summary of driving factors At grid points The total contribution, summed across all its modes:

[0035] Parameter description: For the first l The number of EOF modes truncated by the driving factor; As driving factor l No. q Pattern at grid ,time t Attribution contribution; As driving factor l At grid points ,time t The overall contribution is attributed to the above.

[0036] Therefore, this invention proposes a multi-source temperature attribution analysis method that integrates empirical orthogonal decomposition, XGBoost nonlinear modeling, and SHAP interpretability mechanism. Its core innovations include: (1) a spatiotemporally unified mode decomposition mechanism (EOF): using EOF technology to decompose the multidimensional temperature field and driving factor field into finite spatial modes and temporal principal components (PC), while compressing dimensions, retaining key climate mode features and establishing physically interpretable factor expressions; (2) nonlinear multi-output regression modeling (XGBoost): using XGBoost in the principal component space to establish a nonlinear mapping relationship between temperature and multi-source driving factors, automatically capturing the interaction enhancement and saturation effects between variables, and improving the accuracy of attribution fitting; (3) contribution decomposition and grid reconstruction (SHAP+). (4) Support for multi-scale attribution analysis: By selecting different time resolutions (month, season, year) and the number of EOF modes, attribution analysis of temperature (not limited to temperature) under multi-scale backgrounds such as medium- and long-term changes and extreme events can be flexibly realized.

[0037] Exemplary device Figure 2 This is a schematic diagram of the structure of an attribution device for temperature change driving factors that integrates empirical orthogonal decomposition and interpretable machine learning, provided in an exemplary embodiment of the present invention. Figure 2 As shown, the device 200 includes: The sampling module 210 is used to resample the acquired historical data according to a preset grid scale to obtain preprocessed data, wherein the preprocessed data includes preprocessed temperature data and preprocessed driving factor data. The decomposition module 220 is used to perform empirical orthogonal function decomposition on the preprocessed temperature data and the preprocessed driving factor data respectively to obtain temperature decomposition data and driving factor decomposition data. The fitting module 230 is used to perform XGBoot regression fitting on the temperature sequence of the temperature decomposition data and the driving factor sequence of the driving factor decomposition data to obtain the XGBoost model. The interpretation decomposition module 240 is used to perform SHAP interpretability decomposition on each driving factor of temperature change based on the temperature spatial pattern, XGBoost model and sample data used for interpretation of temperature decomposition data, and to obtain the attribution results of each driving factor affecting temperature change.

[0038] Exemplary electronic devices Figure 3 This is the structure of an electronic device provided in an exemplary embodiment of the present invention. For example... Figure 3 As shown, the electronic device 30 includes one or more processors 31 and memory 32.

[0039] The processor 31 may be a central processing unit (CPU) or other form of processing unit with data processing and / or instruction execution capabilities, and may control other components in the electronic device to perform desired functions.

[0040] The memory 32 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 31 may execute the program instructions to implement the methods of the software programs of the various embodiments of the present invention described above, and / or other desired functions. In one example, the electronic device may also include an input device 33 and an output device 34, these components being interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0041] In addition, the input device 33 may also include, for example, a keyboard, a mouse, etc.

[0042] The output device 34 can output various information to the outside. The output device 34 may include, for example, a display, a speaker, a printer, and a communication network and its connected remote output devices, etc.

[0043] Of course, for the sake of simplicity, Figure 3 Only some of the components of this electronic device relevant to the present invention are shown, omitting components such as buses, input / output interfaces, etc. In addition, the electronic device may include any other suitable components depending on the specific application.

[0044] Exemplary computer program products and computer-readable storage media In addition to the methods and apparatus described above, embodiments of the present invention may also be computer program products, which include computer program instructions that, when executed by a processor, cause the processor to perform the steps in the methods according to various embodiments of the present invention described in the "Exemplary Methods" section above.

[0045] The computer program product can be written in any combination of one or more programming languages ​​to perform the operations of the embodiments of the present invention. The programming languages ​​include object-oriented programming languages ​​such as Java and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computing device, partially on the user's computing device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0046] Furthermore, embodiments of the present invention may also be computer-readable storage media storing computer program instructions thereon, which, when executed by a processor, cause the processor to perform the steps of the methods according to various embodiments of the present invention described in the "Exemplary Methods" section above.

[0047] The computer-readable storage medium may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0048] The basic principles of the present invention have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in the present invention are merely examples and not limitations, and should not be considered as essential features of each embodiment of the present invention. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the present invention to the necessity of employing the aforementioned specific details.

[0049] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For system embodiments, since they largely correspond to method embodiments, the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments.

[0050] The block diagrams of devices, systems, devices, and systems involved in this invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, systems, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0051] The methods and systems of the present invention may be implemented in many ways. For example, they may be implemented by software, hardware, firmware, or any combination of software, hardware, and firmware. The above-described order of steps for the methods is for illustrative purposes only, and the steps of the methods of the present invention are not limited to the order specifically described above unless otherwise specifically stated. Furthermore, in some embodiments, the present invention may also be implemented as a program recorded on a recording medium, the program comprising machine-readable instructions for implementing the methods according to the present invention. Thus, the present invention also covers recording media storing programs for performing the methods according to the present invention.

[0052] It should also be noted that in the systems, apparatus, and methods of the present invention, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered equivalents of the present invention. The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the invention. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the invention. Therefore, the invention is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0053] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the invention to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A method for attributing drivers of temperature change by integrating empirical orthogonal decomposition and interpretable machine learning, characterized in that, include: The acquired historical data is resampled according to a preset grid scale to obtain preprocessed data, which includes preprocessed temperature data and preprocessed driving factor data. The preprocessed temperature data and the preprocessed driving factor data are respectively decomposed using empirical orthogonal functions to obtain temperature decomposition data and driving factor decomposition data. XGBoot regression fitting is performed on the temperature sequence of the temperature decomposition data and the driving factor sequence of the driving factor decomposition data to obtain the XGBoost model; Based on the temperature spatial pattern of the temperature decomposition data, the XGBoost model, and the sample data used for interpretation, SHAP interpretability decomposition is performed on each driving factor of temperature change to obtain the attribution results of each driving factor affecting temperature change.

2. The method according to claim 1, characterized in that, The empirical orthogonal function decomposition expression for the preprocessed temperature data is: The empirical orthogonal function decomposition expression for the preprocessed driving factor data is: In the formula, i , j For spatial grid indexing; For time indexing; m The number of EOF models used for temperature; k l For the first l The number of EOF modes taken by each driving factor; For the temperature p Temporal principal components of the pattern; For the temperature p The spatial EOF mode of the pattern; For the first l The driving factor is the first q Temporal principal components of the pattern; For the first l Driving factor number p The spatial EOF mode of the pattern; For the temperature residual field; for l Individual residual fields.

3. The method according to claim 2, characterized in that, The expression for the XGBoot regression fit is: In the formula, L The number of driving factor types; The first driving factor is the first mode of the time principal component; For the first factor k Temporal principal components of mode 1; For the first l The time principal components of the first mode of the driving factors; For the first l The factor of the first k Temporal principal components of mode 1; For the first L The time principal components of the first mode of the driving factors; For the first L The factor of the first k L Temporal principal components of the pattern; The first principal component of the model predicts the temperature. The second principal component of the model predicts the temperature. The temperature predicted by the model is the first m Principal components of the model; For XGBoost multi-output regression functions; This is the set of model hyperparameters.

4. The method according to claim 3, characterized in that, Based on the temperature spatial pattern of the temperature decomposition data, the XGBoost model, and the sample data used for interpretation, SHAP interpretability decomposition is performed on each driving factor of temperature change to obtain the attribution results of each driving factor affecting temperature change, including: Based on the spatial patterns of temperature in the temperature decomposition data and the sample data used for interpretation, the trained XGBoost model is trained at each time step. t For each output mode p and input driving factors Perform SHAP value decomposition to obtain the driving factors at each grid point (i,j) at time t. t Contributions; Based on the driving factors at each grid point (i,j) at time... t The total contribution of each driving factor at grid point (i,j) is summarized to obtain the attribution results of each driving factor affecting temperature change.

5. The method according to claim 4, characterized in that, The formula for calculating the total contribution of each driving factor at grid point (i,j) is as follows: in, In the formula, As driving factor l No. q Pattern at grid ,time t Attribution contribution; : driving factor l No. q The index of the pattern in the feature vector The obtained SHAP value; For the temperature p The spatial EOF mode of the pattern; m The number of EOF models extracted for temperature; For the first l The number of EOF modes truncated by the driving factor; As driving factor l At grid points ,time t The overall contribution is attributed to the above.

6. A device for attributing temperature change driving factors by integrating empirical orthogonal decomposition and interpretable machine learning, characterized in that, include: The sampling module is used to resample the acquired historical data according to a preset grid scale to obtain preprocessed data, wherein the preprocessed data includes preprocessed temperature data and preprocessed driving factor data. The decomposition module is used to perform empirical orthogonal function decomposition on the preprocessed temperature data and the preprocessed driving factor data respectively to obtain temperature decomposition data and driving factor decomposition data. The fitting module is used to perform XGBoot regression fitting on the temperature sequence of the temperature decomposition data and the driving factor sequence of the driving factor decomposition data to obtain the XGBoost model. The interpretation decomposition module is used to perform SHAP interpretability decomposition on each driving factor of temperature change based on the temperature spatial pattern of the temperature decomposition data, the XGBoost model, and the sample data used for interpretation, so as to obtain the attribution results of each driving factor affecting temperature change.

7. The apparatus according to claim 6, characterized in that, The empirical orthogonal function decomposition expression for the preprocessed temperature data is: The empirical orthogonal function decomposition expression for the preprocessed driving factor data is: In the formula, i , j For spatial grid indexing; For time indexing; m The number of EOF models used for temperature; k l For the first l The number of EOF modes taken by each driving factor; For the temperature p Temporal principal components of the pattern; For the temperature p The spatial EOF mode of the pattern; For the first l The driving factor is the first q Temporal principal components of the pattern; For the first l Driving factor number p The spatial EOF mode of the pattern; For the temperature residual field; for l Individual residual fields.

8. The apparatus according to claim 7, characterized in that, The expression for the XGBoot regression fit is: In the formula, L The number of driving factor types; The first driving factor is the first mode of the time principal component; For the first factor k Temporal principal components of mode 1; For the first l The time principal components of the first mode of the driving factors; For the first l The factor of the first k Temporal principal components of mode 1; For the first L The time principal components of the first mode of the driving factors; For the first L The factor of the first k L Temporal principal components of the pattern; The first principal component of the model predicts the temperature. The second principal component of the model predicts the temperature. The temperature predicted by the model is the first m Principal components of the model; For XGBoost multi-output regression functions; This is the set of model hyperparameters.

9. A computer-readable storage medium, characterized in that, The storage medium stores a computer program for performing the method described in any one of claims 1-5.

10. An electronic device, characterized in that, The electronic device includes: processor; Memory used to store the processor's executable instructions; The processor is configured to read the executable instructions from the memory and execute the instructions to implement the method described in any one of claims 1-5.

Images