A method and system for constructing a user electricity carbon emission portrait based on space-time distribution characteristics

Abstract

The application is suitable for the technical field of carbon emission, and provides a construction method and system of user electricity carbon emission portrait based on space-time distribution characteristics, which comprises the following steps: obtaining information data of a target user, historical electricity data of the target user and power grid generation structure data, and preprocessing the obtained data; extracting target feature data related to time and target feature data related to regional distribution based on the preprocessed data; analyzing time variation characteristics and regional distribution variation characteristics of carbon emission factors of the target user according to the target feature data related to time and the target feature data related to regional distribution respectively; and constructing a dynamic portrait of electricity carbon emission of the target user according to the time variation characteristics and the regional distribution variation characteristics of the carbon emission factors of the target user. The application can accurately reflect the electricity carbon emission of the user under different time and regional conditions, and can provide accurate data for subsequent work.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of carbon emission, and particularly relates to a method and system for constructing a user electricity carbon emission portrait based on space-time distribution characteristics. BACKGROUND

[0002] At present, most power companies can understand the electricity consumption mode and carbon emission of different user groups through the portrait, so as to optimize the power grid planning and power dispatching through accurate observation of the portrait. For government departments, the portrait can provide a strong basis for formulating energy and environmental policies, and scientific and reasonable energy-saving and emission-reducing targets can be set accordingly, and different types of users can be implemented differentiated supervision. From the perspective of users, the carbon emission portrait can enable users to intuitively understand the influence of their own user behavior on the environment, so that they can adjust their electricity consumption habits or replace energy-saving equipment, so as to achieve the purpose of reducing electricity cost and reducing carbon emission.

[0003] In the prior art, the time dimension rich information is ignored in the user electricity carbon emission portrait, and the change of user electricity behavior under different time scales is not considered, such as the significant difference between workdays and holidays, the difference between electricity peak and valley in different time periods of a day, and the influence of seasonal change on electricity demand and carbon emission. In addition, in the spatial dimension, factors such as the geographical location of the user, the functional attribute of the region, the distribution of energy infrastructure in the region, and the spatial aggregation degree of the user are not considered. This leads to the inability to depict the spatial characteristics of carbon emission of users in different regions, so that a comprehensive and accurate user carbon emission portrait cannot be constructed, and the problem of inaccurate and lack of comprehensive user electricity carbon emission portrait occurs.

[0004] Therefore, a method and system for constructing a user electricity carbon emission portrait based on space-time distribution characteristics are needed. SUMMARY

[0005] The embodiments of the present application provide a method and system for constructing a user electricity carbon emission portrait based on space-time distribution characteristics, which are used to solve the problem of inaccurate and lack of comprehensive user electricity carbon emission portrait.

[0006] The first aspect of the embodiments of the present application provides a method for constructing a user electricity carbon emission portrait based on space-time distribution characteristics, comprising:

[0007] Obtaining information data of a target user, historical electricity consumption data of the target user, and power grid generation structure data, and preprocessing the obtained data;

[0008] Extracting target feature data related to time and target feature data related to regional distribution based on the preprocessed data;

[0009] analyze time variation characteristics and regional distribution variation characteristics of the carbon emission factor of the target user according to the target feature data related to time and the target feature data related to regional distribution respectively;

[0010] construct a dynamic image of the carbon emission of the target user according to the time variation characteristics and the regional distribution variation characteristics of the carbon emission factor of the target user.

[0011] Further, the extracting the target feature data related to time and the target feature data related to regional distribution based on the preprocessed data comprises:

[0012] determining the feature vector related to time and the feature vector related to regional distribution respectively by using principal component analysis method;

[0013] determining the target feature data related to time and the target feature data related to regional distribution according to the scores on the time principal component and the regional principal component.

[0014] Further, the determining the target feature data related to time and the target feature data related to regional distribution according to the scores on the time principal component and the regional principal component comprises:

[0015] time principal component score :

[0016]

[0017] wherein: is a row vector of the standardized time series data matrix of the i-th user sample, is a corresponding feature vector; regional principal component score

[0018] :

[0019] wherein:

[0020] is a row vector of the standardized regional data matrix of the i-th user sample, is a corresponding feature vector. Further, the analyzing the time variation characteristics and the regional distribution variation characteristics of the carbon emission factor of the target user according to the target feature data related to time and the target feature data related to regional distribution respectively comprises:

[0021]

[0022] ​​The virtual variables are constructed based on the dependent variable and the target feature data related to time, and the virtual variables are constructed based on the dependent variable and the target feature data related to geographical distribution.

[0023] An initial time virtual variable model is determined based on the virtual variables constructed based on the dependent variable and the target feature data related to time, and a space virtual variable model is determined based on the virtual variables constructed based on the dependent variable and the target feature data related to geographical distribution.

[0024] The coefficients of the time virtual variable model and the coefficients of the space virtual variable model are calculated by using a least square method.

[0025] The target time virtual variable model and the target space virtual variable model are determined based on the calculated coefficients.

[0026] Further, the coefficients of the time virtual variable model and the coefficients of the space virtual variable model are calculated by using a least square method, including:

[0027] The coefficients of the time virtual variable model :

[0028]

[0029]

[0030]

[0031]

[0032] wherein: is the number of observation values time virtual variables in the observation values

[0033] The coefficients of the space virtual variable model :

[0034]

[0035]

[0036]

[0037]

[0038] wherein: is the number of observation values space virtual variables in the observation values

[0039] ​​Further, the determining the target time dummy variable model and the target space dummy variable model according to the calculated coefficients respectively comprises:

[0040] The target time dummy variable model comprises:

[0041]

[0042] Wherein: is a time factor intercept term, is a time dummy variable corresponding coefficient, is a time factor error term.

[0043] Further, the determining the target time dummy variable model and the target space dummy variable model according to the calculated coefficients respectively comprises:

[0044] The target space dummy variable model comprises:

[0045]

[0046] Wherein: is a space factor intercept term, is a space dummy variable corresponding coefficient, is a space factor error term.

[0047] Further, the constructing the dynamic portrait of the carbon emission of the electricity consumption of the target user according to the time variation characteristics and the regional distribution variation characteristics of the carbon emission factor of the target user comprises:

[0048] The time dummy variable coefficient is determined according to the target time dummy variable model, and the variation of the dummy variable coefficient in a preset time is predicted by combining a time series analysis method;

[0049] The space dummy variable coefficient is determined according to the target space dummy variable model, and the influence of the external factors related to the region on the coefficient is analyzed;

[0050] The first dynamic portrait of the carbon emission of the electricity consumption of the user is constructed based on the time coefficient variation and the prediction result, and the second dynamic portrait of the carbon emission of the electricity consumption of the user is constructed based on the space coefficient variation and the external factors;

[0051] The target dynamic portrait of the carbon emission of the electricity consumption of the user is determined according to the first dynamic portrait and the second dynamic portrait.

[0052] Further, the determining the target dynamic portrait of the carbon emission of the electricity consumption of the user according to the first dynamic portrait and the second dynamic portrait comprises:

[0053] The comprehensive model expression of the target dynamic image is as follows:

[0054]

[0055] wherein: is a coefficient of the interaction term .

[0056] The second aspect of the embodiment of the application provides a user electricity carbon emission image construction system based on space-time distribution characteristics, comprising:

[0057] a data acquisition and preprocessing unit configured to acquire information data of a target user, historical electricity consumption data of the target user, and power grid generation structure data, and preprocess the acquired data;

[0058] a target feature data extraction unit configured to extract time-related target feature data and regionally distributed target feature data based on the preprocessed data;

[0059] a time variation feature and regionally distributed variation feature analysis unit configured to analyze time variation features and regionally distributed variation features of carbon emission factors of the target user according to the time-related target feature data and the regionally distributed target feature data, respectively;

[0060] a dynamic image construction unit configured to construct a dynamic image of electricity carbon emission of the target user according to the time variation features and the regionally distributed variation features of the carbon emission factors of the target user.

[0061] As can be seen from the above technical solutions, the embodiment of the application has the following advantages:

[0062] The application collects information data, historical electricity consumption data, and power grid generation structure data of a target user, preprocesses these data, and then extracts time-related target feature data and regionally distributed target feature data. On this basis, the variation features of carbon emission factors of the target user are analyzed from two dimensions of time and region, and finally a dynamic image of electricity carbon emission of the target user is constructed according to the variation features. The target dynamic image can accurately reflect the electricity carbon emission of the target user under different time and regional conditions, and provides strong data support and scientific basis for electric power companies to formulate individualized energy service strategies and for governments to implement accurate carbon emission reduction policies.

[0063] Other advantages, objects, and features of the application will be set forth in part in the following specification, and in part will become apparent to those skilled in the art from the examination of the following, or can be learned from practice of the application. The objectives and other advantages of the application can be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings. BRIEF DESCRIPTION OF DRAWINGS

[0064] Fig. 1 A flowchart of one embodiment of the method for constructing a user electricity carbon emission portrait based on spatial and temporal distribution characteristics;

[0065] Fig. 2 Another flowchart of one embodiment of the method for constructing a user electricity carbon emission portrait based on spatial and temporal distribution characteristics;

[0066] Fig. 3 Another flowchart of one embodiment of the method for constructing a user electricity carbon emission portrait based on spatial and temporal distribution characteristics. DETAILED DESCRIPTION

[0067] The terms "first", "second", "third", "fourth" and the like in the description and claims of the present application and in the above summary of the application, if any, are used for distinguishing between similar objects talking about the application and do not necessarily have to appear in the application in this special order or used to describe a special sequential or chronological order. It is to be understood that the use of these terms is merely for distinguishing between the similar objects talking about the application and that these terms are not necessarily used to describe a special sequential or chronological order among the execution of the steps or units or to signify a special sequential or chronological order of executing the steps or units. Furthermore, the terms "comprising", "containing", "having" and "including" and their conjugates, as used herein, are intended to cover the respective terms and their conjugates as if each of the terms and their conjugates were individually and alternately used throughout the application. The use of the negative "not" in the claims is intended to mean the absence of the feature or characteristics, not an exclusion of the features or characteristics.

[0068] Embodiment one

[0069] The method implemented in this embodiment can be implemented in a system, and can be implemented in a server or a terminal, and the specific implementation is not limited. The photovoltaic station multi-element device fault ride-through coordination control method provided in the present application will be introduced from the perspective of system implementation. Please refer to Figs. 1-3 The method provided in the embodiment of the present application includes the following steps:

[0070] S11. Obtain information data of a target user, historical electricity consumption data of the target user, and power grid power generation structure data, and pre-process the obtained data;

[0071] In this embodiment, the user information data includes geographic location information data, user type, business type of commercial users, production type of industrial users, population, business site size, factory production scale, and number of employees, and the like. The region, industry to which the target user belongs, and group type of the target user are determined by obtaining the user information data.

[0072] The historical power consumption data includes power consumption data, power consumption time data and power consumption power data, wherein the power consumption data includes time resolution data and cumulative power consumption, such as power consumption recorded at different time intervals, total power consumption in a certain period. The power consumption time data includes daily power consumption time distribution, weekly power consumption time rule and seasonal power consumption time change. The power consumption power data includes the power consumption power of the user at each time and the utilization efficiency of the user's power consumption equipment to the electric energy.

[0073] The power grid power generation structure data includes power generation method and proportion data, power generation time sequence data and power grid connection transmission data, the power generation method and proportion data includes the proportion of various power generation methods such as coal-fired power generation, hydropower generation, wind power generation, solar power generation, nuclear power generation and natural gas power generation in the total power generation of the power grid, and the contribution of the power generation of large power stations and distributed power generation facilities in the power grid. The power generation time sequence data includes the power generation time change of each power generation method in a day, a week or a month, and the power generation plan made by the power grid dispatching department according to the power demand prediction and the power generation resource situation. The power grid connection transmission data includes the connection mode between different power grids, the power exchange amount and the exchange time and other information, and the loss rate, transmission distance and other data of different power transmission lines.

[0074] The above data obtained is subjected to data preprocessing including data cleaning and standardization, checking whether the user power consumption data and the power grid power generation structure data are complete, and for the missing data points, the appropriate method is used to fill according to the characteristics and distribution of the data. The data of different orders of magnitude is standardized to make all the data in the same order of magnitude. For example, for the power consumption data and the power generation proportion data, the Z-score standardization method is used here to convert the data into a standard normal distribution with a mean of 0 and a standard deviation of 1.

[0075] S12. Extracting time-related target feature data and regionally distributed target feature data based on the preprocessed data;

[0076] Step S12 includes the following steps:

[0077] 1. Determine the time-related feature vector and the regionally distributed feature vector respectively by using principal component analysis method;

[0078] 2. Determine the time-related target feature data and the regionally distributed target feature data according to the scores on the time principal components and the region principal components.

[0079] Specifically, the preprocessed time sequence data is arranged into a data matrix , wherein the rows represent different user samples and the columns represent different time point variables. According to the standardization formula The standardization process is performed to obtain the standardized matrix. Calculate the covariance matrix. ,in This represents the number of user samples. The covariance matrix reflects the covariance relationship between variables at different time points, i.e., the correlation of time series data at different time points. Perform eigenvalue decomposition to obtain eigenvalues. , ,…, ,in The number of variables at time points, and the corresponding feature vectors are: , ,…, For example, the first eigenvector corresponds to the largest eigenvalue, indicating that the variance of the time series data is largest in that direction, which may correspond to the main trend of electricity consumption within a day. Principal components are selected based on the cumulative variance contribution rate; assuming the first eigenvector is selected... Each principal component makes Reaching the set ratio, Each principal component represents the dominant temporal pattern in the time series data. For each user sample In the The scoring formula for each principal component over time is as follows:

[0080]

[0081] in: For the first The row vectors of each user sample in the standardized time series data matrix.

[0082] These scores constitute new time-related feature data, such as representing a user's performance in primary time patterns.

[0083] Similarly, the preprocessed regional data is organized into a data matrix. The rows represent different user samples, and the columns represent different region-related variables. The data is processed according to a standardized formula. The standardization process is performed to obtain the standardized matrix. Calculate the covariance matrix. The covariance matrix reflects the covariance relationship between variables from different regions, that is, the correlation between regionally related data. Perform eigenvalue decomposition to obtain eigenvalues. , ,…, ,in The number of variables at time points, and the corresponding feature vectors are: , ,…, The first eigenvector corresponds to the largest eigenvalue, indicating the largest variance in the regional data along that direction. This direction may correspond to the main differences between regions. Principal components are selected based on the cumulative variance contribution rate. Assuming the first eigenvector is selected... Each principal component makes Reaching the set ratio, Each principal component represents a major regional variation pattern in the regional data. For each user sample In the The scoring formula for each principal component over time is as follows:

[0084]

[0085] In the district: For the first The row vectors of each user sample in the standardized geographic data matrix.

[0086] S13. Analyze the temporal and geographical variation characteristics of the carbon emission factors of the target users based on the time-related target characteristic data and the geographical distribution-related target characteristic data, respectively.

[0087] In this embodiment, step S13 further includes the following steps:

[0088] S131. Using the carbon emission factor of the target user as the dependent variable, construct dummy variables using time-related target characteristic data and geographically distributed target characteristic data respectively;

[0089] Dummy variables are constructed using time-related target feature data:

[0090] First, analyze the time-related target feature data to determine an appropriate time period division method. For example, it can be divided by time periods of the day, such as 0-6 o'clock (time period 1), 6-12 o'clock (time period 2), 12-18 o'clock (time period 3), and 18-24 o'clock (time period 4); it can also be divided by weekdays, such as weekdays and rest days; or it can be divided by seasons, such as spring, summer, autumn, and winter.

[0091] Dummy variables are constructed based on time periods. Taking time periods of a day as an example, four dummy variables are created. , , , When the data corresponds to a time between 0:00 and 6:00, ,the remaining When the time is between 6 and 12 o'clock, The rest are 0, and so on.

[0092] If divided by week, let For workday dummy variables, workday time , holiday , set as holiday dummy variable, when it is holiday , when it is weekday . For season division, set as spring dummy variable, when it is spring , when it is other season ; similarly set , , set summer, autumn, winter dummy variables respectively.

[0093] Construct dummy variables with target feature data related to geographical distribution:

[0094] Analyze target feature data related to geographical distribution to determine the division method of geographical type. Geographical type can be divided according to administrative region, such as different cities, provinces; it can also be divided according to functional region, such as business district, industrial district, residential district; it can also be divided according to geographical features, such as coastal area, inland area, mountainous area, etc.

[0095] Taking city division as an example, assume there are 3 cities A, B, C. Construct 3 dummy variables , , when the target user is located in city A, , when the target user is located in city B, , when the target user is located in city C, , . If divided according to functional region, set as business district dummy variable, when located in business district , otherwise , set as industrial district dummy variable, when located in industrial district , otherwise ; set as residential district dummy variable, when located in residential district , otherwise .

[0096] In the above way, dummy variables are constructed with target feature data related to time and target feature data related to geographical distribution respectively, which can be used to further analyze the relationship between target user's carbon emission factor and time and geographical factors.

[0097] S132. Determine an initial time virtual variable model based on the dependent variable and the virtual variable constructed based on the target feature data related to time, and determine a spatial virtual variable model based on the dependent variable and the virtual variable constructed based on the target feature data related to geographical distribution;

[0098] In this embodiment, the initial time virtual variable model is established by using the virtual variable constructed based on the target feature data related to time, with the carbon emission factor of the target user as the core dependent variable, so as to mine the change rule of the carbon emission factor in the time dimension. Meanwhile, the spatial virtual variable model is determined based on the virtual variable constructed based on the target feature data related to geographical distribution, so as to explore the change mode of the carbon emission factor under different geographical conditions, and provide a data basis for comprehensively analyzing the carbon emission of the target user.

[0099] S133. Calculate the coefficients of the time virtual variable model and the coefficients of the spatial virtual variable model respectively by using the least square method;

[0100] The coefficients of the time virtual variable model :

[0101]

[0102]

[0103]

[0104]

[0105] Wherein: is the number of time virtual variables in the i th observation value; The coefficients of the spatial virtual variable model :

[0106]

[0107]

[0108]

[0109]

[0110]

[0111] Wherein: is the number of spatial virtual variables in the i th observation value;

[0112] S134. Determine the target time virtual variable model and the target spatial virtual variable model respectively according to the calculated coefficients. ​​​​

[0113] Target time dummy variable model:

[0114]

[0115] wherein: is a time factor intercept term, is a corresponding coefficient of the time dummy variable is a time factor error term. In the time dummy variable model, the change of the coefficient

[0116] may reflect the dynamic change of the carbon emission factor in the time dimension. For example, if increases over time, it means that the carbon emission factor of the user in this time interval is on the rise. Target space dummy variable model:

[0117]

[0118]

[0119] wherein: is a space factor intercept term, is a corresponding coefficient of the space dummy variable is a space factor error term. In the space dummy variable model,

[0120] the change of the coefficient reflects the dynamic change of the carbon emission factor in the spatial dimension. For example, an increase may mean that the carbon emission factor of the user in this region is increasing.

[0121] S14. Construct a dynamic portrait of the carbon emission of the target user according to the time variation characteristics and regional distribution variation characteristics of the carbon emission factor of the target user.

[0122] In this embodiment, step S14 further includes:

[0123] S141. Determine the change of the time dummy variable coefficient over time according to the target time dummy variable model, and predict the change of the dummy variable coefficient in the preset time by combining the time series analysis method;

[0124] Observe the change of the different time dummy variable coefficients over time (such as different seasons, different years). For example, if the time dummy variable coefficient representing the daytime period increases year by year in summer, it means that the carbon emission factor of the user in summer daytime is on the rise. The change of the future time dummy variable coefficient is predicted by combining the time series analysis method. For example, for ​​A simple first-order autoregressive model is established:

[0125]

[0126] where is the autoregressive coefficient, and is the error term. By estimating the future coefficient changes, the future carbon emission behavior of the user can be predicted.

[0127] S142. Determine the changes of spatial dummy variable coefficients with regional development according to the target spatial dummy variable model, and analyze the influence of region-related external factors on the coefficients;

[0128] Observe the changes of different spatial dummy variable coefficients with regional development. For example, if a city is vigorously developing renewable energy, the spatial dummy variable coefficient representing the city may gradually decrease. Consider the influence of region-related external factors on the coefficients. For example, after a new carbon emission reduction policy is implemented in a certain region, evaluate the policy effect by comparing the changes of spatial dummy variable coefficients before and after the policy, and update the user portrait.

[0129] S143. Construct the first dynamic portrait of user electricity carbon emission based on the changes and prediction results of time coefficients, and construct the second dynamic portrait of user electricity carbon emission based on the changes of spatial coefficients and external factors;

[0130] Construct the first dynamic portrait according to the above coefficient changes and prediction results, for example, if it is predicted that the carbon emission factor coefficient of a certain user will rise sharply in the future, and the user is a residential user, the dynamic portrait can be “future high carbon emission growth type residential user”.

[0131] Construct the second dynamic portrait according to the above spatial coefficient changes and external factors, for example, if a region has high energy-consuming industries due to new industrial policies, the spatial dummy variable coefficient will rise, and for users in this region, the dynamic portrait can be “industrial policy affected high carbon emission growth type user”.

[0132] S144. Determine the target dynamic portrait of user electricity carbon emission according to the first dynamic portrait and the second dynamic portrait.

[0133] Combine the analysis results of time and space models, i.e. determine the target dynamic portrait of user electricity carbon emission according to the first dynamic portrait and the second dynamic portrait, and the target dynamic portrait is a more detailed user portrait. For example, through the spatial model, it is determined that the user is in a certain city, through the time model, it is determined that the user's carbon emission factor is high in winter night, and the user type is residential, and the portrait can be “certain city winter night high carbon emission residential user”.

[0134] Specifically, a model containing time and space interaction terms can be constructed by integrating time and space models, and the interaction coefficients can be analyzed to describe the target dynamic portrait. The expression is as follows:

[0135]

[0136] wherein: is the coefficient of the interaction term . If a certain time-space interaction term coefficient is large, it indicates that the user carbon emission factor has a special change pattern under a specific combination of time and space, and the interaction effect can be reflected in the portrait.

[0137] For the above integrated model with interaction terms, the changes in time and space can be considered to predict the carbon emission factor. For example, the carbon emission factor of a user who has moved to a certain region at a future time point can be predicted, which fully embodies the combination of time and space dynamic portrait. Through the fitting and coefficient estimation of these models, the combination of time and space dynamic portrait can be quantitatively represented. By continuously updating the coefficients of the above models, , , , the dynamic changes of the user's carbon emission factor in the time and space dimensions can be timely reflected, and the combination of the time and space dynamic portrait can be accurately presented. For example, due to the adjustment of the energy structure of a certain region, and the change of the user's electricity consumption behavior at different times, the model can be refitted to capture the impact of these changes on the interaction term coefficient, and accurately represent the combination of the dynamic portrait.

[0138] The above embodiment can present the user's electricity consumption carbon emission characteristics in different time and space situations by constructing the user portrait based on the carbon emission factor varying with time and space. The portrait presents the user's carbon emission in a specific time and a specific region. The power company can provide personalized energy-saving suggestions and customized electricity price packages for users based on the portrait, encourage users to adjust their electricity consumption behavior, and achieve energy saving and emission reduction; government departments can develop more targeted and scientific carbon emission reduction policies based on the portrait, and reasonably plan energy structure adjustment and regional development strategies, while also providing intuitive basis for evaluating the implementation effect of the policies; from a broader perspective, it helps to improve the overall social awareness of carbon emissions and promote the whole society to actively change towards low-carbon development.

[0139] Embodiment two

[0140] An embodiment of a user electricity consumption carbon emission portrait construction system based on spatiotemporal distribution characteristics in the present application comprises the following steps:

[0141] a data acquisition and preprocessing unit configured to acquire information data of a target user, historical power consumption data of the target user, and power grid generation structure data, and to preprocess the acquired data;

[0142] a target feature data extraction unit configured to extract time-related target feature data and region-distribution-related target feature data based on the preprocessed data;

[0143] a time variation feature and region-distribution variation feature analysis unit configured to analyze time variation features and region-distribution variation features of carbon emission factors of the target user according to the time-related target feature data and the region-distribution-related target feature data, respectively;

[0144] a dynamic portrait construction unit configured to construct a dynamic portrait of power consumption carbon emission of the target user according to the time variation features and the region-distribution variation features of carbon emission factors of the target user.

[0145] The specific limitations on the system can be referred to the limitations on the method in the above, which will not be repeated here. Each module in the above system can be realized by software, hardware, and combinations thereof, in whole or in part. Each module described above can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0146] Those skilled in the art can appreciate that the units of each example described in combination with the embodiments disclosed herein can be realized in electronic hardware, computer software, or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, each example has been described in the above description in general terms. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0147] In the embodiments provided by the present application, it should be understood that the division of units is only a logical functional division, and actual implementation can have another division manner, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above integrated unit can be realized in the form of hardware, or in the form of software functional unit.

[0148] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store program codes.

[0149] It can be understood that the above embodiments are only used to illustrate the technical solutions of the present application, but not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application, and they should be covered in the scope of the claims and the specification of the present application.

Claims

1. A method for constructing a user electricity consumption carbon emission profile based on spatiotemporal distribution characteristics, characterized in that, include: Acquire target user information data, target user historical electricity consumption data, and power grid power generation structure data, and preprocess the acquired data; Based on the preprocessed data, extract time-related target feature data and geographically distributed target feature data; Analyzing the time-related and geographical distribution characteristics of the carbon emission factors of the target user based on the time-related target feature data and the geographical distribution target feature data, respectively; the analysis of the time-related and geographical distribution characteristics of the carbon emission factors of the target user based on the time-related and geographical distribution target feature data includes: Using the carbon emission factor of the target user as the dependent variable, virtual variables are constructed using time-related target characteristic data and geographically distributed target characteristic data, respectively. The initial time dummy variable model is determined based on the dependent variable and the time-related target feature data, and the spatial dummy variable model is determined based on the dependent variable and the target feature data related to geographical distribution. The coefficients of the time dummy variable model and the coefficients of the space dummy variable model are calculated using the least squares method, respectively. The target time dummy variable model and the target space dummy variable model are determined based on the calculated coefficients, respectively; the determination of the target time dummy variable model and the target space dummy variable model based on the calculated coefficients includes: Target time dummy variable model: in: For the time factor intercept term, For time dummy variables The corresponding coefficients, This is the time factor error term; Target space dummy variable model: in: For spatial factors, For spatial dummy variables The corresponding coefficients, This is the spatial factor error term; A dynamic profile of the target user's electricity consumption carbon emissions is constructed based on the temporal and geographical variation characteristics of the target user's carbon emission factors; the comprehensive model expression for the dynamic profile of the target user's electricity consumption carbon emissions is as follows: in: For interactive items The coefficient.

2. The method for constructing a user electricity consumption carbon emission profile based on spatiotemporal distribution characteristics according to claim 1, characterized in that, The extraction of time-related target feature data and geographic distribution-related target feature data based on preprocessed data includes: Principal component analysis was used to determine the time-related eigenvectors and the regional distribution-related eigenvectors, respectively. Based on the calculated scores of the time principal component and the region principal component, target feature data related to time and target feature data related to regional distribution are determined.

3. The method for constructing a user electricity consumption carbon emission profile based on spatiotemporal distribution characteristics according to claim 2, characterized in that, The process of determining time-related target feature data and geographically related target feature data based on the calculated scores of the time principal component and the geographical principal component includes: Time principal component score : in: For the first The row vectors of each user sample in the standardized time series data matrix The corresponding feature vector; Regional principal component scores : In the district: For the first The row vectors of each user sample in the standardized geographic data matrix. This is the corresponding feature vector.

4. The method for constructing a user electricity consumption carbon emission profile based on spatiotemporal distribution characteristics according to claim 1, characterized in that, The step of calculating the coefficients of the time dummy variable model and the spatial dummy variable model using the least squares method includes: Coefficients of the time dummy variable model : in: for Among the observations A time-based dummy variable; Coefficients of the spatial dummy variable model : in: for Among the observations Spatial dummy variables.

5. The method for constructing a user electricity consumption carbon emission profile based on spatiotemporal distribution characteristics according to any one of claims 1-4, characterized in that, The process of constructing a dynamic profile of the target user's electricity carbon emissions based on the time-varying and geographical-distribution characteristics of the target user's carbon emission factors includes: The changes in the coefficients of the time dummy variables over time are determined based on the target time dummy variable model, and the changes in the coefficients of the dummy variables over a preset time period are predicted by combining time series analysis methods. Based on the target spatial dummy variable model, determine the changes in spatial dummy variable coefficients with regional development, and analyze the influence of regionally related external factors on the coefficients; A first dynamic profile of user electricity carbon emissions is constructed based on changes in time coefficients and prediction results, and a second dynamic profile of user electricity carbon emissions is constructed based on changes in spatial coefficients and external factors. The target dynamic profile of the user's electricity carbon emissions is determined based on the first dynamic profile and the second dynamic profile.

6. A system for constructing a user electricity consumption carbon emission profile based on spatiotemporal distribution characteristics, characterized in that, Using the method according to any one of claims 1-5, comprising: The data acquisition and preprocessing unit is used to acquire information data of target users, historical electricity consumption data of target users, and power grid power generation structure data, and to preprocess the acquired data. The target feature data extraction unit is used to extract time-related target feature data and geographically related target feature data based on the preprocessed data. The time-varying characteristics and geographical distribution variation characteristics analysis unit is used to analyze the time-varying characteristics and geographical distribution variation characteristics of the carbon emission factors of the target user based on the time-related target characteristic data and the geographical distribution-related target characteristic data, respectively. The dynamic profile building unit is used to build a dynamic profile of the target user's electricity carbon emissions based on the time variation characteristics and geographical distribution variation characteristics of the target user's carbon emission factors.

Images