Electric quantity data prediction method and device, terminal equipment and storage medium

Abstract

The invention is suitable for the technical field of electric power and electric quantity data prediction, and provides an electric quantity data prediction method and device, terminal equipment and a storage medium. Meteorological sensitivity factors of all enterprises are obtained in combination with a linear regression model, high-precision and high-efficiency unified medium-and-long-term load prediction is achieved through parameter fitting and anti-normalization processing, and the requirement for power market deviation rate control is met.

Description

Technical Field

[0001] This application belongs to the field of power data prediction technology, and in particular relates to a power data prediction method, device, terminal equipment and storage medium. Background Technology

[0002] With the advancement of a new round of power system reform, a large number of electricity users, covering multiple industries, with large differences in electricity consumption, wide geographical distribution, and whose electricity consumption characteristics are affected by different factors such as holidays, weather, and production plans, are participating in market transactions. Meanwhile, electricity sales companies face the challenge of limited historical data due to frequent changes in the number of agents and the heavy workload of analyzing each agent individually due to the large number of agents.

[0003] Current technologies struggle to construct high-precision models, making it impossible to accurately predict electricity demand from different regions and user types, thus failing to meet the needs of electricity sales companies. Therefore, a method for medium- to long-term load forecasting for electricity users is urgently needed. Summary of the Invention

[0004] This application provides a method, apparatus, terminal device, and storage medium for predicting electricity data, which can improve the accuracy and applicability of medium- and long-term load forecasting and better meet the electricity consumption forecasting needs of electricity sales companies for complex user groups.

[0005] In a first aspect, embodiments of this application provide a method for predicting electricity data, including: Acquire first data; wherein the first data includes first meteorological data and second meteorological data corresponding to the target area, wherein the first meteorological data is meteorological data corresponding to at least one first target month corresponding to the target area; and the second meteorological data is meteorological data corresponding to the same month corresponding to the first target month; Obtain second data; wherein the second data includes first power consumption data and second power consumption data corresponding to the target area; wherein the first power consumption data is power consumption data corresponding to at least one first target month corresponding to the target area; and the second power consumption data is power consumption data corresponding to the same month corresponding to the first target month. The meteorological sensitivity factor is calculated based on the first and second data; the meteorological sensitivity factor is used to represent the variation pattern of electricity data with meteorological data. Obtain the third electricity data; wherein, the third electricity data is the electricity data of the same month corresponding to the second target month; The target electricity data for the second target month is predicted based on meteorological sensitive factors and third electricity data; wherein, the first target month is the preceding month of the second target month.

[0006] In this embodiment, meteorological data (the first target month and its corresponding months) and electricity consumption data for different time periods in the target area are acquired. A meteorological sensitive factor that reflects the pattern of electricity consumption changes with meteorological conditions is calculated, and then the electricity consumption for the subsequent month (the second target month) is predicted based on this factor. By introducing comparative analysis of meteorological and electricity consumption data from the same period, the influence of meteorological factors on electricity consumption (meteorological sensitive factor) is accurately captured. This solves the error problem caused by ignoring meteorological variables or using only data from a single period in traditional forecasting, improves the accuracy and applicability of medium- and long-term load forecasting, and better meets the electricity consumption forecasting needs of electricity sales companies for complex user groups.

[0007] In one possible implementation of the first aspect, calculating the meteorological sensitivity factor based on the first data and the second data includes: The difference between the first and second meteorological data in the first data is calculated to obtain the third meteorological data; The first electricity consumption coefficient is calculated based on the first electricity consumption data in the second data; wherein, the first electricity consumption coefficient represents the ratio coefficient between the first type of date and the working day type date corresponding to the first target month; wherein, the first type of date includes at least one of working days, public holidays and statutory holidays; The second electricity consumption coefficient is calculated based on the second electricity consumption data in the second data; wherein, the second electricity consumption coefficient represents the ratio coefficient between the second type date and the working day type date corresponding to the same month of the first target month; wherein, the second type date includes at least one of working days, public holidays and statutory holidays; Meteorological sensitivity factors are calculated based on the third meteorological data, the first power coefficient, and the second power coefficient.

[0008] In this embodiment, the meteorological sensitive factor is calculated step by step. First, the difference in meteorological data is quantified to quantify meteorological changes. Then, two types of electricity coefficients are used to eliminate the interference of electricity consumption differences on different date types. Finally, the accurate meteorological sensitive factor is obtained by combining multi-dimensional data. Its beneficial effects are: through systematic data processing and correlation analysis, the influence of date type and meteorological factors on electricity consumption is effectively separated, so that the calculated meteorological sensitive factor can more realistically reflect the intrinsic relationship between meteorological changes and electricity fluctuations, providing reliable parameter support for subsequent electricity consumption forecasting and improving the scientificity and accuracy of the forecasting model.

[0009] In one possible implementation of the first aspect, the step of calculating the target electricity coefficient corresponding to the target type date in the first target month based on the target electricity data includes: Calculate the average daily electricity consumption data for the first day corresponding to each target type date based on the target electricity consumption data; Obtain the second daily average electricity consumption data from multiple first daily average electricity consumption data; wherein, the second daily average electricity consumption data is the daily average electricity consumption data corresponding to the target type date being a weekday; Calculate the ratio between the average daily power consumption data of the first day and the average daily power consumption data of the second day to obtain the target power consumption coefficient; wherein, when the target power consumption data is the first power consumption data, the target type date is the first type date, and the target power consumption coefficient is the first power consumption coefficient; when the target power consumption data is the second power consumption data, the target type date is the second type date, and the target power consumption coefficient is the second power consumption coefficient.

[0010] In this embodiment, the target electricity consumption coefficient is obtained through the steps of "calculating the average daily electricity consumption of each target type date → extracting the average daily electricity consumption of working days → calculating the ratio". The beneficial effects of distinguishing the scenarios corresponding to the first and second electricity consumption data are as follows: it can accurately quantify the electricity consumption ratio of different target type dates (such as public holidays and public holidays) relative to working days, effectively remove the interference of date type differences on electricity consumption data, provide a reliable basis for subsequent electricity consumption normalization correction and elimination of the influence of differences in cross-period dates, and improve the pertinence and accuracy of data processing.

[0011] In one possible implementation of the first aspect, the meteorological sensitivity factor is calculated based on third meteorological data, a first electrical energy coefficient, and a second electrical energy coefficient, including: The first power data is normalized according to the first power coefficient to obtain the first corrected power of the first power data; The second power data is normalized according to the second power coefficient to obtain the second corrected power data. Meteorological sensitivity factors are calculated based on the first corrected electricity amount, the second corrected electricity amount, and the third meteorological data.

[0012] In this embodiment of the application, by normalizing the first and second power consumption data with the first and second power consumption coefficients respectively to eliminate date type interference, and then combining the third meteorological data to calculate the meteorological sensitivity factor, the influence of date differences on power consumption data can be effectively removed, so that the calculated meteorological sensitivity factor can more accurately reflect the correlation between meteorological changes and power consumption fluctuations, and provide reliable parameter support for subsequent power consumption forecasting.

[0013] In one possible implementation of the first aspect, the meteorological sensitivity factor is calculated based on the first corrected electrical quantity, the second corrected electrical quantity, and the third meteorological data, including: The difference between the first and second corrected power quantities is calculated to obtain the third corrected power quantity. A linear regression model incorporating meteorological sensitive factors was established based on the third revised electricity volume and the third meteorological data. The meteorological sensitivity factors are obtained by solving the linear regression model.

[0014] In this embodiment, the meteorological sensitivity factor is calculated through a step-by-step logic of "normalization to eliminate date interference → calculation of power difference to focus on changes → building a linear model to solve". The beneficial effect is that: first, the date type interference in the power data is removed by normalization correction, and then the model is built with power difference and meteorological difference data. This can accurately quantify the correlation between meteorological changes and power fluctuations, making the solved meteorological sensitivity factor more reliable and providing scientific and accurate core parameter support for subsequent power forecasting.

[0015] In one possible implementation of the first aspect, the target electricity data for the second target month is predicted based on meteorological sensitive factors and third electricity data, including: The third electricity consumption coefficient is calculated based on the third electricity consumption data; wherein, the third electricity consumption coefficient represents the ratio coefficient between the third type of date and the working day type date in the same month corresponding to the second target month; wherein, the third type of date includes at least one of working days, public holidays and statutory holidays; The third energy data is normalized and corrected based on the third energy coefficient to obtain the third corrected energy. Obtain third data; wherein, the third data includes fourth meteorological data and fifth meteorological data; wherein, the fourth meteorological data is the meteorological data corresponding to the second target month; and the fifth meteorological data is the meteorological data corresponding to the same month as the second target month; Target power data is predicted based on the third corrected power volume, the third data, and meteorological sensitive factors.

[0016] In this embodiment of the application, the electricity volume of the second target month is predicted by the steps of "calculating the third electricity coefficient → normalizing to obtain the third corrected electricity volume → taking the target and the meteorological data of the same period → combining the sensitive factors for prediction". The beneficial effect is that: firstly, the date type interference of the electricity volume of the same period in history is eliminated, and then the meteorological data of the target month and the same period and the obtained meteorological sensitive factors are combined, so that the prediction process relies on reliable historical benchmarks and conforms to the actual meteorological conditions of the target month, effectively improving the accuracy and rationality of the electricity volume prediction of the second target month.

[0017] In one possible implementation of the first aspect, the target electricity data is predicted based on the third corrected electricity level, the third data, and meteorological sensitive factors, including: The sixth meteorological data is obtained by calculating the difference between the fourth and fifth meteorological data in the third data set. The first value is obtained by weighting and summing the sixth meteorological data with the meteorological sensitive factors; The first value and the third corrected energy value are summed to obtain the fourth corrected energy value; The fourth corrected energy level is inversely normalized to obtain the target energy level data.

[0018] In this embodiment of the application, the target electricity consumption is predicted by the steps of "calculating meteorological differences → calculating the incremental impact of meteorological conditions on electricity consumption → obtaining the baseline predicted electricity consumption → restoring the actual date scenario". The beneficial effect is that: firstly, the meteorological differences between the target and the same period and their impact on electricity consumption are quantified, and then the historical baseline electricity consumption is combined with the restoration of the actual date type composition, so that the prediction is not only accurately related to the meteorological change pattern, but also fits the real electricity consumption scenario of the target month, and finally outputs accurate target electricity consumption data that meets the actual needs.

[0019] Secondly, embodiments of this application provide a power data prediction device, comprising: A first data acquisition module is used to acquire first data; wherein the first data includes first meteorological data and second meteorological data corresponding to a target area, wherein the first meteorological data is meteorological data corresponding to at least one first target month corresponding to the target area; and the second meteorological data is meteorological data corresponding to the same month corresponding to the first target month. The second data acquisition module is used to acquire second data; wherein the second data includes first power data and second power data corresponding to the target area; wherein the first power data is power data corresponding to at least one first target month corresponding to the target area; and the second power data is power data corresponding to the same month as the first target month. A meteorological sensitivity factor calculation module is used to calculate a meteorological sensitivity factor based on the first data and the second data; wherein, the meteorological sensitivity factor is used to represent the variation pattern of electricity data with meteorological data. The third power data acquisition module is used to acquire third power data; wherein, the third power data is the power data of the same month corresponding to the second target month; The power data prediction module is used to predict the target power data corresponding to the second target month based on the meteorological sensitive factor and the third power data; wherein, the first target month is the preceding month of the second target month.

[0020] Thirdly, embodiments of this application provide a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the power data prediction method as described in any of the first aspects above.

[0021] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the power data prediction method as described in any of the first aspects above.

[0022] Fifthly, embodiments of this application provide a computer program product that, when run on a terminal device, causes the terminal device to execute the power data prediction method of any one of the first aspects described above.

[0023] It is understood that the beneficial effects of the second to the third aspects mentioned above can be found in the relevant descriptions in the first aspect mentioned above, and will not be repeated here. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic flowchart of a power data prediction method provided in an embodiment of this application; Figure 2 This is a schematic diagram of the process for obtaining meteorological sensitive factors provided in an embodiment of this application. Figure 1 ; Figure 3 This is a schematic flowchart illustrating the calculation of the power coefficient according to an embodiment of this application; Figure 4 This is a schematic diagram of the process for obtaining meteorological sensitive factors provided in the embodiments of this application. Figure 2 ; Figure 5 This is a schematic diagram of the process for obtaining meteorological sensitive factors provided in the embodiments of this application. Figure 3 ; Figure 6 This is a flowchart illustrating the prediction of target power data provided in an embodiment of this application. Figure 1 ; Figure 7 This is a flowchart illustrating the prediction of target power data provided in an embodiment of this application. Figure 2 ; Figure 8 This is a power data prediction device provided in one embodiment of the present application; Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0026] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0027] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0028] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0029] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0030] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0031] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0032] With the advancement of the new round of power system reform, a large number of electricity users, spanning multiple industries, varying in electricity consumption volume, widely distributed geographically, and whose electricity consumption characteristics are affected by different factors such as holidays, weather, and production plans, are participating in market transactions. However, electricity sales companies face challenges such as limited historical data due to frequent changes in their represented users, and the heavy workload of analyzing each user individually due to the large number of represented users. Existing technologies struggle to construct high-precision models, making it impossible to accurately predict the electricity demand of different regions and types of users, thus failing to meet the needs of electricity sales companies. Therefore, a method for medium- and long-term load forecasting of electricity users is urgently needed.

[0033] To address the aforementioned technical issues, this application provides a method for predicting electricity consumption data. Based on the monthly historical electricity consumption of regional and enterprise users, this application calculates the year-on-year difference in meteorological data and the electricity consumption coefficient for each date type, and combines a linear regression model to derive the meteorological sensitivity factor for each enterprise. After parameter fitting and inverse normalization, a unified medium- and long-term load forecast with high accuracy and efficiency is achieved, meeting the requirements for controlling the deviation rate of the electricity market.

[0034] See Figure 1 This is a flowchart illustrating a power data prediction method according to an embodiment of this application. It is intended as an example and not a limitation. The method may include the following steps: S101, acquire first data; wherein the first data includes first meteorological data and second meteorological data corresponding to the target area, wherein the first meteorological data is meteorological data corresponding to at least one first target month corresponding to the target area; and the second meteorological data is meteorological data corresponding to the same month corresponding to the first target month.

[0035] In this embodiment of the application, the first meteorological data refers to the meteorological data of at least one first target month corresponding to the target area (such as the location area corresponding to a certain enterprise). The first target month is the "preceding historical month" used for modeling in the scheme (e.g., when predicting October 2024, the first target month can be set to July-August 2024). The data includes the highest temperature, lowest temperature, relative humidity, pressure and other elements of the region in these months, which are the first meteorological data and serve as the "current benchmark data" for subsequent calculation of the year-on-year difference in meteorological data.

[0036] The second meteorological data refers to the meteorological data of the "same-year month corresponding to the first target month." The "same-year month" is the same month of the previous year that is "year-aligned" with the first target month (e.g., if the first target month is July-August 2024, the same-year month is July-August 2023). The data type is consistent with the first meteorological data and is used for comparison to calculate the differences in changes in meteorological elements between years (e.g., the year-on-year differences in temperature and humidity in Formula 3-6). Simply put, the first data is a combination of meteorological data from the previous historical months of the target area plus the same-year month of the previous year. Its core function is to provide basic data for subsequent quantification of the "impact of meteorological changes on electricity consumption" and is a key input for calculating the meteorological sensitivity factor.

[0037] Specifically, the system first indexes meteorological data (first or second meteorological data) for each enterprise's region from the input meteorological data source. This data includes various meteorological elements such as maximum temperature (Tmax), minimum temperature (Tmin), relative humidity (H), and pressure (P). Meteorological data is one of the important influencing factors considered in subsequent analysis because meteorological conditions have a significant impact on enterprises' electricity consumption behavior; for example, enterprises' electricity consumption for refrigeration equipment will increase during hot weather.

[0038] S102, Obtain second data; wherein, the second data includes first power data and second power data corresponding to the target area; wherein, the first power data is power data corresponding to at least one first target month corresponding to the target area; and the second power data is power data corresponding to the same month corresponding to the first target month.

[0039] In this embodiment, the first electricity consumption data refers to the electricity consumption data of the target area corresponding to "at least one first target month," where the "first target month" is the "preceding historical month of the current year" used for modeling in the scheme (e.g., when predicting October 2024, the first target month can be set to July-August 2024). The data type includes the monthly historical electricity consumption of enterprises in the area (used for subsequent electricity consumption correction calculations) and time-of-use electricity consumption data (recorded hourly), which is the core data characterizing the current electricity consumption characteristics of enterprises.

[0040] The second electricity data refers to the electricity data of the "same-year month corresponding to the first target month". The "same-year month" is the same month of the previous year that is "year-aligned" with the first target month (e.g., if the first target month is July-August 2024, the same-year month would be July-August 2023). The data type is the same as the first electricity data, and it also includes monthly historical electricity and hourly electricity data.

[0041] The historical months preceding the current year can be represented by the following formula (time-based): (1) Historical electricity consumption data for the corresponding month of the previous year (the same month) can be expressed using the following formula (year-aligned): (2) Where n is the set historical window length, usually 3 to 6, determined according to actual business needs and data integrity.

[0042] S103, calculate the meteorological sensitivity factor based on the first data and the second data; wherein, the meteorological sensitivity factor is used to represent the variation pattern of electricity data with meteorological data.

[0043] In this application embodiment, a key factor that reflects "how the weather changes, the electricity consumption changes accordingly" is calculated using "first data" (meteorological data of specific historical months in the target area and the same month of the previous year) and "second data" (electricity data of the target area corresponding to these months). This factor is the meteorological sensitivity factor.

[0044] In one embodiment, see Figure 2 This is a schematic diagram of the process for obtaining meteorological sensitive factors provided in an embodiment of this application. Figure 1 ,like Figure 2 As shown, step S103 includes: S201, perform difference calculation on the first meteorological data and the second meteorological data in the first data to obtain the third meteorological data.

[0045] In this application embodiment, the third meteorological data obtained by calculating the difference between the first meteorological data and the second meteorological data in the first data is essentially the annual year-on-year difference in meteorological elements corresponding to the "first target month" and the "same period month" in the target area. Its core function is to quantify the changes in meteorological conditions within the same period of two years, and to provide a quantitative basis for meteorological dimensions for subsequent related changes in electricity consumption.

[0046] Specifically, the first meteorological data consists of meteorological data for at least one "first target month" (e.g., July-August 2024) in the target area, including elements such as maximum temperature, minimum temperature, relative humidity, and pressure. The second meteorological data consists of the same type of meteorological data for the corresponding "months of the same period" (e.g., July-August 2023) of these "first target months". The third meteorological data is obtained by subtracting the corresponding elements from the two sets of data one by one (e.g., subtracting the maximum temperature in July 2023 from the maximum temperature in July 2024).

[0047] Specifically, the calculation of year-on-year differences in meteorological information (i.e., third-party meteorological data) involves processing meteorological information for each enterprise's location on a monthly basis. Specifically, the meteorological data is aggregated and processed at a monthly granularity. Let a certain meteorological element in region r for year Y and month M be... The value corresponding to the same month of the previous year The year-on-year difference of this factor is then defined as: (3) Taking the highest temperature as an example, the year-on-year difference is expressed as follows: (4) Similarly, the difference in minimum temperature is expressed as: (5) By calculating the above differences, the changing trends of regional climate elements can be systematically depicted from year to year, providing input features for subsequent meteorological-driven modeling of electricity load changes.

[0048] Furthermore, this method supports parallel processing of multiple meteorological factors to construct a multi-dimensional difference function: (6) in , These values ​​represent the year-on-year changes in humidity, air pressure, and other factors. These differences visually reflect the changes in meteorological conditions from year to year, providing quantitative indicators for subsequent analysis of the impact of meteorology on electricity consumption.

[0049] S202, calculate the first power consumption coefficient based on the first power consumption data in the second data; wherein, the first power consumption coefficient represents the ratio coefficient between the first type of date and the working day type date corresponding to the first target month; wherein, the first type of date includes at least one of working days, public holidays and statutory holidays.

[0050] In this embodiment of the application, the first electricity consumption coefficient is calculated based on the first electricity consumption data (electricity consumption data of at least one first target month in the target area) in the second data. Essentially, it is to quantify the ratio of electricity consumption of "first type of date" (such as public holidays and statutory holidays) to "working day" in the first target month. The core is to extract the electricity consumption patterns of different date types through statistical analysis.

[0051] S203, calculate the second electricity coefficient based on the second electricity data in the second data; wherein, the second electricity coefficient represents the ratio coefficient between the second type date and the working day type date corresponding to the same month of the first target month; wherein, the second type date includes at least one of working days, public holidays and statutory holidays.

[0052] In this embodiment of the application, the second electricity consumption coefficient is calculated based on the second electricity consumption data (electricity consumption data of the first target month corresponding to the same month) in the second data. Essentially, it is to quantify the electricity consumption ratio of the second type of date (such as public holidays and statutory holidays) and working days within the "same month". The core is to extract the electricity consumption pattern of date type in the same month, so as to provide parameters for subsequent cross-year electricity consumption comparison and elimination of date type interference.

[0053] In one embodiment, see Figure 3 This is a schematic flowchart illustrating the calculation of the power coefficient according to an embodiment of this application, as shown below. Figure 3 As shown, it includes: S301, calculate the first daily average electricity data corresponding to each target type date based on the target electricity data.

[0054] In the embodiments of this application, "calculating the first daily average electricity consumption data corresponding to each target type date based on the target electricity consumption data" essentially involves extracting the electricity consumption data of the target type date (such as weekdays, public holidays, statutory holidays, etc.) from the target electricity consumption data, and then obtaining the daily average electricity consumption of each date type through statistical averaging.

[0055] To achieve structured modeling and feature summarization of enterprise electricity consumption behavior, this invention first reads enterprise... Historical time-of-use electricity data (target electricity data), through daily Sum the electricity consumption across all time periods to construct a daily total electricity consumption index. Then, based on the time-based tag library, the system assigns each date... Mapped to a specific date type t( )∈{weekdays, weekends, statutory holidays, adjusted workdays…}, and associate them with the corresponding electricity consumption data: ( (7) For each enterprise ei, calculate its average daily electricity consumption (first day's average electricity consumption data) under different date types: (8) in = { |T( Let ) = t} represent the set of all dates belonging to date type t. S302, Obtain the second daily average power consumption data from multiple first daily average power consumption data; wherein, the second daily average power consumption data is the daily average power consumption data corresponding to the target type date being a working day.

[0056] In this embodiment of the application, the operation of "obtaining the second daily average electricity consumption data from multiple first daily average electricity consumption data" essentially involves filtering out the data set where "the target type date is a weekday" from the statistically compiled "first daily average electricity consumption data corresponding to each target type date" as the second daily average electricity consumption data. The core is to clearly define the electricity consumption benchmark value for the specific target type date of "weekday". The specific logic is as follows: First, the "multiple first-day average electricity consumption data" are the results previously calculated using target electricity consumption data—that is, the average daily electricity consumption for each group of dates is calculated separately based on "target date type" (such as weekdays, public holidays, statutory holidays, etc.) (corresponding to "average daily electricity consumption under different date types" in Formula 8). These grouped daily average electricity consumption data together constitute "multiple first daily average electricity consumption data".

[0057] Next, based on the definition that "the second daily average power consumption data is the daily average power consumption data corresponding to the target type date being a working day", the daily average power consumption data of the group with "target type date = working day" is located and extracted from the above "multiple first daily average power consumption data". This data is the second daily average power consumption data.

[0058] S303, calculate the ratio between each first daily average power consumption data and the second daily average power consumption data to obtain the target power consumption coefficient; wherein, when the target power consumption data is the first power consumption data, the target type date is the first type date, and the target power consumption coefficient is the first power consumption coefficient; when the target power consumption data is the second power consumption data, the target type date is the second type date, and the target power consumption coefficient is the second power consumption coefficient.

[0059] In the embodiments of this application, "calculating the ratio of each first daily average electricity consumption data to the second daily average electricity consumption data to obtain the target electricity consumption coefficient" essentially quantifies the electricity consumption difference ratio of different date types relative to workdays by "daily average electricity consumption of a specific date type ÷ daily average electricity consumption of workdays", and generates a first electricity consumption coefficient or a second electricity consumption coefficient according to "whether the target electricity consumption data is the first electricity consumption data or the second electricity consumption data".

[0060] Specifically, using "working day" as a reference benchmark, the proportionality coefficient of electricity consumption for other date types (t) relative to working days is calculated: (9) This ratio can be used to measure the intensity of electricity consumption differences between special date types such as holidays and weekends and normal working days, and is an important preprocessing coefficient for modeling enterprise electricity consumption patterns.

[0061] For example, when the target electricity data is the first electricity data, the target type date is the first type date (various dates in the first target month). It is necessary to calculate "the first daily average electricity data for each type of the first type date ÷ the second daily average electricity data (the average daily electricity consumption on weekdays in the first target month)". The resulting ratio is the first electricity coefficient. For instance, if the first daily average electricity consumption on public holidays in the first target month is 500 kWh, and the second daily average electricity consumption (the average daily electricity consumption on weekdays in that month) is 1000 kWh, the ratio of 0.5 is the first electricity coefficient corresponding to public holidays.

[0062] In another example, when the target electricity data is the second type of electricity data, the target type date is the second type of date (various dates within the same month). The calculation involves "the first daily average electricity data for each type of the second type of date ÷ the second daily average electricity data (the average daily electricity consumption on weekdays within the same month)". The resulting ratio is the second electricity coefficient. For example, if the first daily average electricity consumption for statutory holidays in the same month is 300 kWh, and the second daily average electricity consumption (the average daily electricity consumption on weekdays in the same month) is 1000 kWh, the ratio of 0.3 is the second electricity coefficient corresponding to the statutory holidays.

[0063] The above method obtains the target electricity consumption coefficient through the steps of "calculating the average daily electricity consumption of each target type date → extracting the average daily electricity consumption of working days → calculating the ratio" and distinguishing the scenarios corresponding to the first / second electricity consumption data. Its beneficial effects are: it can accurately quantify the electricity consumption ratio of different target type dates (such as public holidays and weekends) relative to working days, effectively remove the interference of date type differences on electricity consumption data, provide a reliable basis for subsequent electricity consumption normalization correction and elimination of the impact of differences in cross-period date composition, and improve the pertinence and accuracy of data processing.

[0064] S204, calculate the meteorological sensitivity factor based on the third meteorological data, the first power coefficient, and the second power coefficient.

[0065] In this embodiment of the application, based on "third meteorological data" (meteorological difference data between the first target month and the same period month), "first electricity consumption coefficient" (electricity consumption ratio of different date types in the first target month relative to working days), and "second electricity consumption coefficient" (electricity consumption ratio of different date types in the same period month relative to working days), and combined with relevant electricity consumption data, a meteorological sensitive factor that can reflect "how electricity consumption changes when the weather changes" is calculated. The core is to use these three types of data to jointly correlate the changing patterns of weather and electricity consumption.

[0066] The above method calculates meteorological sensitivity factors step by step. First, it quantifies the difference in meteorological data to quantify meteorological changes. Then, it eliminates the interference of electricity consumption differences on different date types through two types of electricity coefficients. Finally, it combines multi-dimensional data to obtain accurate meteorological sensitivity factors. Its beneficial effects are: through systematic data processing and correlation analysis, it effectively isolates the influence of date type and meteorological factors on electricity consumption, so that the calculated meteorological sensitivity factors can more realistically reflect the intrinsic relationship between meteorological changes and electricity fluctuations, providing reliable parameter support for subsequent electricity consumption forecasting and improving the scientificity and accuracy of the forecasting model.

[0067] In one embodiment, see Figure 4 This is a schematic diagram of the process for obtaining meteorological sensitive factors provided in the embodiments of this application. Figure 2 ,like Figure 4 As shown, step S204 includes: S401, Normalize the first power data according to the first power coefficient to obtain the first corrected power of the first power data.

[0068] In this embodiment of the application, "normalizing the first electricity consumption data according to the first electricity consumption coefficient to obtain the first corrected electricity consumption" essentially uses the first electricity consumption coefficient (the electricity consumption ratio of different date types in the first target month relative to working days) to eliminate the interference of "date type differences" in the first electricity consumption data (electricity consumption data of the first target month), making the electricity consumption data more focused on core influencing factors such as weather and production. The correction formula is as follows: First, utilize the above-mentioned proportional coefficients for different date types. Electricity consumption of enterprises in previous months (The initial battery data) is normalized to obtain the corrected battery data: Normalization yields corrected power data: (10) in The total number of days in month Mn of year Y; This indicates a date type representing year Y and month Mn.

[0069] First, the first electricity consumption data is the monthly historical electricity consumption data for the first target month (such as the total monthly electricity consumption for July-August 2024). However, this type of data is affected by the proportion of working days, public holidays, and public holidays in that month (for example, if there are many public holidays in a certain month, the total electricity consumption may be lower, which does not reflect changes in actual electricity demand). The first electricity consumption coefficient is the previously calculated "electricity consumption ratio of each date type relative to working days in the first target month" (such as a public holiday coefficient of 0.6 and a public holiday coefficient of 0.3).

[0070] During normalization, the total number of days in the first target month and the number of days for each date type are combined, and the first electricity consumption coefficient is used to adjust the first electricity consumption data. Essentially, the first electricity consumption data is "converted" into the electricity consumption under the scenario of "all working days", eliminating the impact of differences in the composition of date types in different months on the electricity consumption. The final adjusted data is the first corrected electricity consumption.

[0071] In short, this step corrects the date interference of the first electricity consumption data by using the first electricity consumption coefficient, so that the corrected first electricity consumption can more accurately reflect the actual electricity consumption level of the first target month, laying the foundation for subsequent calculation of meteorological sensitive factors and improving the accuracy of forecasts by combining meteorological data.

[0072] S402, normalize the second power data according to the second power coefficient to obtain the second corrected power of the second power data.

[0073] In this embodiment of the application, the second electricity consumption data is normalized according to the second electricity consumption coefficient to obtain the second corrected electricity consumption. The core is to use the second electricity consumption coefficient (the electricity consumption ratio of different date types relative to working days in the same period of the first target month) to eliminate the interference of "date type composition differences" in the second electricity consumption data (electricity consumption data of the same period month), so that the electricity consumption data of the same period month can form a comparable benchmark with the electricity consumption data of the first target month. The correction formula is as follows: Electricity consumption of enterprises in the corresponding month of the previous year (Second battery data) is processed: (11) First, the second electricity consumption data is the monthly historical electricity consumption data of the corresponding month for the first target month (e.g., if the first target month is July-August 2024, the corresponding month is the total monthly electricity consumption of July-August 2023). This type of data will also be affected by the proportion of working days, public holidays, and statutory holidays in the same month (e.g., July 2023 has more holiday days than July 2024, and directly comparing the total electricity consumption of the two months will result in errors due to the difference in date types); while the second electricity consumption coefficient is the previously calculated "electricity consumption ratio of each date type relative to the working days of the same month" (e.g., the public holiday coefficient is 0.55 and the holiday coefficient is 0.28 in July 2023).

[0074] During normalization, the total number of days in the same month and the number of days for each date type are combined, and the second electricity consumption coefficient is used to adjust the second electricity consumption data. Essentially, the electricity consumption data of the same month is "converted" into the electricity consumption under the scenario of "all working days", eliminating the interference of the date type of the same month itself on the electricity consumption. The final adjusted data is the second corrected electricity consumption.

[0075] In short, this step involves "correcting" the electricity consumption data of the same month by using a second electricity consumption coefficient. This allows the second corrected electricity consumption to more accurately reflect the actual electricity consumption level of the same month, thereby forming a cross-year comparison benchmark without date interference with the first corrected electricity consumption (the electricity consumption after correction for the first target month). This provides accurate data support for subsequent calculations of "electricity consumption difference" and solutions to meteorological sensitive factors, perfectly aligning with the core purpose of load forecast data preprocessing in the document.

[0076] S403, calculate the meteorological sensitivity factor based on the first corrected power, the second corrected power, and the third meteorological data.

[0077] In this embodiment of the application, based on "first corrected electricity consumption" (electricity consumption data of the first target month after date type correction), "second corrected electricity consumption" (electricity consumption data of the same month after date type correction), and "third meteorological data" (meteorological difference data between the first target month and the same month), the meteorological sensitive factor that can reflect the "law of change in electricity consumption due to meteorological changes" is calculated by correlation analysis of these three types of data. The core is to use the corrected electricity consumption data and meteorological difference data to remove irrelevant interference and focus on the correlation between meteorology and electricity consumption.

[0078] The above method predicts the electricity consumption of the second target month through the steps of "calculating the third electricity consumption coefficient → normalizing to obtain the third corrected electricity consumption → taking the target and the meteorological data of the same period → combining the forecast with the sensitive factors". Its beneficial effect is that: first, the date type interference of the electricity consumption of the same period in history is eliminated, and then the meteorological data of the target month and the same period and the obtained meteorological sensitive factors are combined, so that the forecasting process relies on reliable historical benchmarks and conforms to the actual meteorological conditions of the target month, effectively improving the accuracy and rationality of the electricity consumption forecast of the second target month.

[0079] In one embodiment, see Figure 5 This is a schematic diagram of the process for obtaining meteorological sensitive factors provided in the embodiments of this application. Figure 3 ,like Figure 5 As shown, step S403 includes: S501, calculate the difference between the first corrected power and the second corrected power to obtain the third corrected power.

[0080] In this application embodiment, the third corrected electricity amount is essentially calculated by removing date type interference by calculating the difference between "corrected electricity amount of the first target month" and "corrected electricity amount of the same period month", focusing on reflecting the electricity amount change across years, and providing the core input of "electricity amount change" dimension for subsequent correlation of meteorological differences and solving meteorological sensitive factors.

[0081] First, the first corrected electricity consumption is "the result of normalizing the electricity consumption data of the first target month by the first electricity consumption coefficient" (corresponding to formula 10), and the second corrected electricity consumption is "the result of normalizing the electricity consumption data of the same month by the second electricity consumption coefficient" (corresponding to formula 11). Both have eliminated the interference of "different proportions of working days / public holidays / holidays" within their respective months, and can truly reflect the actual electricity consumption level of the corresponding month, providing a basis for cross-year comparison.

[0082] Under this premise, the difference between the two is calculated (i.e., "first corrected electricity consumption - second corrected electricity consumption"), and the resulting third corrected electricity consumption essentially quantifies "the change in electricity consumption in the first target month relative to the same period month". For example, if the first corrected electricity consumption for the first target month (July 2024) is 1.2 million kWh, and the second corrected electricity consumption for the same period month (July 2023) is 1 million kWh, the difference of 200,000 kWh is the third corrected electricity consumption. This value directly reflects the change in electricity consumption in July 2024 relative to July 2023.

[0083] S502, establish a linear regression model that includes meteorological sensitive factors based on the third corrected electricity and third meteorological data.

[0084] In this embodiment of the application, the core of the "linear regression model" is to construct a correlation model of "meteorological change → electricity change", and to use meteorological sensitive factors as the core parameters to be solved in the model, so as to quantify the linear relationship between the two.

[0085] Specifically, based on the above third correction of the electricity amount, , integrate target enterprises The meteorological difference value of the region r corresponding to the month M is used to calculate the sensitivity parameter of the meteorological factor, and a multi-source variable linear model is constructed to serve as the weight value of the characteristic parameter of the future power prediction model.

[0086] (12) in, It is a meteorological sensitive factor.

[0087] Transform formula (12) into matrix form: (13) in: A matrix representing the differences in historical weather data (one sample per row, one factor per column). (14) Represents the sensitivity vector of the meteorological factors to be calculated. (15) This represents the vector of historical electricity consumption correction values ​​(current correction value - previous year's correction value). (16) S503, solve the linear regression model to obtain the meteorological sensitivity factors.

[0088] In this embodiment of the application, the basic form of the linear regression model to be solved is the matrix equation F·a = Y (Formula 13 in the document): where “F” is a meteorological difference matrix composed of third meteorological data (the year-on-year difference in meteorological data between the first target month and the same period month, such as temperature and humidity differences), each row corresponds to the multi-dimensional meteorological difference of a historical month, and each column corresponds to the difference of a meteorological element; “a” is the meteorological sensitive factor vector to be solved, each element corresponds to the influence weight of a meteorological element (such as the highest temperature and humidity) on electricity; “Y” is an electricity change vector composed of third corrected electricity (the difference between the first corrected electricity and the second corrected electricity), representing the electricity change across years.

[0089] The solution must satisfy the condition that "sample size ≥ number of meteorological factors" (i.e., the number of rows T of matrix F ≥ the number of columns n) to ensure the validity of the solution. Then, the least squares method is used to minimize the "sum of squared errors between the model's predicted values ​​and the actual changes in electricity consumption," thus deriving the meteorological sensitive factor vector (i.e., the meteorological sensitive factors). The formula for solving " is: (17) S104, Obtain the third power consumption data; wherein, the third power consumption data is the power consumption data of the same month corresponding to the second target month.

[0090] In this embodiment of the application, after the sensitivity parameter estimation is completed, it can be used to predict the correction value for the next Y year and M month. The core of the third electricity data is to collect the electricity data of the same month corresponding to the second target month. The electricity of the "second target month" (i.e., the target month to be predicted, such as October 2024) is predicted; while the "same month corresponding to the second target month" refers to the same month of the previous year that is aligned with the year of the "second target month" (e.g., the same month of October 2024 is October 2023).

[0091] Based on this, the third electricity data is the electricity data for the same month (such as the same month of the previous year), and the data type is consistent with the second data (electricity data of the first target month and the same month) mentioned above. It includes the monthly historical electricity data of the same month (used for subsequent comparison with the predicted value and calculation of the basic electricity) and hourly electricity data (recorded by hour, which can be used as input if further processing of the date-type electricity coefficient of the same month is required).

[0092] S105, based on meteorological sensitive factors and third electricity data, predict the target electricity data corresponding to the second target month; wherein, the first target month is the preceding month of the second target month.

[0093] In this embodiment of the application, the electricity consumption of the "second target month" is predicted based on the calculated "meteorological sensitive factor" and the "electricity consumption data of the corresponding month of the second target month (third electricity consumption data)". It should be noted that the "first target month" used to assist in the calculation of the meteorological sensitive factor is a month before the "second target month" (for example, if the second target month is October 2024, the first target month can be a month before October, such as July-August 2024).

[0094] The above method acquires meteorological data (the first target month and its contemporaneous months) and corresponding electricity consumption data for different time periods in the target area. It then calculates a meteorological sensitive factor that reflects the pattern of electricity consumption changes with weather conditions, and uses this factor to predict electricity consumption for subsequent months (the second target month). By introducing comparative analysis of contemporaneous meteorological and electricity consumption data, the method accurately captures the influence of meteorological factors on electricity consumption (the meteorological sensitive factor), solving the error problems caused by ignoring meteorological variables or using only single-period data in traditional forecasting. This improves the accuracy and applicability of medium- and long-term load forecasting, better meeting the electricity consumption forecasting needs of power sales companies for complex user groups.

[0095] In one embodiment, see Figure 6 This is a flowchart illustrating the prediction of target power data provided in an embodiment of this application. Figure 1 ,like Figure 6 As shown, step S105 includes: S601, calculate the third electricity coefficient based on the third electricity data; wherein, the third electricity coefficient represents the ratio coefficient between the third type date and the working day type date of the same month corresponding to the second target month; wherein, the third type date includes at least one of working days, public holidays and statutory holidays.

[0096] In this application embodiment, the third electricity consumption coefficient is essentially based on the electricity consumption data (third electricity consumption data) of the corresponding month of the second target month, quantifying the electricity consumption ratio of "third type dates" (such as public holidays and statutory holidays) and "working days" within the same month. The core logic is completely consistent with the idea of ​​"calculation of date type electricity consumption coefficient" (formula 7-9) above.

[0097] Specifically, the hourly electricity data for the same month is extracted from the third electricity data, and the hourly electricity data for each day is summed according to the logic of the document formula (7) to obtain the total electricity for each day in the same month.

[0098] Then, based on the time dimension tag library, the daily days of the same month are mapped to the corresponding third type of date (such as working days and public holidays), and according to the logic of the above formula (8), the daily average electricity consumption of "working days" and each "third type of date" is calculated respectively - that is, after summing the daily total electricity consumption of the same type of date, it is divided by the total number of days of that type of date to obtain the daily average electricity consumption of each date type.

[0099] Finally, based on the "average daily electricity consumption on weekdays" of the same month, and following the calculation logic of the "date type electricity consumption ratio coefficient" in formula (9) above, the "average daily electricity consumption of each third type date" is divided by the "average daily electricity consumption on weekdays," and the resulting ratio is the third electricity consumption coefficient. For example, if the average daily electricity consumption on public holidays in the same month is 58% of that on weekdays, then the third electricity consumption coefficient corresponding to the public holidays in that month is 0.58.

[0100] S602, the third energy data is normalized and corrected according to the third energy coefficient to obtain the third corrected energy.

[0101] In this embodiment of the application, the third corrected electricity consumption is to use the third electricity consumption coefficient (the electricity consumption ratio of the date type in the same month corresponding to the second target month) to eliminate the interference of "date type composition difference" in the third electricity consumption data (the electricity consumption data of the same month), and convert it into a unified "working day benchmark" electricity consumption, so as to provide a comparable historical reference for the electricity consumption forecast of the subsequent target month. The specific logic is consistent with the core idea of ​​the "electricity consumption correction" section in the document (consistent with the normalization logic of formulas 10 and 11).

[0102] The third electricity consumption data is the electricity consumption data of the "second target month corresponding to the same period month" (e.g., if the second target month is October 2024 and the same period month is October 2023, the third electricity consumption data is the monthly historical electricity consumption of October 2023). This type of data will be affected by the proportion of working days, public holidays, and statutory holidays in the same period month (for example, October 2023 has many holidays, and directly using its total monthly electricity consumption as a benchmark will underestimate the actual electricity consumption level); while the third electricity consumption coefficient is the previously calculated "electricity consumption ratio of each third type of date in the same period month relative to working days" (e.g., the public holiday coefficient is 0.58 and the holiday coefficient is 0.29 in October 2023), which is a key parameter for quantifying the differences in electricity consumption by date type.

[0103] During normalization correction, the total number of days in the same month and the actual number of days for each of the third-type dates are considered. A third-electricity coefficient is then used to adjust the third-electricity data. Essentially, this "converts" the third-electricity data to the electricity consumption under the scenario of "assuming the month consists entirely of working days," eliminating the interference from the date types within the same month. For example, October 2023 has 31 days (22 working days and 9 public holidays). The third-electricity data (total monthly electricity consumption) is 1.1 million kWh. Using a public holiday coefficient of 0.58, the corrected third-electricity data is the "working-day baseline" electricity consumption after removing the differences in public holiday electricity consumption, more accurately reflecting the actual electricity demand level for the same month.

[0104] S603, Obtain the third data; wherein the third data includes the fourth meteorological data and the fifth meteorological data; wherein the fourth meteorological data is the meteorological data corresponding to the second target month; and the fifth meteorological data is the meteorological data corresponding to the same period month corresponding to the second target month.

[0105] In this application embodiment, the two types of meteorological data included in the third data have clearly defined functions: First, the fourth meteorological data, which is the meteorological data of the "second target month" to be predicted (such as October 2024), which needs to cover core elements closely related to electricity consumption, such as daily maximum temperature, minimum temperature, relative humidity, atmospheric pressure, etc. These data directly reflect the actual meteorological conditions of the predicted month; Second, the fifth meteorological data, which is the meteorological data of the "same period month corresponding to the second target month" (such as October 2023 corresponding to October 2024), whose data type is completely consistent with the fourth meteorological data, and is used to form a cross-year meteorological comparison with the fourth meteorological data to quantify the meteorological changes of the predicted month relative to the historical same period.

[0106] Similarly, the system first indexes meteorological data for each enterprise's region from the input meteorological data source. This data includes various meteorological elements such as maximum temperature (Tmax), minimum temperature (Tmin), relative humidity (H), and pressure (P). Meteorological data is one of the important influencing factors considered in subsequent analysis because meteorological conditions have a significant impact on enterprises' electricity consumption behavior; for example, enterprises' electricity consumption for refrigeration equipment will increase during hot weather.

[0107] S604 predicts target electricity data based on third corrected electricity, third data, and meteorological sensitive factors.

[0108] The above method predicts the electricity consumption of the second target month through the steps of "calculating the third electricity consumption coefficient → normalizing to obtain the third corrected electricity consumption → taking the target and the meteorological data of the same period → combining the forecast with the sensitive factors". Its beneficial effect is that: first, the date type interference of the electricity consumption of the same period in history is eliminated, and then the meteorological data of the target month and the same period and the obtained meteorological sensitive factors are combined, so that the forecasting process relies on reliable historical benchmarks and conforms to the actual meteorological conditions of the target month, effectively improving the accuracy and rationality of the electricity consumption forecast of the second target month.

[0109] In this embodiment of the application, the target electricity data for the second target month (the month to be predicted) is calculated by using the logic of "historical baseline electricity + impact of meteorological changes".

[0110] In one embodiment, see Figure 7 This is a flowchart illustrating the prediction of target power data provided in an embodiment of this application. Figure 2 ,like Figure 7 As shown, step S604 includes: S701, perform difference calculation on the fourth and fifth meteorological data in the third data to obtain the sixth meteorological data.

[0111] In this embodiment of the application, the fourth meteorological data is the meteorological data of the "second target month" to be predicted (such as October 2024), which includes core elements closely related to electricity consumption such as maximum temperature, minimum temperature, relative humidity, and pressure; the fifth meteorological data is the meteorological data of the "corresponding month of the second target month" (such as October 2023), and the data type is completely consistent with the fourth meteorological data (same elements, same statistical dimensions), ensuring the effectiveness of the difference calculation.

[0112] When calculating the difference, the corresponding elements of the two types of meteorological data need to be subtracted one by one (i.e., "the value of a certain element in the fourth meteorological data - the value of the same element in the fifth meteorological data"). The difference result for each meteorological element is the sixth meteorological data. For example, the average maximum temperature in October 2024 (the second target month) is 22℃, and the average maximum temperature in October 2023 (the same month) is 19℃. The difference between the two + 3℃ is the element value of "average maximum temperature" in the sixth meteorological data. If the difference in relative humidity is -5%, it means that the second target month is 5 percentage points drier than the same month.

[0113] S702, the first value is obtained by weighting and summing the sixth meteorological data with the meteorological sensitive factor.

[0114] In this embodiment of the application, the sixth meteorological data is the difference in meteorological elements between the second target month and the same month (such as the difference in maximum temperature +3℃, the difference in relative humidity -5%, etc.), representing the "amplitude of meteorological change"; the meteorological sensitive factor is a parameter previously solved by a linear regression model (such as a change in electricity consumption of 20,000 kWh per ℃ of temperature change, a change in electricity consumption of 5,000 kWh per % of humidity change, etc.), representing the "weight of the impact of unit meteorological change on electricity consumption", and the meteorological element dimensions of the two are completely corresponding (such as temperature difference corresponding to temperature sensitive factor, humidity difference corresponding to humidity sensitive factor).

[0115] When performing weighted summation, the difference of each meteorological element in the sixth meteorological data needs to be multiplied one by one by the corresponding meteorological sensitivity factor (i.e., "difference of a meteorological element × sensitivity factor of that element") to obtain the single impact value of the element's change on electricity. Then, the single impact values ​​of all meteorological elements are added together, and the final sum is the first value. For example, temperature difference + 3℃ × temperature sensitivity factor 20,000 kWh / ℃ = +60,000 kWh, humidity difference - 5% × humidity sensitivity factor 5,000 kWh / % = -25,000 kWh. The sum of the two, 35,000 kWh, is the first value, representing the total incremental impact of the combined change of the two meteorological elements on electricity.

[0116] S703, sum the first value and the third corrected quantity to obtain the fourth corrected quantity.

[0117] In this embodiment of the application, the third corrected electricity consumption is the electricity consumption of the "second target month corresponding to the same period month" after date type correction (date interference has been eliminated, such as 1 million kWh after correction for October 2023), representing the real electricity consumption benchmark of the same period in history; the first value is the result of the weighted sum of the sixth meteorological data (the meteorological difference between the second target month and the same period month) and the meteorological sensitive factor (such as +80,000 kWh), representing the total increase in electricity consumption caused by meteorological changes in the second target month relative to the same period month (positive value is an increase, negative value is a decrease).

[0118] When summing the values, the two are directly added together (i.e., "third corrected electricity consumption + first value"), and the result is the fourth corrected electricity consumption. For example, the third corrected electricity consumption of 1 million kWh plus the first value of 80,000 kWh results in 1.08 million kWh, which is the fourth corrected electricity consumption. This result is essentially the "predicted electricity consumption assuming all working days in the second target month," which references historical electricity consumption benchmarks for the same period and incorporates the impact of current month weather changes, laying the foundation for subsequent elimination of date type interference and obtaining the final target electricity consumption prediction. The formula is as follows: The revised value forecast for month M in year Y is shown in the following formula: (18) S704, the fourth corrected power quantity is denormalized to obtain the target power quantity data.

[0119] In this embodiment of the application, the target electricity data is obtained by inverse normalization of the fourth corrected electricity amount. The core is to restore the fourth corrected electricity amount (the predicted electricity amount based on the second target month being "all working days") to the actual electricity amount composed of actual date types, eliminating the interference of the "working day basis" assumption. Specifically: First, we need to determine the actual date type composition of the second target month (e.g., October 2024 has 31 days, including 22 working days and 9 public holidays), and the corresponding date type electricity consumption coefficient. (Referring to the logic of the previous "third electricity consumption coefficient," which is the proportion of electricity consumption for each date type in the second target month relative to weekdays, such as a public holiday coefficient of 0.58). These data are key parameters for converting "weekday baseline electricity consumption" into "actual date electricity consumption."

[0120] During the denormalization process, first, based on the number of days and corresponding coefficients of each date type in the second target month, calculate the "date type correction factor": divide the sum of "(number of days for a certain date type × electricity coefficient for that type)" by the "total number of days in the second target month" to obtain the average correction factor per unit number of days; then multiply the fourth corrected electricity (the monthly baseline electricity for "all working days") by the "average correction factor" to restore the electricity under the actual date type composition of the second target month, which is the final target electricity data. The formula is: (19) For example, the fourth revised electricity volume is 1.08 million kWh (assuming that all the predicted electricity volume in October 2024 is working days). The monthly electricity volume is "(22 days × 1 + 9 days × 0.58) = 22 + 5.22 = 27.22", and the average correction factor is "27.22 ÷ 31 ≈ 0.878". Therefore, the target electricity volume data is "1.08 million kWh × 0.878 ≈ 948,000 kWh", which is the final predicted electricity volume that matches the actual working days / public holidays of that month.

[0121] In short, inverse normalization is a key step in "regressing from baseline assumptions to real-world scenarios". Its core function is to match the prediction results with the actual date type distribution of the second target month, ensuring that the target electricity data can directly reflect the demand of the actual electricity consumption scenario and complete the final output of the entire prediction process.

[0122] The above method predicts the target electricity consumption through the steps of "calculating meteorological differences → calculating the incremental impact of meteorological conditions on electricity consumption → obtaining the baseline predicted electricity consumption → restoring the actual date scenario". Its beneficial effect is that it first quantifies the meteorological differences between the target and the same period and their impact on electricity consumption, and then combines the historical baseline electricity consumption and restores the actual date type composition, so that the prediction is not only accurately related to the meteorological change pattern, but also fits the real electricity consumption scenario of the target month, and finally outputs accurate target electricity consumption data that meets actual needs.

[0123] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0124] Corresponding to the power data prediction method described in the above embodiments, Figure 8 This is a structural block diagram of a power data prediction device provided in an embodiment of this application. For ease of explanation, only the parts related to the embodiment of this application are shown.

[0125] Reference Figure 8 The device 8 includes: The first data acquisition module 81 is used to acquire first data; wherein the first data includes first meteorological data and second meteorological data corresponding to the target area, wherein the first meteorological data is meteorological data corresponding to at least one first target month corresponding to the target area; and the second meteorological data is meteorological data corresponding to the same month corresponding to the first target month. The second data acquisition module 88 is used to acquire second data; wherein the second data includes first power data and second power data corresponding to the target area; wherein the first power data is power data corresponding to at least one first target month corresponding to the target area; and the second power data is power data corresponding to the same month corresponding to the first target month. The meteorological sensitivity factor calculation module 83 is used to calculate the meteorological sensitivity factor based on the first data and the second data; wherein, the meteorological sensitivity factor is used to represent the variation pattern of electricity data with meteorological data. The third power data acquisition module 84 is used to acquire third power data; wherein, the third power data is the power data of the same month corresponding to the second target month; The power data prediction module 85 is used to predict the target power data corresponding to the second target month based on the meteorological sensitive factor and the third power data; wherein, the first target month is the preceding month of the second target month.

[0126] Optionally, the meteorological sensitivity factor calculation module 83 is also used for: The difference between the first meteorological data and the second meteorological data in the first data is calculated to obtain the third meteorological data; A first power consumption coefficient is calculated based on the first power consumption data in the second data; wherein, the first power consumption coefficient represents the ratio coefficient between a first type of date and a working day type of date corresponding to the first target month; wherein, the first type of date includes at least one of working days, public holidays and statutory holidays; A second electricity coefficient is calculated based on the second electricity data in the second data; wherein, the second electricity coefficient represents the ratio coefficient between the second type of date and the working day type date corresponding to the same period of the first target month; wherein, the second type of date includes at least one of working days, public holidays and statutory holidays; The meteorological sensitivity factor is calculated based on the third meteorological data, the first power coefficient, and the second power coefficient.

[0127] Optionally, the meteorological sensitivity factor calculation module 83 is also used for: Calculate the first daily average electricity consumption data corresponding to each of the target types of dates based on the target electricity consumption data; Obtain second average daily power consumption data from multiple first average daily power consumption data; wherein, the second average daily power consumption data is the average daily power consumption data corresponding to the working day with the target type date; The ratio between each of the first daily average power consumption data and the second daily average power consumption data is calculated to obtain the target power consumption coefficient; wherein, when the target power consumption data is the first power consumption data, the target type date is the first type date, and the target power consumption coefficient is the first power consumption coefficient; when the target power consumption data is the second power consumption data, the target type date is the second type date, and the target power consumption coefficient is the second power consumption coefficient.

[0128] Optionally, the meteorological sensitivity factor calculation module 83 is also used for: The first power data is normalized according to the first power coefficient to obtain the first corrected power of the first power data; The second power data is normalized according to the second power coefficient to obtain the second corrected power data. The meteorological sensitivity factor is calculated based on the first corrected electrical charge, the second corrected electrical charge, and the third meteorological data.

[0129] Optionally, the meteorological sensitivity factor calculation module 83 is also used for: A third electricity coefficient is calculated based on the third electricity data; wherein the third electricity coefficient represents the ratio between the third type of date and the working day type date of the same month corresponding to the second target month; wherein the third type of date includes at least one of working days, public holidays and statutory holidays; The third power data is normalized and corrected according to the third power coefficient to obtain the third corrected power. Obtain third data; wherein the third data includes fourth meteorological data and fifth meteorological data; wherein the fourth meteorological data is the meteorological data corresponding to the second target month; and the fifth meteorological data is the meteorological data corresponding to the same month as the second target month; The target power data is predicted based on the calculated third corrected power, the third data, and the meteorological sensitive factor.

[0130] Optionally, the power data prediction module 85 is also used for: The difference between the fourth and fifth meteorological data in the third data is calculated to obtain the sixth meteorological data. The first value is obtained by weighting and summing the sixth meteorological data with the meteorological sensitive factor; The first value and the third corrected quantity are summed to obtain the fourth corrected quantity; The fourth corrected power quantity is inversely normalized to obtain the target power quantity data.

[0131] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.

[0132] in addition, Figure 8 The power data prediction device shown can be a software unit, a hardware unit, or a combination of software and hardware built into an existing terminal device. It can also be integrated into the terminal device as a separate component, or exist as a standalone terminal device.

[0133] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0134] Figure 9 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. For example... Figure 9 As shown, the terminal device 9 of this embodiment includes: at least one processor 90 ( Figure 9 (Only one is shown in the diagram) a processor, a memory 91, and a computer program 92 stored in the memory 91 and executable on at least one processor 90, wherein the processor 90 executes the computer program 92 to implement the steps in any of the above-described embodiments of the power data prediction method.

[0135] The terminal device can be a computing device such as a desktop computer, laptop, handheld computer, or cloud server. This terminal device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that... Figure 9This is merely an example of terminal device 9 and does not constitute a limitation on terminal device 9. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0136] The processor 90 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0137] In some embodiments, memory 91 may be an internal storage unit of terminal device 9, such as a hard disk or memory of terminal device 9. In other embodiments, memory 91 may be an external storage device of terminal device 9, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on terminal device 9. Furthermore, memory 91 may include both internal storage units and external storage devices of terminal device 9. Memory 91 is used to store operating system, application programs, boot loader, location information, and other programs, such as program code of computer programs. Memory 91 may also be used to temporarily store location information that has been output or will be output.

[0138] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, can implement the steps in the above-described method embodiments.

[0139] This application provides a computer program product that, when run on a terminal device, enables the terminal device to implement the steps described in the various method embodiments above.

[0140] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. A computer-readable medium can include at least: any entity or device capable of carrying computer program code to a device / terminal equipment, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0141] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0142] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0143] In the embodiments provided in this application, it should be understood that the disclosed apparatus / terminal devices and methods can be implemented in other ways. For example, the apparatus / terminal device embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0144] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0145] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for predicting electricity data, characterized in that, The method includes: Acquire first data; wherein the first data includes first meteorological data and second meteorological data corresponding to the target area, wherein the first meteorological data is meteorological data corresponding to at least one first target month corresponding to the target area; and the second meteorological data is meteorological data corresponding to the same month corresponding to the first target month; Obtain second data; wherein the second data includes first power consumption data and second power consumption data corresponding to the target area; wherein the first power consumption data is power consumption data corresponding to at least one first target month corresponding to the target area; and the second power consumption data is power consumption data corresponding to the same month as the first target month. A meteorological sensitivity factor is calculated based on the first data and the second data; wherein, the meteorological sensitivity factor is used to represent the variation pattern of electricity data with meteorological data. Obtain third electricity data; wherein, the third electricity data is the electricity data of the same month corresponding to the second target month; Based on the meteorological sensitive factors and the third electricity data, the target electricity data corresponding to the second target month is predicted; wherein, the first target month is the preceding month of the second target month.

2. The power data prediction method as described in claim 1, characterized in that, The calculation of meteorological sensitivity factors based on the first data and the second data includes: The difference between the first meteorological data and the second meteorological data in the first data is calculated to obtain the third meteorological data; A first power consumption coefficient is calculated based on the first power consumption data in the second data; wherein, the first power consumption coefficient represents the ratio coefficient between a first type of date and a working day type of date corresponding to the first target month; wherein, the first type of date includes at least one of working days, public holidays and statutory holidays; A second electricity coefficient is calculated based on the second electricity data in the second data; wherein, the second electricity coefficient represents the ratio coefficient between the second type of date and the working day type date corresponding to the same period of the first target month; wherein, the second type of date includes at least one of working days, public holidays and statutory holidays; The meteorological sensitivity factor is calculated based on the third meteorological data, the first power coefficient, and the second power coefficient.

3. The power data prediction method as described in claim 2, characterized in that, The steps for calculating the target electricity coefficient corresponding to the target type date in the first target month based on the target electricity data include: Calculate the first daily average electricity consumption data corresponding to each of the target types of dates based on the target electricity consumption data; Obtain second average daily power consumption data from multiple first average daily power consumption data; wherein, the second average daily power consumption data is the average daily power consumption data corresponding to the working day with the target type date; The ratio between each of the first daily average power consumption data and the second daily average power consumption data is calculated to obtain the target power consumption coefficient; wherein, when the target power consumption data is the first power consumption data, the target type date is the first type date, and the target power consumption coefficient is the first power consumption coefficient; when the target power consumption data is the second power consumption data, the target type date is the second type date, and the target power consumption coefficient is the second power consumption coefficient.

4. The power data prediction method as described in claim 3, characterized in that, The calculation of the meteorological sensitivity factor based on the third meteorological data, the first power coefficient, and the second power coefficient includes: The first power data is normalized according to the first power coefficient to obtain the first corrected power of the first power data; The second power data is normalized according to the second power coefficient to obtain the second corrected power data. The meteorological sensitivity factor is calculated based on the first corrected electrical charge, the second corrected electrical charge, and the third meteorological data.

5. The power data prediction method as described in claim 4, characterized in that, The calculation of the meteorological sensitivity factor based on the first corrected electrical quantity, the second corrected electrical quantity, and the third meteorological data includes: The difference between the first corrected power and the second corrected power is calculated to obtain the third corrected power. A linear regression model incorporating meteorological sensitive factors is established based on the third corrected electricity and the third meteorological data. The meteorological sensitivity factor is obtained by solving the linear regression model.

6. The power data prediction method as described in claim 5, characterized in that, The prediction of the target electricity data corresponding to the second target month based on the meteorological sensitive factor and the third electricity data includes: A third electricity coefficient is calculated based on the third electricity data; wherein the third electricity coefficient represents the ratio between the third type of date and the working day type date of the same month corresponding to the second target month; wherein the third type of date includes at least one of working days, public holidays and statutory holidays; The third power data is normalized and corrected according to the third power coefficient to obtain the third corrected power. Obtain third data; wherein the third data includes fourth meteorological data and fifth meteorological data; wherein the fourth meteorological data is the meteorological data corresponding to the second target month; and the fifth meteorological data is the meteorological data corresponding to the same period month corresponding to the second target month; The target power data is predicted based on the third corrected power, the third data, and the meteorological sensitive factor.

7. The power data prediction method as described in claim 6, characterized in that, The prediction of the target electricity data based on the third corrected electricity level, the third data, and the meteorological sensitivity factor includes: The difference between the fourth and fifth meteorological data in the third data is calculated to obtain the sixth meteorological data. The first value is obtained by weighting and summing the sixth meteorological data with the meteorological sensitive factor; The first value and the third corrected quantity are summed to obtain the fourth corrected quantity; The fourth corrected power quantity is denormalized to obtain the target power quantity data.

8. A power data prediction device, characterized in that, include: A first data acquisition module is used to acquire first data; wherein the first data includes first meteorological data and second meteorological data corresponding to a target area, wherein the first meteorological data is meteorological data corresponding to at least one first target month corresponding to the target area; and the second meteorological data is meteorological data corresponding to the same month corresponding to the first target month. The second data acquisition module is used to acquire second data; wherein the second data includes first power data and second power data corresponding to the target area; wherein the first power data is power data corresponding to at least one first target month corresponding to the target area; and the second power data is power data corresponding to the same month as the first target month. A meteorological sensitivity factor calculation module is used to calculate a meteorological sensitivity factor based on the first data and the second data; wherein, the meteorological sensitivity factor is used to represent the variation pattern of electricity data with meteorological data. The third power data acquisition module is used to acquire third power data; wherein, the third power data is the power data of the same month corresponding to the second target month; The power data prediction module is used to predict the target power data corresponding to the second target month based on the meteorological sensitive factor and the third power data; wherein, the first target month is the preceding month of the second target month.

9. A terminal device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1 to 7.

Images