Three-dimensional multi-time scale load prediction method based on transverse and longitudinal multi-dimensional check

Abstract

The invention discloses a three-dimensional multi-time scale load prediction method based on transverse and longitudinal multi-dimensional check, and relates to the technical field of power load prediction, and the method comprises the steps: collecting historical load data and environment data of a power grid, and dividing the historical load data into a short-term load data set, a middle-term load data set and a long-term load data set according to multi-time scale division; under each time scale, carrying out load characteristic analysis based on a transverse and longitudinal multi-dimensional checking method to obtain a load change rule under each time scale; and according to the load characteristic analysis result of each time scale, constructing a multi-dimensional load prediction model to predict the future load. Through hierarchical analysis and comprehensive modeling of load characteristics of multiple time scales, dynamic change characteristics of the load are accurately captured, prediction precision is improved, scientific support is provided for short-term scheduling and medium-and-long-term planning of a power system, power resource distribution is optimized, and the overall efficiency and stability of the system are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power load forecasting, and in particular to a three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification. Background Art

[0002] With the continuous growth of electricity demand in modern society, power load forecasting technology has become an important guarantee for the stable operation and scientific dispatch of power systems. Traditional load forecasting methods usually use a single time scale for forecasting, which makes it difficult to fully consider the differences in short-term, medium-term and long-term load characteristics. At the same time, existing load forecasting methods do not adequately consider the impact of environmental factors such as light intensity, resulting in limited forecasting accuracy.

[0003] In the existing technology, linear regression, time series analysis, machine learning and other methods are mainly used to predict load. However, these methods are mostly based on a fixed time scale, ignoring the dynamic changes of load characteristics at different time scales, and cannot adapt to the complex and changeable load environment. In addition, the impact of dynamic changes in environmental data on load has not been fully quantified, especially when the light intensity fluctuates violently, the accuracy of load prediction will be significantly affected. Summary of the invention

[0004] Based on the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification to solve the above technical problems.

[0005] To achieve the above object, the present invention provides the following technical solution: a three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification, comprising:

[0006] Collecting historical load data and environmental data of the power grid, and dividing the historical load data into short-term, medium-term and long-term load data sets according to multiple time scales;

[0007] At each time scale, load characteristics analysis is performed based on the horizontal and vertical multi-dimensional verification method to obtain the load variation law at each time scale;

[0008] According to the analysis results of load characteristics at each time scale, a multi-dimensional load forecasting model is constructed to predict future loads.

[0009] The present invention is further configured such that the historical load data includes a load timestamp and a grid load; the environmental data includes a light timestamp and light intensity, wherein the light intensity is a normalized value.

[0010] The present invention is further configured to perform time pairing according to the load timestamp and the light timestamp to obtain the grid load and light intensity in the time series;

[0011] Divide the time series according to the time scale to obtain short-term, medium-term and long-term data sets;

[0012] Perform horizontal and vertical multi-dimensional verification based on short-term data sets to analyze short-term load characteristics;

[0013] Conduct horizontal and vertical multi-dimensional verification based on the mid-term data set to analyze mid-term load characteristics;

[0014] Conduct horizontal and vertical multi-dimensional verification based on long-term data sets to analyze long-term load characteristics;

[0015] According to the short-term load characteristics, medium-term load characteristics and long-term load characteristics, a multi-dimensional load forecasting model is constructed to predict future loads.

[0016] The present invention is further configured to perform horizontal and vertical multi-dimensional verification based on the short-term data set to analyze the short-term load characteristics, including:

[0017] Based on the short-term data set, the horizontal dimension is checked and the response function of short-term load change to light intensity is calculated;

[0018] The longitudinal dimension is calibrated based on the short-term data set and the short-term load change rate function is calculated.

[0019] The present invention is further configured such that the calculation logic of the response function of the short-term load change to the light intensity is: Among them, R s (G s ) is the response function of short-term load change to light intensity, τ0 and τ n is the start and end time of the short-term load data, P s (τ) is the grid load at time τ, G s (τ) is the light intensity at time τ, α is the response coefficient, and δ is the regularization constant;

[0020] The calculation logic of the short-term load change rate function is: Among them, V s( τ ) is the short-term load change rate at time τ.

[0021] The present invention is further configured to perform horizontal and vertical multi-dimensional verification based on the mid-term data set to analyze the mid-term load characteristics, including:

[0022] Perform lateral dimension calibration based on the mid-term data set and calculate the harmonic response function of mid-term load changes;

[0023] The longitudinal dimension is calibrated based on the medium-term data set, and the medium-term load fluctuation correlation function is calculated.

[0024] The present invention is further configured such that the calculation logic of the harmonic response function of the mid-term load change is: Among them, H m( G m) is the harmonic response function of the mid-term load change, K is the total number of mid-term data points, θ k is the timestamp corresponding to the kth mid-term data point, P m (θ k ) is the time at time θ k The grid load at that time, G m (θ k ) is the time at time θ k The light intensity at that time;

[0025] The calculation logic of the medium-term load fluctuation correlation function is: Among them, C m (θ) is the medium-term load fluctuation correlation function, θ0 and θ n are the start and end time of the mid-term load data, and η is the integral variable.

[0026] The present invention is further configured to perform horizontal and vertical multi-dimensional verification based on a long-term data set to analyze long-term load characteristics, including:

[0027] Perform lateral dimension calibration based on long-term data sets and calculate environmental deviation functions for long-term loads;

[0028] The longitudinal dimension is calibrated based on the long-term data set and the long-term load increment cumulative function is calculated.

[0029] The present invention is further configured such that the calculation logic of the environmental deviation function of the long-term load is: Among them, D l (G l ) is the environmental deviation function of the long-term load, φ0 and φ n is the start and end time of long-term load data, P l (φ) is the grid load at time φ, G l (φ) is the light intensity at time φ, G avg is the long-term average light intensity;

[0030] The calculation logic of the long-term load increment accumulation function is: Among them, J l (φ) is the cumulative function of long-term load increment, J is the total number of long-term data points, φ j is the timestamp corresponding to the j-th long-term data point, P l (φ j ) is the time at which j The grid load at that time, G l (φj ) is the time at which j The light intensity at that time.

[0031] The present invention is further configured to construct a multi-dimensional load forecasting model according to short-term load characteristics, medium-term load characteristics and long-term load characteristics. The construction logic of the multi-dimensional load forecasting model is:

[0032] Where Q(t) is the load forecast at time t+1, τ t and τ t-1 is the timestamp corresponding to the current moment and the previous moment in the short-term data set, J l (φ t ) and J l (φ t-1 ) is the cumulative function value of the long-term load increment corresponding to the current moment and the previous moment in the long-term data set, λ1, λ2 and λ3 are weight coefficients, all greater than 0, and λ1+λ2+λ3=1.

[0033] The present invention provides a three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification. The method collects historical load data and environmental data of the power grid, and divides the historical load data into short-term, medium-term and long-term load data sets according to multi-time scale division; at each time scale, load characteristic analysis is performed based on horizontal and vertical multi-dimensional verification methods to obtain the load change law at each time scale; according to the load characteristic analysis results at each time scale, a multi-dimensional load forecasting model is constructed to predict future loads, and the beneficial effects produced include:

[0034] 1. Improve the accuracy of load forecasting: By dividing historical load data into short-term, medium-term and long-term load data, and performing feature extraction and analysis at different time scales, the dynamic change characteristics of load at each time scale can be effectively captured. Compared with the traditional single time scale forecasting method, the present invention can describe the load characteristics more finely, thereby improving the forecasting accuracy;

[0035] 2. Achieve comprehensive predictions at different time scales: By analyzing short-term, medium-term and long-term load characteristics in layers and conducting comprehensive modeling, it is possible to take into account both short-term fluctuations and long-term trends, thereby more comprehensively reflecting the changing patterns of loads. This method is not only suitable for short-term load dispatching decisions, but also can provide an effective reference for medium- and long-term power planning;

[0036] 3. Improve the dispatching and planning of power systems: Since the present invention introduces a multi-dimensional and multi-time scale load characteristic analysis method in load forecasting and constructs a multi-dimensional prediction model, it can provide more scientific decision support for the dispatching and planning of power systems. By accurately predicting load changes at different time scales, the allocation of power resources can be optimized and the overall efficiency and stability of the power system can be improved.

[0037] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0039] Figure 1 The flowchart of a three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification is shown as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0040] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0041] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0042] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0043] A three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification, such as Figure 1 As shown, including:

[0044] Collecting historical load data and environmental data of the power grid, and dividing the historical load data into short-term, medium-term and long-term load data sets according to multiple time scales;

[0045] At each time scale, load characteristics analysis is performed based on the horizontal and vertical multi-dimensional verification method to obtain the load variation law at each time scale;

[0046] According to the analysis results of load characteristics at each time scale, a multi-dimensional load forecasting model is constructed to predict future loads.

[0047] The present invention is further configured such that the historical load data includes a load timestamp and a grid load; the environmental data includes a light timestamp and a light intensity, wherein the light intensity is a normalized value. Specifically, the introduction of historical load data includes a load timestamp and a grid load. The load timestamp is the recording time of the load data, which can be used to accurately track and analyze the changes in the load at different times; the grid load represents the actual load value of the grid at a certain moment; the introduction of environmental data includes a light timestamp and a light intensity. The light timestamp is paired with the load timestamp to synchronize the environmental data and the load data; the light intensity affects the photovoltaic power generation, and the photovoltaic power generation significantly affects the grid load. Therefore, when there is a photovoltaic power generation unit in the power system, only the environmental factor of light intensity is considered, and the influence of other factors can be ignored when the light intensity exists. Furthermore, in order to adapt to the model requirements, the light intensity is normalized to ensure that the value range of the variable in the model is consistent, which is convenient for subsequent analysis and calculation.

[0048] The present invention is further configured to perform time pairing according to the load timestamp and the light timestamp to obtain the grid load and light intensity in the time series;

[0049] Divide the time series according to the time scale to obtain short-term, medium-term and long-term data sets;

[0050] Perform horizontal and vertical multi-dimensional verification based on short-term data sets to analyze short-term load characteristics;

[0051] Conduct horizontal and vertical multi-dimensional verification based on the mid-term data set to analyze mid-term load characteristics;

[0052] Conduct horizontal and vertical multi-dimensional verification based on long-term data sets to analyze long-term load characteristics;

[0053] According to the short-term load characteristics, medium-term load characteristics and long-term load characteristics, a multi-dimensional load forecasting model is constructed to predict future loads.

[0054] The present invention is further configured to perform horizontal and vertical multi-dimensional verification based on the short-term data set to analyze the short-term load characteristics, including:

[0055] A lateral dimension check is performed based on the short-term data set to calculate the response function of the short-term load change to the light intensity. The present invention is further configured such that the calculation logic of the response function of the short-term load change to the light intensity is: Among them, R s (G s ) is the response function of short-term load change to light intensity, τ0 and τ n is the start and end time of the short-term load data, P s (τ) is the grid load at time τ, G s (τ) is the light intensity at time τ, α is the response coefficient, and δ is the regularization constant; specifically, in order to analyze the relationship between short-term load changes and light intensity, a response function for calculating short-term load changes to light intensity is proposed. This response function cumulatively analyzes the dynamic response of load changes under different levels of light intensity through an integral method. Specifically, this response function accumulates the interaction between grid load and light intensity over a time period, and describes the sensitivity of load changes to light intensity through an exponential function, which helps to build environmental adaptability in the load forecasting model. The response coefficient and the regularization constant both range from 0 to 1.

[0056] Based on the short-term data set, the longitudinal dimension is checked and the short-term load change rate function is calculated. The calculation logic of the short-term load change rate function is: Among them, V s (τ) is the short-term load change rate at time τ. Specifically, the short-term load change rate is calculated to capture the sensitivity of short-term load fluctuations, thereby better adapting to rapidly changing environmental conditions. The load change rate function is obtained by fuzzy dimensional kernel construction of the short-term load data set; by calculating the short-term load change rate, the model can quickly identify sudden changes in the load, is suitable for rapidly changing environments, and enhances the response speed of the model.

[0057] The present invention is further configured to perform horizontal and vertical multi-dimensional verification based on the mid-term data set to analyze the mid-term load characteristics, including:

[0058] Based on the mid-term data set, a lateral dimension check is performed to calculate the harmonic response function of the mid-term load change. The present invention is further configured such that the calculation logic of the harmonic response function of the mid-term load change is: Among them, H m (G m ) is the harmonic response function of the mid-term load change, K is the total number of mid-term data points, θ k is the timestamp corresponding to the kth mid-term data point, P m (θ k ) is the time at time θ k The grid load at that time, G m (θ k ) is the time at time θ k Specifically, the load variation law on the medium-term scale is analyzed by accumulating medium-term load data and harmonic response of light intensity. Specifically, by combining the grid load with light intensity data and calculating its harmonic response at each time point, the comprehensive impact of load change on the medium-term time scale is obtained. By introducing the harmonic response of light intensity, the law of medium-term load change with the environment can be analyzed more accurately. The introduction of the harmonic function helps the model better capture the dynamic changes of load under light intensity fluctuations.

[0059] Based on the mid-term data set, the longitudinal dimension is checked to calculate the mid-term load fluctuation correlation function; the calculation logic of the mid-term load fluctuation correlation function is: Among them, C m (θ is the medium-term load fluctuation correlation function, θ0 and θ n is the start and end time of the medium-term load data, and η is the integral variable. Specifically, by integrating the medium-term load data and the light intensity data, the cumulative correlation between the load fluctuation and the light intensity is obtained, which is used to quantify the correlation between the grid load and the light intensity fluctuation on the medium-term time scale. By calculating the medium-term load fluctuation correlation function, the correlation between the load change and the light intensity change can be quantified, providing auxiliary support for load forecasting with environmental data.

[0060] The present invention is further configured to perform horizontal and vertical multi-dimensional verification based on a long-term data set to analyze long-term load characteristics, including:

[0061] Based on the long-term data set, the horizontal dimension calibration is performed to calculate the environmental deviation function of the long-term load; the present invention is further configured that the calculation logic of the environmental deviation function of the long-term load is: Among them, D l (G l ) is the environmental deviation function of the long-term load, φ0 and φ n is the start and end time of long-term load data, Pl (φ) is the grid load at time φ, G l (φ) is the light intensity at time φ, G avg is the long-term average light intensity; specifically, by l (φ) Deviation of light intensity from the average light intensity G l (φ) and multiply by 1+|G l (φ) is normalized, and the formula accumulates the difference between load and light over a long period of time to quantify the long-term response characteristics of the load to light intensity fluctuations. By calculating the long-term environmental deviation function, the long-term response of the load under light intensity fluctuations can be quantified, providing a long-term basis for load forecasting based on environmental changes.

[0062] Based on the long-term data set, the longitudinal dimension is checked and the long-term load increment accumulation function is calculated. The calculation logic of the long-term load increment accumulation function is: Among them, J l (φ) is the cumulative function of long-term load increment, J is the total number of long-term data points, φ j is the timestamp corresponding to the j-th long-term data point, P l (φ j ) is the time at which j The grid load at that time, G l (φ j ) is the time at which j Specifically, by calculating the grid load P at each time point l (φ j ) multiplied by ln(1+G l (φ j The accumulation of the load increment function can be used to quantitatively describe the growth of the grid load over a long period of time, providing an important reference for the long-term prediction of the power load.

[0063] The present invention is further configured to construct a multi-dimensional load forecasting model according to short-term load characteristics, medium-term load characteristics and long-term load characteristics. The construction logic of the multi-dimensional load forecasting model is:

[0064] Where Q(t) is the load forecast at time t+1, τ t and τ t-1 is the timestamp corresponding to the current moment and the previous moment in the short-term data set, J l (φ t ) and J l (φt-1 ) is the cumulative function value of the long-term load increment corresponding to the current moment and the previous moment in the long-term data set, λ1, λ2 and λ3 are weight coefficients, all greater than 0, and λ1+λ2+λ3=1; specifically, based on the short-term load characteristics, the medium-term load characteristics and the long-term load characteristics, the characteristics of different time scales are weighted and combined to more accurately predict the dynamic changes of the load. The formula calculates the load forecast value by multiplying the characteristics of each time scale (short-term characteristics, medium-term characteristics, long-term characteristics) with their corresponding weight coefficients λ1, λ2, λ3, and then summing up the results. This multi-dimensional load forecasting model comprehensively considers the impact of different time scales on the load by combining short-term, medium-term and long-term characteristics, so that the forecasting model can more accurately reflect the load change trend.

[0065] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0066] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0067] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0068] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean 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 the present application.

[0069] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example 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 performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0070] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0071] In the several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0072] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0073] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0074] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0075] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification, characterized in that: include: Collecting historical load data and environmental data of the power grid, and dividing the historical load data into short-term, medium-term and long-term load data sets according to multiple time scales; At each time scale, load characteristics analysis is performed based on the horizontal and vertical multi-dimensional verification method to obtain the load variation law at each time scale; According to the analysis results of load characteristics at each time scale, a multi-dimensional load forecasting model is constructed to predict future loads.

2. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 1 is characterized in that: The historical load data includes a load timestamp and a grid load; the environmental data includes a light timestamp and light intensity, wherein the light intensity is a normalized value.

3. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 2 is characterized in that: Perform time pairing according to the load timestamp and the light timestamp to obtain the grid load and light intensity in the time series; Divide the time series according to the time scale to obtain short-term, medium-term and long-term data sets; Perform horizontal and vertical multi-dimensional verification based on short-term data sets to analyze short-term load characteristics; Conduct horizontal and vertical multi-dimensional verification based on the mid-term data set to analyze mid-term load characteristics; Conduct horizontal and vertical multi-dimensional verification based on long-term data sets to analyze long-term load characteristics; According to the short-term load characteristics, medium-term load characteristics and long-term load characteristics, a multi-dimensional load forecasting model is constructed to predict future loads.

4. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 3 is characterized in that: Based on the short-term data set, multi-dimensional verification is performed horizontally and vertically to analyze the short-term load characteristics, including: Based on the short-term data set, the horizontal dimension is checked and the response function of short-term load change to light intensity is calculated; The longitudinal dimension is calibrated based on the short-term data set and the short-term load change rate function is calculated.

5. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 4 is characterized in that: The calculation logic of the response function of the short-term load change to the light intensity is: Among them, R s (G s ) is the response function of short-term load change to light intensity, τ0 and τ n is the start and end time of the short-term load data, P s (τ) is the grid load at time τ, G s (τ) is the light intensity at time τ, α is the response coefficient, and δ is the regularization constant; The calculation logic of the short-term load change rate function is: Among them, V s (τ) is the short-term load change rate at time τ.

6. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 3 is characterized in that: Based on the mid-term data set, multi-dimensional verification is performed in horizontal and vertical directions to analyze the mid-term load characteristics, including: Perform lateral dimension calibration based on the mid-term data set and calculate the harmonic response function of mid-term load changes; The longitudinal dimension is calibrated based on the medium-term data set, and the medium-term load fluctuation correlation function is calculated.

7. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 6 is characterized in that: The calculation logic of the harmonic response function of the mid-term load change is: Among them, H m (G m ) is the harmonic response function of the mid-term load change, K is the total number of mid-term data points, θ k is the timestamp corresponding to the kth mid-term data point, P m (θ k ) is the time at time θ k The grid load at that time, G m (θ k ) is the time at time θ k The light intensity at that time; The calculation logic of the medium-term load fluctuation correlation function is: Among them, C m (θ) is the medium-term load fluctuation correlation function, θ0 and θ n are the start and end time of the mid-term load data, and η is the integral variable.

8. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 3 is characterized in that: Based on the long-term data set, multi-dimensional verification is carried out horizontally and vertically to analyze the long-term load characteristics, including: Perform lateral dimension calibration based on long-term data sets and calculate environmental deviation functions for long-term loads; The longitudinal dimension is calibrated based on the long-term data set and the long-term load increment cumulative function is calculated.

9. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to claim 8 is characterized in that: The calculation logic of the environmental deviation function of the long-term load is: Among them, D l( G l) is the environmental deviation function of the long-term load, φ0 and φ n is the start and end time of long-term load data, P l (φ) is the grid load at time φ, G l (φ) is the light intensity at time φ, G avg is the long-term average light intensity; The calculation logic of the long-term load increment accumulation function is: Among them, J l (φ) is the cumulative function of long-term load increment, J is the total number of long-term data points, φ j is the timestamp corresponding to the j-th long-term data point, P l (φ j ) is the time at which j The grid load at that time, G l (φ j ) is the time at which j The light intensity at that time.

10. The three-dimensional multi-time scale load forecasting method based on horizontal and vertical multi-dimensional verification according to any one of claims 1 to 9, characterized in that: According to the short-term load characteristics, medium-term load characteristics and long-term load characteristics, a multi-dimensional load forecasting model is constructed. The construction logic of the multi-dimensional load forecasting model is: Where Q(t) is the load forecast at time t+1, τ t and τ t-1 is the timestamp corresponding to the current moment and the previous moment in the short-term data set, J l (φ t ) and J l (φ t-1 ) is the cumulative function value of the long-term load increment corresponding to the current moment and the previous moment in the long-term data set, λ1, λ2 and λ3 are weight coefficients, all greater than 0, and λ1+λ2+λ3=1.

Images