Combined user missing electricity fitting method and system based on typical load curve

Abstract

The present disclosure belongs to the technical field of power marketing, and particularly relates to a combined user missing electric quantity fitting method and system based on a typical load curve, which comprises the following steps: obtaining user power consumption information; obtaining a user power consumption typical load curve according to the obtained user power consumption information; judging the number of missing electric quantities in the obtained typical load curve; selecting a missing data fitting method according to the obtained number of missing electric quantities to obtain a fitting proportion coefficient; and completing fitting of combined user missing electric quantities according to the obtained fitting proportion coefficient.

Description

Technical Field

[0001] This disclosure belongs to the field of power marketing technology, specifically involving a combined user missing power fitting method and system based on typical load curves. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] With the deepening of power market reform and the gradual implementation of spot trading, the coordinated development of long-term and short-term spot trading has placed higher demands on the settlement capabilities of power market transactions.

[0004] Different types of transactions have different settlement requirements, and short-term transactions place significantly higher demands on the time accuracy of metering data. Especially with the continuous operation of the spot market, transaction types with minute-level time scales, such as demand response and ancillary services markets, will gradually be launched. While enriching the types of transactions, the types and details of user settlement bills will become increasingly numerous and detailed. In this context, the hourly electricity data currently provided by most users' electricity meters is not only the accurate basis for settlement of hourly transaction types but also a crucial foundation for forming settlement data for even smaller time scales. The adverse effects of missing data cannot be ignored. Since errors and malfunctions in metering devices or data communication systems are inevitable, the occurrence of missing user electricity data cannot be completely eliminated. Therefore, a high-precision fitting method for missing electricity data urgently needs to be established.

[0005] According to the inventor, current methods for fitting user electricity consumption data in the electricity market mainly involve linear interpolation or fitting the data using the average of historical data from the same period at points where data is missing, as stipulated in the electricity spot market trading rules. Because neither of these methods fully considers the patterns of user electricity consumption, the fitted data obtained using current rules tends to be inaccurate and fails to meet the high standards required for market development.

[0006] Electricity consumption by users follows certain patterns over time, exhibiting both a longitudinal correlation between the overall load on different days and a horizontal correlation where local differences arise based on the basic daily load patterns. Current methods for fitting electricity market user electricity consumption data primarily use linear interpolation and similar daily average imputation, which uses the average of historical data from the same period to fit missing data points. Linear interpolation mainly considers the horizontal correlation of daily load characteristics, while similar daily average imputation mainly considers the periodic patterns of user electricity consumption. A method for fitting missing electricity consumption data that comprehensively considers both aspects has not yet been established.

[0007] While existing studies have noted that user electricity load exhibits both periodic longitudinal correlations and horizontal correlations relative to the basic daily load, a scientific quantitative representation of these longitudinal and horizontal correlations has not been established. Some literature utilizes characteristic curves of similar users and adjusts outlier data by scaling them proportionally, considering the similarity between the curve to be fitted and the characteristic curve, but it does not address the differences between the curve to be fitted and the characteristic curve. Compressed sensing theory can be used to supplement missing electricity consumption and shows good fitting performance when there are few sampling points for missing electricity consumption data, but the method is complex, lacks transparency, and is difficult to adapt well to engineering needs. Summary of the Invention

[0008] To address the aforementioned issues, this disclosure proposes a combined method and system for fitting missing user electricity consumption based on typical load curves. This method utilizes typical load curves from historically similar days, comprehensively considers the overall and local correlations of user electricity consumption data, and clarifies weighting coefficients for various scenarios. It then generates missing data in a combined manner, effectively improving the accuracy of user electricity consumption data fitting and promoting the rapid development of the electricity market.

[0009] According to some embodiments, the first solution of this disclosure provides a combined user missing power fitting method based on typical load curves, employing the following technical solution:

[0010] A combined user missing power fitting method based on typical load curves includes:

[0011] Obtain user electricity consumption information;

[0012] Based on the obtained user electricity consumption information, a typical user electricity load curve is obtained;

[0013] Determine the number of missing electrical quantities in the obtained typical load curve;

[0014] Based on the number of missing electricity values, select the fitting method for the missing data and obtain the fitting ratio coefficient;

[0015] Based on the obtained fitting ratio coefficients, the fitting of the combined user missing power is completed.

[0016] As a further technical limitation, the acquired user electricity consumption information includes the user's longitudinal time attributes and the user's horizontal correlation attributes.

[0017] Furthermore, based on the obtained user electricity consumption information, the user's longitudinal time attributes are meticulously divided to determine the historical similar days for various types of electricity consumption information; the obtained historical similar days are then tagged, and electricity consumption tags are extracted.

[0018] Furthermore, in the process of obtaining the typical load curve of user electricity consumption, based on the determined historical similar days and historical electricity consumption data, the typical load curve of historical similar days is constructed. Specifically, the electricity consumption data is initially classified according to the historical similar days to which it belongs, and electricity consumption tags are obtained; based on the obtained electricity consumption tags, the classified user electricity consumption data is normalized to obtain normalized data; based on the obtained normalized data, the daily load curves of different historical similar days are clustered by combining clustering algorithms and the number of clusters, and the cluster center of each cluster is the typical load curve.

[0019] As a further technical limitation, in the process of obtaining the fitting ratio coefficient, the number of missing power is determined based on the typical load curve, and the fitting analysis is performed based on the number of missing power to determine the fitting ratio coefficient and fitting parameters.

[0020] As a further technical limitation, considering the pattern between the daily load curve of the data to be fitted and the typical load curve of similar historical days, the missing electricity data is fitted by the relative load change rate; the relative load change rate is the relative change rate between the daily load curve of the data to be fitted and the corresponding time point data of the typical load curve of similar historical days.

[0021] As a further technical limitation, the fitted data is adjusted by a scaling factor β; if the data for 24:00 on a given day is missing, the fitted data is not adjusted, and β = 1; if the data for 24:00 on a given day is available, the total electricity consumption for that day is known, and β is: Where P represents the cumulative daily electricity consumption data at 24:00. x is the sum of the data available on the daily load curve to which the fitted data belongs. j This represents user load data, where j represents a time point, and M ranges from 0 to 23. x is the sum of the missing data obtained from the fitting. i This represents the fitted user load data, where i represents the time point, and N ranges from 0 to 23.

[0022] According to some embodiments, the second aspect of this disclosure provides a combined user missing power fitting system based on typical load curves, employing the following technical solution:

[0023] A combined user missing power fitting system based on typical load curves includes:

[0024] The acquisition module is configured to acquire user electricity consumption information;

[0025] The judgment module is configured to obtain a typical load curve of user electricity consumption based on the acquired user electricity consumption information; and to determine the number of missing electricity quantities in the obtained typical load curve.

[0026] The fitting module is configured to select a fitting method for the missing data based on the number of missing power values ​​obtained, and obtain a fitting ratio coefficient; based on the obtained fitting ratio coefficient, it completes the fitting of the combined user missing power values.

[0027] According to some embodiments, a third aspect of this disclosure provides a computer-readable storage medium, employing the following technical solution:

[0028] A computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the steps of the combined user missing power fitting method based on typical load curves as described in the first aspect of this disclosure.

[0029] According to some embodiments, the fourth solution of this disclosure provides an electronic device that adopts the following technical solution:

[0030] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the steps of the combined user missing power fitting method based on a typical load curve as described in the first aspect of this disclosure.

[0031] Compared with the prior art, the beneficial effects of this disclosure are as follows:

[0032] This disclosure comprehensively considers the correlation between user electricity consumption data both vertically and horizontally, establishing a quantifiable method for fitting missing data, effectively improving the accuracy of user electricity consumption data fitting. After establishing typical load curves for historically similar days, the load curve of the day to which the data to be fitted belongs is compared with it. Different values ​​of weighting coefficients correspond to the utilization of overall or partial information provided by the typical load curves of historically similar days under different scenarios, and the fitted data is generated after necessary scaling. The method has high applicability, and a single formula can be used to unify the fitting of electricity consumption data for multiple scenarios. In different cases, only the corresponding weighting coefficients and typical load curves of historically similar days need to be selected to complete the fitting of missing electricity consumption. The introduction of fitting scaling coefficients ensures the accuracy of the total daily electricity consumption after fitting. Attached Figure Description

[0033] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0034] Figure 1 This is a flowchart of the combined user missing power fitting method based on typical load curves in Embodiment 1 of this disclosure;

[0035] Figure 2This is a schematic diagram illustrating the specific steps of the combined user missing power fitting method based on typical load curves in Embodiment 1 of this disclosure;

[0036] Figure 3 This is a schematic diagram of the relative rate of change δ in Embodiment 1 of this disclosure;

[0037] Figure 4 It is δ in Embodiment 1 of this disclosure m and δ n A diagram illustrating the same number;

[0038] Figure 5 This is a schematic diagram of the case where the number of consecutive defects is greater than or equal to 2 and contains 24-point table codes in Embodiment 1 of this disclosure;

[0039] Figure 6 This is a structural block diagram of the combined user missing power fitting system based on typical load curves in Embodiment 2 of this disclosure. Detailed Implementation

[0040] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0041] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0042] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0043] Where there is no conflict, the embodiments and features described herein can be combined with each other.

[0044] Example 1

[0045] Embodiment 1 of this disclosure introduces a combined user missing power fitting method based on typical load curves.

[0046] like Figure 1 and Figure 2 The method shown is a combined user missing power fitting method based on typical load curves, including:

[0047] Obtain user electricity consumption information;

[0048] Based on the obtained user electricity consumption information, a typical user electricity load curve is obtained;

[0049] Determine the number of missing electrical quantities in the obtained typical load curve;

[0050] Based on the number of missing electricity values, select the fitting method for the missing data and obtain the fitting ratio coefficient;

[0051] Based on the obtained fitting ratio coefficients, the fitting of the combined user missing power is completed.

[0052] User electricity consumption exhibits both periodic patterns and is subject to random fluctuations in certain areas due to uncertainties. Therefore, the fitting method for user electricity consumption data needs to take both of these factors into account.

[0053] The periodic patterns of user electricity consumption exhibit repetition at certain time cycles, and can therefore be referred to as the user's longitudinal time attribute. Referring to existing electricity market rules and related research, the user's longitudinal time attribute represents the pattern of their electricity load being affected by factors such as working conditions and seasonal changes. Therefore, to improve the accuracy of the longitudinal time attribute, it can generally be divided according to weekdays, weekends, and national statutory holidays [holidays are divided into short holidays (New Year's Day, May Day, Qingming Festival, etc.) and long holidays (Spring Festival, National Day)]. In other words, user electricity consumption data can be subdivided into multiple categories based on time attributes.

[0054] The random fluctuations in user electricity consumption have two characteristics: first, the fluctuations conform to a normal distribution with the typical load curve as the mean; second, the consistency of fluctuations is higher over shorter time scales, meaning the data has a certain degree of horizontal correlation. Therefore, considering only the longitudinal time attribute and fitting missing user electricity consumption data based solely on historical similar day data is unlikely to achieve satisfactory results.

[0055] If cluster analysis is performed on user data under each longitudinal time attribute, and the typical load curve of the user under that time attribute is obtained, further combined data fitting by combining the horizontal correlation between the user's current load data and the typical load curve will effectively improve the accuracy of data fitting.

[0056] Using historical similar day information as a longitudinal correlation reference when electricity consumption data is missing, and using typical load curves as a lateral correlation reference when electricity consumption data is missing, an electricity consumption data fitting formula is established, as shown in equation (1):

[0057]

[0058] Where, x i This represents the fitted user load data, where i represents the time point and its value ranges from 0 to 23; x′ iThis represents the value of the typical load curve for historically similar days at time point i. Represents x′ i The normalized value; m and n are the time points corresponding to the known data immediately before and after the missing data area. The difference between n and m is the duration of the missing data interval in terms of time periods. α1 and α2 represent the weighting coefficients of the combined electricity fitting method based on typical load curves of historical similar days. The α1 coefficient represents that the data fitting needs to comprehensively consider the vertical and horizontal correlation between the load curve of the day to which the data to be fitted belongs and the typical curve, both as a whole and in the local context. The α2 coefficient represents that the data fitting only needs to consider the horizontal correlation of the load curve of the day to which the data to be fitted belongs. The specific values ​​are determined according to different scenarios.

[0059] A unified expression for fitting missing electricity data can be established. For different situations, only the corresponding weighting coefficients and typical load curves need to be selected to fit the missing electricity. In addition, considering that the goal of data fitting is to provide accurate settlement of electricity trading, in order to ensure that the cumulative electricity of the daily load curve after fitting is consistent with the actual situation, this embodiment defines a fitting scaling factor β to further adjust the fitted data, as shown in equation (2).

[0060]

[0061] Among them, X i This is the final fitting result for the missing data.

[0062] Establishment of typical load curves for historically similar days

[0063] To fit the missing electricity consumption data for users, a reference basis is needed. For continuously metered user electricity consumption data, the data at different time points are not independent; they can form a daily load curve for the user. The daily load curve reflects the overall periodic pattern of user electricity consumption, as well as its local variations relative to the basic pattern. Therefore, the reference basis for data fitting should also possess both of these characteristics. This invention proposes using the typical load curve of historically similar days as a reference. Establishing the typical load curve of historically similar days based on the user's historical electricity consumption data can be mainly divided into three steps:

[0064] First, the data is initially classified based on the historical similar dates to which it belongs.

[0065] Secondly, in order to eliminate the influence of the magnitude of the data and to cluster curves with similar shapes into one class, the obtained data is normalized. The traditional normalization formula is shown in equation (3).

[0066]

[0067] Where, x iThis represents the raw data at time point i on the daily load curve, max(x i This represents the maximum value in the raw daily load data. This represents the normalized data at time point i on the daily load curve. Since this invention is designed for fitting missing data, the maximum value in the data sequence cannot be determined when some data is missing; therefore, the normalization method is improved.

[0068] If 24-hour load data exists, the normalization formula is as shown in equation (4):

[0069]

[0070] Among them, avg(x i () represents the average value of the daily load curve at each time point.

[0071] If the 24-hour load data is unavailable, the normalization formula is as shown in equation (5):

[0072]

[0073] Where, x i This represents the raw data at time point i on the daily load curve, pavg(x) i () represents the average value of known data for the daily load curve. This represents the normalized data at time point i on the daily load curve.

[0074] Before clustering the original data, normalization is performed using either equation (4) or equation (5) depending on whether the data sequence is complete (when the data sequence is complete, equation (4) and equation (5) are equivalent).

[0075] Finally, a clustering algorithm was used to cluster the daily load curves of different historically similar days, and the cluster center of each cluster was regarded as the typical load curve of that class.

[0076] Determination of fitting parameters

[0077] Using the fitting method shown in equation (6), α1 and α2 are the weighting coefficients to be determined:

[0078]

[0079] To determine the values ​​of α1 and α2, the cases of missing data are categorized as follows:

[0080] (1) When data for one time point (not 00:00) is missing, the missing data for a single point does not affect the overall characteristics of the daily load. It is not necessary to supplement the overall information based on the historical similar day patterns. Only the horizontal correlation of the local information of the daily load curve needs to be considered. In this case, α1 = 0 and α2 = 0.5 are determined. That is, the missing data is determined by taking the average of the data before and after the missing time point.

[0081] (2) When 2 to 12 consecutive time points of data are missing (the number of consecutive missing data points is no more than 1 / 2 of the entire data sequence), as the number of missing data increases, it is difficult to accurately restore the overall situation of daily load based on local information alone. It is necessary to supplement it with the historical similar day patterns. At this time, the corresponding typical load curve can be matched according to the historical similar day labels of the user's current date, and the values ​​of α1 and α2 can be determined by comparing the curve to be fitted with the selected typical load curve.

[0082] Since α2 is a correlation coefficient that only considers the local information between the missing data to be fitted and the data at the previous and subsequent time points, when two or more data points are missing consecutively, the direct contribution of the data at the previous and subsequent time points to the fitting of the missing data becomes weaker as the number of missing data points increases. Therefore, we set α2 = 0 and use α1, a coefficient that comprehensively considers the overall correlation between the daily load curve of the data to be fitted and the typical load curve of similar historical days, to achieve the fitting.

[0083] Considering that the daily load curve of the data to be fitted has the same overall pattern but different local values ​​compared with the typical load curve of similar historical days, this invention defines the relative load change rate δ, which is the relative change rate between the daily load curve of the data to be fitted and the typical load curve of similar historical days at the corresponding time point. The specific formula is shown in Equation (7).

[0084]

[0085] Among them, A i Let D be the load value corresponding to time point i of the daily load curve to which the data to be fitted belongs. i This represents the load value at point i on a historically similar day's typical load curve. Figure 3 Taking the two curves in the figure as examples, we will further explain the relative rate of change δ.

[0086] according to Figure 3 We can obtain,

[0087]

[0088]

[0089] There are generally two scenarios regarding the relative rate of change between the daily load curve of the data to be fitted and the typical load curve of similar historical days: the two values ​​have opposite signs (one positive and one negative), such as... Figure 3 As shown, the values ​​of both have the same sign (both negative or both positive), such as... Figure 4 As shown.

[0090] Since the typical load curve is generated by clustering historical load data of users and represents the typical situation of user load, the actual daily load curve fluctuates around this typical load curve. Without sacrificing accuracy, and to facilitate the engineering implementation of the algorithm, a linear function is chosen to calculate the relative rate of change of load δ at each time point within the missing data interval. The formula is derived based on the known data at two time points before and after the missing data interval.

[0091]

[0092] The formula shown in equation (10) is obtained:

[0093]

[0094] By substituting the parameter values ​​under different conditions, δ can be determined. i The value of .

[0095] According to equation (11), we obtain α1 and δ i The relationship between them is shown in equation (12):

[0096] α1=δ i +1 (12)

[0097] The value of α1 is determined by the above steps. Substituting it into equation (1), the missing data can be fitted according to equation (1) in the corresponding case.

[0098] (3) When the number of consecutive missing data points is greater than or equal to two and includes data from the last hour of the day (24:00), only one δ value can be calculated, such as... Figure 5 As shown.

[0099] At this point, α2 = 0, and α1 is determined as shown in formula (13):

[0100]

[0101] Where, ∑δ i R is the sum of all relative rates of change that can be obtained from this daily load curve, where R is the number of relative rates of change that can be obtained.

[0102] (4) When all day data is missing, the typical load curve of a historical similar day with the corresponding time attribute label is used for fitting. Since all data for the day is missing, the formula in (1) becomes formula (14):

[0103]

[0104] At this point, the value of 1 is determined as shown in formula (15):

[0105]

[0106] in, This represents the average of all data at that point in the current or previous month that share the same data series label. This represents the normalized value of the typical load curve for historically similar days under this label sequence at time point i.

[0107] Determining the fitting scaling factor

[0108] To ensure that the cumulative electricity consumption of users after fitting the missing data is consistent with the actual situation, this embodiment defines a fitting scaling factor β to further adjust the fitted data. If the data at 24:00 on a given day is missing, no adjustment is made since the total electricity consumption for that day cannot be determined, and β is set to 1 in this case; if the data at 24:00 on a given day is available, the total electricity consumption for that day is known, and β is defined according to formula (16):

[0109]

[0110] Where P represents the cumulative daily electricity consumption data at 24:00. The sum of data available on the daily load curve to which the fitted data belongs. This is the sum of the missing data obtained by fitting according to equation (1). The missing data obtained by fitting according to equation (1) is further adjusted using β, as shown in equation (17):

[0111] X i =β·x i (17)

[0112] Among them, X i This represents the final fitting result for the missing data.

[0113] The values ​​of α1 and α2 are determined for different scenarios. The complete representation of the fitting with missing data is shown in Equation (16).

[0114]

[0115] This embodiment addresses the issue of low accuracy in fitting existing linear interpolation and similar daily average fitting methods when the number of consecutive missing daily load data points exceeds two, reaching half of the entire data sequence. It proposes a combined missing data fitting method based on historical similar daily typical load curves. After establishing the historical similar daily typical load curves, the daily load curve of the data to be fitted is compared with them. Different weighting coefficients correspond to different ways of utilizing the overall or partial information provided by the historical similar daily typical load curves in different scenarios, and the fitted data is generated after necessary scaling. The method has high applicability; a single formula can unify the fitting of electricity data for various scenarios. In different cases, only the corresponding weighting coefficients and historical similar daily typical load curves need to be selected to complete the fitting of missing electricity consumption. The introduction of fitting scaling coefficients ensures the accuracy of the total daily electricity consumption after fitting.

[0116] Example 2

[0117] Embodiment 2 of this disclosure introduces a combined user missing power fitting system based on a typical load curve.

[0118] like Figure 6 The system shown is a combined user missing power fitting system based on a typical load curve, comprising:

[0119] The acquisition module is configured to acquire user electricity consumption information;

[0120] The judgment module is configured to obtain a typical load curve of user electricity consumption based on the acquired user electricity consumption information; and to determine the number of missing electricity quantities in the obtained typical load curve.

[0121] The fitting module is configured to select a fitting method for the missing data based on the number of missing power values ​​obtained, and obtain a fitting ratio coefficient; based on the obtained fitting ratio coefficient, it completes the fitting of the combined user missing power values.

[0122] The detailed steps are the same as those of the combined user missing power fitting method based on typical load curves provided in Example 1, and will not be repeated here.

[0123] Example 3

[0124] Embodiment 3 of this disclosure provides a computer-readable storage medium.

[0125] A computer-readable storage medium having a program stored thereon that, when executed by a processor, implements the steps of the combined user missing power fitting method based on a typical load curve as described in Embodiment 1 of this disclosure.

[0126] The detailed steps are the same as those of the combined user missing power fitting method based on typical load curves provided in Example 1, and will not be repeated here.

[0127] Example 4

[0128] Embodiment 4 of this disclosure provides an electronic device.

[0129] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps in the combined user missing power fitting method based on a typical load curve as described in Embodiment 1 of this disclosure.

[0130] The detailed steps are the same as those of the combined user missing power fitting method based on typical load curves provided in Example 1, and will not be repeated here.

[0131] The above description is merely a preferred embodiment of this disclosure and is not intended to limit this disclosure. Various modifications and variations can be made to this disclosure by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

[0132] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A combined user missing power fitting method based on typical load curves, characterized in that, include: Obtain user electricity consumption information, including user longitudinal time attributes and user horizontal correlation attributes; Based on the obtained user electricity consumption information, a typical user electricity load curve is obtained; Determine the number of missing electrical quantities in the obtained typical load curve; Based on the number of missing electricity values, select the fitting method for the missing data and obtain the fitting ratio coefficient; Based on the obtained fitting ratio coefficients, complete the fitting of the combined user missing power. In obtaining the typical load curve of user electricity consumption, based on the determined historical similar days and historical electricity consumption data, the typical load curve of historical similar days is constructed. Specifically, the electricity consumption data is initially classified according to the historical similar days to which it belongs, and electricity consumption tags are obtained; based on the obtained electricity consumption tags, the classified user electricity consumption data is normalized to obtain normalized data; based on the obtained normalized data, the daily load curves of different historical similar days are clustered by combining clustering algorithms and the number of clusters, and the cluster center of each cluster is the typical load curve. Considering the relationship between the daily load curve of the data to be fitted and the typical load curve of similar historical days, the missing electricity data is fitted by the relative load change rate; the relative load change rate is the relative change rate between the daily load curve of the data to be fitted and the corresponding time point data of the typical load curve of similar historical days. Using historical similar day information as a longitudinal correlation reference when electricity consumption data is missing, and using typical load curves as a lateral correlation reference when electricity consumption data is missing, a formula for fitting electricity consumption data is established: in, This represents the fitted user load data. i This represents a point in time, with values ​​ranging from 0 to 23. This indicates the typical load curve for historically similar days at time points. i The value at that location, express The normalized value of m; n The time points corresponding to the known data immediately before and after the missing data area; n and m The difference represents the duration of the missing data interval, expressed as a time period. and This represents the weighting coefficients of a combined electricity consumption fitting method based on typical load curves of historically similar days. The coefficients represent the longitudinal and lateral correlation between the daily load curve of the data to be fitted and the typical curve, which needs to be considered comprehensively from both the overall and local perspectives when fitting the data. The coefficients represent that data fitting only needs to consider the horizontal correlation of the daily load curve to which the data to be fitted belongs; By fitting the scaling factor Adjust the fitted data; if the data for 24:00 on that day is missing, then do not adjust the fitted data. If the data for 24:00 on that day is available, then the total electricity consumption for that day is known. for: ;in, This is the cumulative electricity consumption data for the day at 24:

00. The sum of data available on the daily load curve to which the fitted data belongs. This represents user load data, where j represents a time point, and M ranges from 0 to 23. The sum of missing data obtained from the fitting is N, which ranges from 0 to 23. Using the fitting scaling factor Further adjustments were made to the fitted data: ;in, This represents the final fitting result for the missing data.

2. The combined user missing power fitting method based on typical load curves as described in claim 1, characterized in that, Based on the obtained user electricity consumption information, the user's longitudinal time attributes are meticulously divided to determine the historical similar days for various types of electricity consumption information; the obtained historical similar days are then tagged, and electricity consumption tags are extracted.

3. The combined user missing power fitting method based on typical load curves as described in claim 1, characterized in that, In the process of obtaining the fitting proportional coefficient, the number of missing power is determined based on the typical load curve, and the fitting analysis is performed based on the number of missing power to determine the fitting proportional coefficient and fitting parameters.

4. A combined user missing power fitting system based on typical load curves, employing the combined user missing power fitting method based on typical load curves as described in any one of claims 1-3, characterized in that, include: The acquisition module is configured to acquire user electricity consumption information, including user longitudinal time attributes and user horizontal correlation attributes. The judgment module is configured to obtain a typical load curve of user electricity consumption based on the acquired user electricity consumption information; and to determine the number of missing electricity quantities in the obtained typical load curve. The fitting module is configured to select a fitting method for the missing data based on the number of missing power values ​​obtained, and obtain a fitting ratio coefficient; based on the obtained fitting ratio coefficient, it completes the fitting of the combined user missing power values.

5. A computer-readable storage medium having a program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the combined user missing power fitting method based on typical load curves as described in any one of claims 1-3.

6. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the combined user missing power fitting method based on typical load curves as described in any one of claims 1-3.

Images