A method and medium for predicting electric quantity based on component decomposition and correction

Abstract

The present invention relates to a method and medium for predicting electric quantity based on component decomposition and correction, and relates to the field of electric quantity prediction. The method comprises: constructing a daily electric quantity data decomposition model; correcting the daily average temperature of historical days according to an average temperature correction model; obtaining a trend electric quantity prediction value and a period electric quantity prediction value through a trend electric quantity model and a period electric quantity model; obtaining the meteorological electric quantity of historical days through a meteorological electric quantity model, performing segmented fitting on the meteorological electric quantity of historical days, and obtaining a meteorological electric quantity curve; correcting the average temperature of a day to be predicted according to an average temperature correction model; predicting the meteorological electric quantity according to the daily average temperature to be predicted and the meteorological electric quantity curve; and obtaining an electric quantity prediction value according to the trend electric quantity, the period electric quantity and the meteorological electric quantity. The present invention fully considers the characteristics of electric quantity per day, the limit temperature, the maximum cumulative days and the cumulative effect coefficient and other factors of electric quantity per day through the electric quantity per day decomposition model and the temperature correction model, and improves the accuracy of electric quantity prediction.

Description

Technical Field

[0001] The invention relates to an electric quantity prediction method and medium based on component decomposition and correction, belonging to the technical field of electric power demand analysis. Background Art

[0002] Electricity consumption is the "barometer" and "weather vane" of the national economy. Electricity consumption forecasting plays an important role in the operation and management of the power system. On the one hand, it can help the power system to reasonably arrange power generation plans and peak-shaving capabilities, and improve the stability and reliability of the power system; on the other hand, it can provide users with scientific and reasonable power services to ensure the quality and efficiency of power supply.

[0003] The research on electricity consumption forecasting is very mature, and there are a large number of forecasting methods in both industry and academia. Traditional time series methods include ARIMA, ARMA models, etc., machine learning methods include random forest, GBDT, etc., and deep neural networks include CNN, LSTM, etc.

[0004] Since summer load is significantly affected by temperature factors, continuous high temperature weather will trigger the accumulated temperature effect, resulting in a certain degree of abnormal load growth. The existing short-term load forecasting method does not take this into account and therefore cannot meet actual needs. Summary of the invention

[0005] In order to overcome the above problems, the present invention provides a method and medium for predicting electric quantity based on component decomposition and correction.

[0006] The technical solution of the present invention is as follows:

[0007] In a first aspect, a method for predicting power based on component decomposition and correction comprises the following steps:

[0008] Get daily electricity data and daily average temperature for several historical days;

[0009] Constructing a daily electricity data decomposition model, wherein the daily electricity data decomposition model includes a trend electricity model, a cycle electricity model and a meteorological electricity model;

[0010] Constructing a mean temperature correction model;

[0011] Correct the historical daily average temperature according to the average temperature correction model to obtain the corrected historical daily average temperature;

[0012] Draw the trend power curve and the cycle power curve respectively through the trend power model and the cycle power model;

[0013] The trend power forecast value of the day to be predicted is obtained according to the trend power curve, and the cycle power forecast value of the day to be predicted is obtained according to the cycle power curve;

[0014] The meteorological electricity quantity of several historical days is obtained through the meteorological electricity quantity model, and the meteorological electricity quantity of the historical days is segmentedly fitted through the corrected historical daily average temperature to obtain a meteorological electricity quantity curve representing the change of meteorological electricity quantity with the corrected historical daily average temperature;

[0015] Obtain the average temperature of the day to be predicted, and correct the average temperature of the day to be predicted according to the average temperature correction model to obtain the corrected average temperature of the day to be predicted;

[0016] The meteorological power quantity on the forecast day is forecasted according to the corrected average temperature on the forecast day and the meteorological power quantity curve, so as to obtain the forecast value of meteorological power quantity;

[0017] The predicted value of electricity for the day to be predicted is obtained by summing the predicted value of trend electricity for the day to be predicted, the predicted value of periodic electricity for the day to be predicted and the predicted value of meteorological electricity for the day to be predicted.

[0018] Furthermore, the trend electricity model is constructed in the following way:

[0019] Select daily electricity data with the same meteorological factors and on the nth day of each week, exclude daily electricity data on statutory holidays, and use the linear regression model to simulate the natural growth trend of electricity. The formula is as follows:

[0020] ;

[0021] ;

[0022] in, x i For the i sky, y i For the i Daily electricity consumption, w , b is the fitting parameter, f ( x i ) is the i Fitted electricity consumption for the day.

[0023] Furthermore, the cycle power model is constructed in the following way:

[0024] For weekdays and weekends, the first time period is defined. If the day to be decomposed is the nth day of the week, the nth day of each week in the past first time period is taken as the first basic day;

[0025] Cycle power of the decomposition day L z as follows:

[0026] ;

[0027] in, N is the number of days in the first time period, and the number of days in the time period is greater than 7 days, L u For the u The daily electricity consumption of the first basic day, L qu For the u The trend electricity quantity of the first basic day is obtained through the trend electricity quantity model;

[0028] For statutory holidays, set a second time period and select the time before the statutory holidays. D 1 day to after statutory holidays D 2 days is the time period affected by the statutory holiday. If the day to be decomposed is the nth day of this month ' day, then the nth day of each month in the past second time period ' Day as the second base day;

[0029] Cycle power of the decomposition day L z' as follows:

[0030] ;

[0031] in, N' is the number of days in the second time period, and the number of days in the time period is greater than 31 days, L i ' For the i' The daily electricity consumption of the second basic day, L qi ' For the i' The trend electricity quantity of the second basic day is obtained through the trend electricity quantity model.

[0032] Furthermore, the meteorological electricity model is specifically:

[0033] Obtain meteorological power based on the power, trend power and cycle power of the day to be decomposed L t :

[0034] ;

[0035] in, L is the amount of electricity to be decomposed on the day, L q is the trend electricity of the day to be decomposed, L m for L z or L z' , Trend Power Lq The calculation formula is as follows:

[0036] .

[0037] Furthermore, the average temperature correction model is constructed as follows:

[0038] Establish a temperature correction formula that takes into account the accumulated temperature effect:

[0039] ;

[0040] in, T v ' The revised date v The average temperature, T v For date v The average temperature, T min is the limit temperature, T v-j For date v The previous chapter j The average temperature of the day, k vj is the cumulative effect coefficient and satisfies k v1> k v2>…> k vp , p=min ( o , d max ), o For date v The number of consecutive days with a temperature higher than the limit, d max The maximum cumulative number of days.

[0041] Further, the limit temperature and the maximum cumulative days in the temperature correction formula are determined in the following manner:

[0042] The temperature correction problem is formulated as the following optimization problem:

[0043] ;

[0044] in, RSS is the residual sum of squares, TSS is the total sum of squares, For the i The weather electricity of the day, is the average value of meteorological electricity, For the i The meteorological power forecast value obtained by piecewise fitting of the corrected temperature and meteorological power;

[0045] Solve the above optimization problem to obtain the limit temperature and maximum cumulative days corresponding to the optimal value, and substitute the limit temperature and maximum cumulative days corresponding to the optimal value into the temperature correction formula.

[0046] Furthermore, the optimization problem is solved by the COBYLA algorithm.

[0047] Further, the trend power forecast value of the day to be forecasted is obtained according to the trend power curve, and the cycle power forecast value of the day to be forecasted is obtained according to the cycle power curve, specifically:

[0048] According to the trend of power curve The trend electricity of the forecast day is predicted. x i Set to the number of days from the first data day to the day to be predicted, the first data day being the earliest day of the daily electricity data selected when constructing the trend electricity curve;

[0049] The predicted value of the periodic electricity quantity for the day to be predicted is obtained according to the periodic electricity curve. Specifically, it is determined whether the day to be predicted is a statutory holiday. If so, the average of the periodic electricity quantity for the day to be predicted on the same holiday as the day to be predicted is obtained from the periodic electricity curve as the periodic electricity quantity for the day to be predicted. Otherwise, the position of the day to be predicted in a week is obtained, and the average of the periodic electricity quantity for the day to be predicted on the same holiday as the day to be predicted in the week is used as the periodic electricity quantity for the day to be predicted.

[0050] Furthermore, the meteorological power of the day to be predicted is predicted according to the corrected average temperature and meteorological power curve of the day to be predicted, specifically:

[0051] Correcting the average temperature of the predicted day according to the temperature correction formula;

[0052] The corrected average temperature of the day to be predicted is substituted into the meteorological power curve to obtain the meteorological power of the day to be predicted.

[0053] In a second aspect, a computer-readable storage medium, when a computer program stored thereon is executed, implements the power prediction method based on component decomposition and correction as described in the first aspect.

[0054] The present invention has the following beneficial effects:

[0055] The present invention fully considers the characteristics of daily electricity consumption through the daily electricity consumption decomposition model, analyzes the various factors affecting the electricity consumption, quantifies the growth of economic and population electricity-consuming equipment as trend electricity consumption, quantifies factors such as weekdays, weekends, and statutory holidays as periodic electricity consumption, and thereby obtains the meteorological electricity consumption affected by meteorological factors.

[0056] The temperature correction model of meteorological electricity of the present invention comprehensively considers factors such as limit temperature, maximum cumulative number of days and cumulative effect coefficient, establishes a temperature correction model considering the accumulated temperature effect, and obtains the temperature correction model by converting the temperature correction problem into an optimization problem and solving the limit temperature and maximum cumulative number of days. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 The figure is a flow chart of a method embodiment of the present invention.

[0058] Figure 2 Schematic diagram of daily electricity consumption decomposition according to an embodiment of the present invention.

[0059] Figure 3 The error box plots of each method. DETAILED DESCRIPTION

[0060] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution of the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings of the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the described embodiments of the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0061] Unless otherwise defined, the technical terms or scientific terms used in the present disclosure should be understood by people with ordinary skills in the field to which the present disclosure belongs. The "first", "second" and similar words used in the present disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. In order to keep the following description of the embodiments of the present disclosure clear and concise, the present disclosure omits the detailed description of some known functions and known components.

[0062] The present disclosure is described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] First, reference Figure 1 , a method for predicting electric quantity based on component decomposition and correction, comprising the following steps:

[0064] Get daily electricity data and daily average temperature for several historical days;

[0065] Constructing a daily electricity data decomposition model, wherein the daily electricity data decomposition model includes a trend electricity model, a cycle electricity model and a meteorological electricity model;

[0066] Constructing a mean temperature correction model;

[0067] Correct the historical daily average temperature according to the average temperature correction model to obtain the corrected historical daily average temperature;

[0068] Draw the trend power curve and the cycle power curve respectively through the trend power model and the cycle power model;

[0069] The trend power forecast value of the day to be predicted is obtained according to the trend power curve, and the cycle power forecast value of the day to be predicted is obtained according to the cycle power curve;

[0070] The meteorological electricity quantity of several historical days is obtained through the meteorological electricity quantity model, and the meteorological electricity quantity of the historical days is segmentedly fitted through the corrected historical daily average temperature to obtain a meteorological electricity quantity curve representing the change of meteorological electricity quantity with the corrected historical daily average temperature;

[0071] Obtain the average temperature of the day to be predicted, and correct the average temperature of the day to be predicted according to the average temperature correction model to obtain the corrected average temperature of the day to be predicted;

[0072] The meteorological power quantity on the forecast day is forecasted according to the corrected average temperature on the forecast day and the meteorological power quantity curve, so as to obtain the forecast value of meteorological power quantity;

[0073] The predicted value of electricity for the day to be predicted is obtained by summing the predicted value of trend electricity for the day to be predicted, the predicted value of periodic electricity for the day to be predicted and the predicted value of meteorological electricity for the day to be predicted.

[0074] Daily electricity data is related to trend electricity, periodic electricity and meteorological electricity. Trend electricity is the electricity that grows naturally due to economic growth, population and electricity consumption, etc. Periodic electricity is the electricity affected by weekdays, weekends and statutory holidays. Weekends include Saturdays and Sundays, statutory holidays include Spring Festival, New Year's Day and other statutory holidays, and meteorological electricity is the electricity related to the daily average temperature. Dividing daily electricity data according to these three types of electricity and building corresponding models can accurately reflect and predict electricity.

[0075] In one embodiment of the present invention, the trend electricity model is constructed in the following manner:

[0076] Select daily electricity data with the same meteorological factors and on the nth day of each week, exclude daily electricity data on statutory holidays, and use the linear regression model to simulate the natural growth trend of electricity. The formula is as follows:

[0077] ;

[0078] ;

[0079] in, x i For the i sky, y i For the i Daily electricity consumption, w , b is the fitting parameter, f ( x i ) is the i Fitted electricity consumption for the day.

[0080] When selecting daily electricity data for building a trend electricity model, it is necessary to eliminate the influence of cycles and weather. Therefore, daily electricity data with the same meteorological factors are selected, and meteorological factors can be screened by two conditions: daily average temperature and meteorology. In a specific embodiment, meteorological conditions are divided into sunny days, cloudy days, and rainy days, and the daily average temperature is divided into 5-degree gears, such as 0-5°C (excluding 5°C) as a gear, 5-10°C (excluding 10°C) as a gear, and so on. The dates with the same meteorological conditions and the same daily average temperature are regarded as dates with the same meteorological factors.

[0081] When eliminating the conditions affected by periodic factors, the daily electricity consumption data of statutory holidays is not used as fitting data because the sample size of statutory holidays is small. Select the date at the same position in the week for fitting, such as Monday as the fitting data.

[0082] In a specific embodiment, it also includes fitting the data from the 1st to the 7th day of each week to obtain a trend electricity model, and then performing weighted averaging of the trend electricity models to obtain a weighted trend electricity model, wherein weighting is performed according to the amount of data.

[0083] In one embodiment of the present invention, the cycle power model is constructed in the following manner:

[0084] For weekdays and weekends, the first time period is defined. If the day to be decomposed is the nth day of the week, the nth day of each week in the past first time period is taken as the first basic day;

[0085] Cycle power of the decomposition day L z as follows:

[0086] ;

[0087] in,N is the number of days in the first time period, and the number of days in the time period is greater than 7 days, L u For the u The daily electricity consumption of the first basic day, L qu For the u The trend electricity quantity of the first basic day is obtained through the trend electricity quantity model;

[0088] For statutory holidays, set a second time period and select the time before the statutory holidays. D 1 day to after statutory holidays D 2 days is the time period affected by the statutory holiday. If the day to be decomposed is the nth day of this month ' day, then the nth day of each month in the past second time period ' Day as the second base day;

[0089] Cycle power of the decomposition day L z' as follows:

[0090] ;

[0091] in, N' is the number of days in the second time period, and the number of days in the time period is greater than 31 days, L i ' For the i' The daily electricity consumption of the second basic day, L qi ' For the i' The trend electricity quantity of the second basic day is obtained through the trend electricity quantity model.

[0092] In one embodiment of the present invention, the meteorological electricity model is specifically:

[0093] Obtain meteorological power based on the power, trend power and cycle power of the day to be decomposed L t :

[0094] ;

[0095] in, L is the amount of electricity to be decomposed on the day, L q is the trend electricity of the day to be decomposed, L m for L z or L z' , Trend Power L qThe calculation formula is as follows:

[0096] .

[0097] In one embodiment of the present invention, constructing the average temperature correction model is specifically as follows:

[0098] Establish a temperature correction formula that takes into account the accumulated temperature effect:

[0099] ;

[0100] in, T v ' The revised date v The average temperature, T v For date v The average temperature, T min is the limit temperature, T v-j For date v The previous chapter j The average temperature of the day, k vj is the cumulative effect coefficient and satisfies k v1> k v2>…> k vp , p=min ( o , d max ), o For date v The number of consecutive days with a temperature above the limit, d max The maximum cumulative number of days.

[0101] In one embodiment of the present invention, the limit temperature and the maximum cumulative days in the temperature correction formula are determined in the following manner:

[0102] The temperature correction problem is formulated as the following optimization problem:

[0103] ;

[0104] in, RSS is the residual sum of squares, TSS is the total sum of squares, For the i The weather electricity of the day, is the average value of meteorological electricity, For the i The meteorological power forecast value obtained by piecewise fitting of the corrected temperature and meteorological power;

[0105] Solve the above optimization problem to obtain the limit temperature and maximum cumulative days corresponding to the optimal value, and substitute the limit temperature and maximum cumulative days corresponding to the optimal value into the temperature correction formula.

[0106] In one embodiment of the present invention, the optimization problem is solved by the COBYLA algorithm.

[0107] In one embodiment of the present invention, the trend power forecast value of the day to be predicted is obtained according to the trend power curve, and the cycle power forecast value of the day to be predicted is obtained according to the cycle power curve, specifically:

[0108] According to the trend of power curve The trend electricity of the forecast day is predicted. x i Set to the number of days from the first data day to the day to be predicted, the first data day being the earliest day of the daily electricity data selected when constructing the trend electricity curve;

[0109] The predicted value of the periodic electricity quantity for the day to be predicted is obtained according to the periodic electricity curve. Specifically, it is determined whether the day to be predicted is a statutory holiday. If so, the average of the periodic electricity quantity for the day to be predicted on the same holiday as the day to be predicted is obtained from the periodic electricity curve as the periodic electricity quantity for the day to be predicted. Otherwise, the position of the day to be predicted in a week is obtained, and the average of the periodic electricity quantity for the day to be predicted on the same holiday as the day to be predicted in the week is used as the periodic electricity quantity for the day to be predicted.

[0110] In one embodiment of the present invention, the meteorological power on the day to be predicted is predicted according to the corrected average temperature and meteorological power curve of the day to be predicted, specifically:

[0111] Correcting the average temperature of the predicted day according to the temperature correction formula;

[0112] The corrected average temperature of the day to be predicted is substituted into the meteorological power curve to obtain the meteorological power of the day to be predicted.

[0113] The following is a comparison between the present invention and embodiments of daily electricity forecasting based on daily electricity forecasting and daily electricity forecasting based on electricity decomposition.

[0114] like Figure 3 As shown, the method of the present invention is compared with the daily power forecast based on daily power forecast and the daily power forecast based on power decomposition. The data are all based on Figure 2 data, Figure 2 The curve after daily electricity consumption is decomposed by the method of the present invention and the historical daily electricity consumption data are shown, wherein the horizontal axis represents time and the vertical axis represents electricity consumption, with the unit being 100 million kWh. Figure 3It is a box plot, which is a statistical chart used to display data distribution and outliers. It mainly includes the following key elements:

[0115] 1. Box: represents the interquartile range (IQR) of the data. The lower boundary (bottom of the box) is the first quartile (Q1, 25% quartile); the upper boundary (top of the box) is the third quartile (Q3, 75% quartile).

[0116] 2. Median line: A horizontal line inside the box, indicating the median of the data.

[0117] 3. Whiskers: Indicates the range of the data (usually Q1-1.5*IQR and Q3 +1.5*IQR).

[0118] 4. Outlier points: Data points that fall outside the whiskers, usually marked as discrete dots.

[0119] from Figure 3 It can be seen that the method of the present invention (modified train represents the prediction result of the method of the present invention for the training set, and modified test represents the prediction result for the test set) performs daily electricity prediction based on electricity decomposition and temperature correction, and the accuracy is higher than that based on daily electricity prediction (original train represents the prediction result based on the daily electricity prediction method for the training set, and original test represents the prediction result for the test set) and daily electricity prediction based on electricity decomposition (decomposed train represents the prediction result based on the electricity decomposition method for the training set, and decomposed test represents the prediction result for the test set), that is, the method of the present invention has a smaller quarterback spacing, the loss represented by the median line is the smallest, the whisker range is smaller, and there are fewer outlier points falling outside the whiskers.

[0120] Table 1 Predicted R in some regions 2 contrast

[0121]

[0122] As shown in Table 1, R 2 It is much better than the first two methods and can accurately predict power consumption.

[0123] In a second aspect, a computer-readable storage medium, when a computer program stored thereon is executed, implements the power prediction method based on component decomposition and correction as described in the first aspect.

[0124] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0125] The units involved in the embodiments described in the present disclosure may be implemented by software or hardware, wherein the name of a unit does not, in some cases, limit the unit itself.

[0126] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0127] The above description is only a preferred embodiment of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of disclosure involved in the present disclosure is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present disclosure (but not limited to) by each other to form a technical solution.

[0128] In addition, although each operation is described in a specific order, this should not be understood as requiring these operations to be performed in the specific order shown or in a sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Similarly, although some specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of the present disclosure. Some features described in the context of a separate embodiment can also be implemented in a single embodiment in combination. On the contrary, the various features described in the context of a single embodiment can also be implemented in multiple embodiments individually or in any suitable sub-combination mode.

[0129] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.

[0130] There are a few points to note about this disclosure:

[0131] (1) The drawings of the embodiments of the present disclosure only relate to the structures related to the embodiments of the present disclosure. Other structures may refer to the general design.

[0132] (2) In the absence of conflict, the embodiments of the present disclosure and the features therein may be combined with each other to obtain new embodiments.

[0133] The above descriptions are merely embodiments of the present disclosure and are not intended to limit the patent scope of the present disclosure. Any equivalent structures made using the contents of the present disclosure and the drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present disclosure.

Claims

1. A method for predicting power based on component decomposition and correction, characterized in that: The following steps are involved: Get daily electricity data and daily average temperature for several historical days; Constructing a daily electricity data decomposition model, wherein the daily electricity data decomposition model includes a trend electricity model, a cycle electricity model and a meteorological electricity model; Constructing a mean temperature correction model; Correct the historical daily average temperature according to the average temperature correction model to obtain the corrected historical daily average temperature; Draw the trend power curve and the cycle power curve respectively through the trend power model and the cycle power model; The trend power forecast value of the day to be predicted is obtained according to the trend power curve, and the cycle power forecast value of the day to be predicted is obtained according to the cycle power curve; The meteorological electricity quantity of several historical days is obtained through the meteorological electricity quantity model, and the meteorological electricity quantity of the historical days is segmentedly fitted through the corrected historical daily average temperature to obtain a meteorological electricity quantity curve representing the change of meteorological electricity quantity with the corrected historical daily average temperature; Obtain the average temperature of the day to be predicted, and correct the average temperature of the day to be predicted according to the average temperature correction model to obtain the corrected average temperature of the day to be predicted; The meteorological power quantity on the forecast day is forecasted according to the corrected average temperature on the forecast day and the meteorological power quantity curve, so as to obtain the forecast value of meteorological power quantity; The predicted value of the electricity for the day to be predicted is obtained by summing the predicted value of the trend electricity for the day to be predicted, the predicted value of the periodic electricity for the day to be predicted and the predicted value of the meteorological electricity for the day to be predicted; The trend electricity model is constructed in the following way: Select daily electricity data with the same meteorological factors and on the nth day of each week, exclude daily electricity data on statutory holidays, and use the linear regression model to simulate the natural growth trend of electricity. The formula is as follows: ； ； in, x i For the i sky, y i For the i Daily electricity consumption, w , b is the fitting parameter, f ( x i ) is the i Fitted electricity consumption for the day; The cycle power model is constructed in the following way: For weekdays and weekends, the first time period is defined. If the day to be decomposed is the nth day of the week, the nth day of each week in the past first time period is taken as the first basic day; Cycle power of the decomposition day L z as follows: ； in, N is the number of days in the first time period, and the number of days in the time period is greater than 7 days, L u For the u The daily electricity consumption of the first basic day, L qu For the u The trend electricity quantity of the first basic day is obtained through the trend electricity quantity model; For statutory holidays, set a second time period and select the time before the statutory holidays. D 1 day to after statutory holidays D 2 days is the time period affected by the statutory holiday. If the day to be decomposed is the nth day of this month ' day, then the nth day of each month in the past second time period ' Day as the second base day; Cycle power of the decomposition day L z' as follows: ； in, N' is the number of days in the second time period, and the number of days in the time period is greater than 31 days, L i ' For the i' The daily electricity consumption of the second basic day, L qi ' For the i' The trend electricity quantity of the second basic day is obtained through the trend electricity quantity model.

2. The method for predicting electric quantity based on component decomposition and correction according to claim 1, characterized in that: The meteorological electricity model is specifically: Obtain meteorological power based on the power, trend power and cycle power of the day to be decomposed L t : ； in, L is the amount of electricity to be decomposed on the day, L q is the trend electricity of the day to be decomposed, L m for L z or L z' , Trend Power L q The calculation formula is as follows: 。 3. The method for predicting electric quantity based on component decomposition and correction according to claim 2 is characterized in that: The specific construction of the average temperature correction model is as follows: Establish a temperature correction formula that takes into account the accumulated temperature effect: ； in, T v ' The revised date v The average temperature, T v For date v The average temperature, T min is the limit temperature, T v-j For date v The previous chapter j The average temperature of the day, k vj is the cumulative effect coefficient and satisfies k v1> k v2>…> k vp , p=min ( o , d max ), o For date v The number of consecutive days with a temperature higher than the limit, d max The maximum cumulative number of days.

4. The method for predicting electric quantity based on component decomposition and correction according to claim 3 is characterized in that: The limit temperature and the maximum cumulative days in the temperature correction formula are determined by: The temperature correction problem is formulated as the following optimization problem: ； in, RSS is the residual sum of squares, TSS is the total sum of squares, For the i The weather electricity of the day, is the average value of meteorological electricity, For the i The meteorological power forecast value obtained by piecewise fitting of the corrected temperature and meteorological power; Solve the above optimization problem to obtain the limit temperature and maximum cumulative days corresponding to the optimal value, and substitute the limit temperature and maximum cumulative days corresponding to the optimal value into the temperature correction formula.

5. The method for predicting electric quantity based on component decomposition and correction according to claim 4 is characterized in that: The optimization problem is solved by the COBYLA algorithm.

6. The method for predicting electric quantity based on component decomposition and correction according to claim 4 is characterized in that: The trend power forecast value of the day to be predicted is obtained according to the trend power curve, and the cycle power forecast value of the day to be predicted is obtained according to the cycle power curve. Specifically: According to the trend of power curve The trend electricity of the forecast day is predicted. x i Set to the number of days from the first data day to the day to be predicted, the first data day being the earliest day of the daily electricity data selected when constructing the trend electricity curve; The predicted value of the periodic electricity quantity for the day to be predicted is obtained according to the periodic electricity curve. Specifically, it is determined whether the day to be predicted is a statutory holiday. If so, the average of the periodic electricity quantity for the day to be predicted on the same holiday as the day to be predicted is obtained from the periodic electricity curve as the periodic electricity quantity for the day to be predicted. Otherwise, the position of the day to be predicted in a week is obtained, and the average of the periodic electricity quantity for the day to be predicted on the same holiday as the day to be predicted in the week is used as the periodic electricity quantity for the day to be predicted.

7. The method for predicting electric quantity based on component decomposition and correction according to claim 4 is characterized in that: According to the revised average temperature and meteorological power curve of the day to be predicted, the meteorological power of the day to be predicted is predicted as follows: Correcting the average temperature of the predicted day according to the temperature correction formula; The corrected average temperature of the day to be predicted is substituted into the meteorological power curve to obtain the meteorological power of the day to be predicted.

8. A computer-readable storage medium, characterized in that: When the computer program stored thereon is executed, the power prediction method based on component decomposition and correction as described in any one of claims 1-7 is implemented.

Images