Resident load similar day selection method considering high temperature cumulative effect

Abstract

The invention relates to a resident load similar day selection method considering a high-temperature cumulative effect, and belongs to the technical field of short-term power load prediction. The method comprises the following steps: S1, considering the influence of a high-temperature cumulative effect, and carrying out daily maximum temperature correction by adopting a method of combining a weighting function with parameter optimization on the basis of determining a critical temperature; s2, according to the rainfall level, carrying out rainfall assignment through a segmentation mapping method; s3, based on the historical resident load and meteorological data, the daily maximum temperature, the rainfall, the maximum sensible temperature and the average sensible temperature are selected as characteristic quantities through a Pearson's correlation coefficient method; and S4, constructing an Euclidean distance index based on the selected characteristic quantity, substituting the data of the day to be measured and the data of the historical day into an Euclidean distance index formula, and selecting a date with a relatively small index value as a similar day. According to the method, the accuracy of selecting the summer resident load similar day can be improved, and the method has important significance in improving the prediction precision of the short-term resident load in the peak-meeting summer period.

Description

Technical Field

[0001] The invention belongs to the technical field of short-term power load forecasting and relates to a method for selecting similar days for residential loads taking into account high temperature cumulative effects. Background Art

[0002] Short-term residential load forecasting is the basis of power grid operation and production planning, especially during the peak summer period. Residential load forecasting is of great significance to ensuring the safety of people's livelihood electricity use. Similar day selection is a key link in short-term residential load forecasting. At present, most of the existing similar day selection methods consider temperature factors such as maximum temperature, minimum temperature, and average temperature. Under continuous high temperatures in summer, the temperature accumulation effect will significantly affect the residential load, while the existing similar day selection methods usually ignore the high temperature accumulation effect factor. Therefore, in order to improve the accuracy of similar day selection for residential load during the peak summer period, it is particularly important to design a similar day selection method for residential load that takes into account the high temperature accumulation effect to improve the accuracy of similar day selection in view of the low accuracy of the existing similar day selection. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a method for selecting similar days for residential loads taking into account the cumulative effect of high temperature, so as to solve the problem that the prior art does not consider the cumulative effect of high temperature in the selection of similar days, resulting in low accuracy in the selection of similar days. The method of the present invention takes into account the correction of the maximum daily temperature by the cumulative effect of high temperature, and selects similar days by using the corrected maximum daily temperature, which can improve the accuracy of similar day selection and is of great significance to improving the accuracy of short-term residential load forecasting.

[0004] In order to achieve the above object, the present invention provides the following technical solutions:

[0005] A method for selecting similar days of residential load taking into account the cumulative effect of high temperature includes the following steps:

[0006] S1: Considering the influence of high temperature accumulation effect, based on the determination of critical temperature, the daily maximum temperature is corrected by using weighted function combined with parameter optimization method;

[0007] S2: The influence of rainfall on the load of residents in summer is different. Specifically, when the rainfall is heavy, the load of residents decreases accordingly as the rainfall increases. However, when the rainfall is light, the small rainfall will aggravate the hot and humid air, thereby increasing the load of residents. When similar days are selected, in order to consider the influence of rainfall factors on the load of residents, the rainfall is mapped to the interval [0,1] according to the rainfall level through the segmented mapping method;

[0008] S3: Based on historical residential load data and meteorological data, the daily maximum temperature, rainfall, maximum perceived temperature and average perceived temperature were selected as quantities through the Pearson correlation coefficient method;

[0009] S4: Construct a Euclidean distance index based on the selected feature quantity, substitute the data of the day to be tested and the historical day data into the Euclidean distance index formula, obtain the Euclidean distance index value between the day to be tested and the historical day, and select the date with the smaller Euclidean distance index value as the similar day.

[0010] Further, step S1 specifically includes the following steps:

[0011] S11: Determine the lowest critical temperature point of high temperature cumulative effect;

[0012] Firstly, based on multi-day historical load data, the fitting curve L between the maximum temperature and the maximum power load in summer is obtained by polynomial fitting method. max , as shown in formula (1):

[0013]

[0014] Among them, L max is the daily maximum power load; T0 is the daily maximum temperature; a0, a1, a2 are polynomial parameters;

[0015] Then, by fitting the curve L max The sensitivity curve S of the daily maximum power load relative to the daily maximum temperature is obtained by the derivation method, as shown in formula (2):

[0016]

[0017] Finally, the average sensitivity curve S can be obtained from the sensitivity curve S avg The temperature at the intersection of the two curves is determined as the lowest critical temperature point T of the high temperature cumulative effect. low ;

[0018] S12: Corrected daily maximum temperature;

[0019] If the daily maximum temperature is higher than the minimum critical temperature point T for three consecutive days or more, low When the maximum temperature of the day is , the cumulative effect of high temperature is considered and the correction is made using the formula shown in formula (3);

[0020] T0'=aT0+bT1+cT2 (3)

[0021] Among them, T0′ is the daily maximum temperature after correction considering the cumulative effect of high temperature; T0 is the maximum temperature of the day to be corrected; T1 is the daily maximum temperature of the day before the correction; T2 is the maximum temperature of the two days before the correction; a, b, c are coefficients, 0≤a,b,c≤1.

[0022] Further, in step S12, the coefficients a, b, and c are obtained by using a parameter optimization method, specifically as follows: construct the Pearson correlation coefficient between the daily maximum temperature and the daily maximum power load, as shown in formula (4); take the maximum Pearson correlation coefficient as the objective function, take 0≤a, b, c≤1 as the constraint condition, and use a genetic algorithm to globally optimize the parameters a, b, and c;

[0023]

[0024] Among them, r(T0′,P) is the Pearson correlation coefficient between the corrected daily maximum temperature T0′ and the daily maximum power load P; N is the number of samples.

[0025] Further, in step S2, according to the rainfall level, the rainfall is mapped to the [0,1] interval by a segmented mapping method, as shown in formula (5);

[0026]

[0027] Among them, R is the rainfall mapping value.

[0028] Furthermore, in step S3, the calculation formula of the body temperature AT is as follows:

[0029] AT=1.07T+0.2E-0.65V-2.7 (6)

[0030]

[0031] Wherein, AT is the perceived temperature (°C), T is the air temperature (°C), E is the water vapor pressure (hPa), V is the wind speed (m / sec), and RH is the relative humidity (%).

[0032] Further, in step S4, the Euclidean distance indicator formula is as shown in formula (8):

[0033]

[0034] Among them, b1, b2, b3, b4 are respectively the normalized historical daily maximum temperature value, rainfall mapping value, maximum perceived temperature value and average perceived temperature value; c1, c2, c3, c4 are respectively the normalized maximum temperature value, rainfall mapping value, maximum perceived temperature value and average perceived temperature value of the day to be measured;

[0035] Specifically, the method for determining the historical daily maximum temperature value b1 and the maximum temperature value c1 of the day to be measured is as follows: when the maximum temperature of the historical day or the day to be measured is greater than or equal to the critical temperature, it is determined as the corrected daily maximum temperature T0′; when the maximum temperature of the historical day or the day to be measured is less than the critical temperature, it is determined as the uncorrected daily maximum temperature T0.

[0036] The beneficial effects of the present invention are as follows: the present invention takes into account the cumulative effect of high temperature to correct the daily maximum temperature, and selects similar days through the corrected daily maximum temperature, which can improve the accuracy of selecting similar days for summer residential load, and is of great significance to improving the accuracy of short-term residential load prediction during the peak summer period.

[0037] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0039] Figure 1 A flow chart of the method for selecting similar days of residential loads taking into account the cumulative effect of high temperature according to the present invention;

[0040] Figure 2 Selecting a similar day result graph in the embodiment of the present invention;

[0041] Figure 3 Result diagram of evaluation index in the embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following describes the embodiments of the present invention by specific examples, and 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 noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0043] See also Figure 1 to Figure 3 The present invention provides a method for selecting similar days of residential loads taking into account the cumulative effect of high temperature, such as Figure 1 As shown, the method specifically comprises the following steps:

[0044] S1: daily maximum temperature correction considering the cumulative effect of high temperature;

[0045] Taking into account the cumulative effect of high temperature, based on the determination of critical temperature, the daily maximum temperature is corrected by using a weighted function combined with parameter optimization method. The details are as follows:

[0046] S11: Determination of the lowest critical temperature point of high temperature cumulative effect;

[0047] Firstly, based on the multi-day historical load data, the fitting curve L between the maximum temperature and the maximum power load in summer is obtained by the polynomial fitting method. max , L max , as shown in formula (1).

[0048]

[0049] Among them, L max is the daily maximum power load; T0 is the daily maximum temperature; a0, a1, a2 are polynomial parameters.

[0050] Then, by fitting the curve L max The sensitivity curve S of the daily maximum power load relative to the daily maximum temperature is obtained by the derivation method, as shown in formula (2):

[0051]

[0052] Finally, the average sensitivity curve S can be obtained from the sensitivity curve S avg The temperature at the intersection of the two curves is determined as the lowest critical temperature point T of the high temperature cumulative effect. low .

[0053] S12: daily maximum temperature correction method;

[0054] If the daily maximum temperature is higher than the minimum critical temperature point T for three consecutive days or more, low When the maximum temperature of the day is , the cumulative effect of high temperature is taken into account to correct the maximum temperature of the day. The correction is made using the formula shown in formula (3).

[0055] T0'=aT0+bT1+cT2 (3)

[0056] Among them, T0′ is the daily maximum temperature after correction considering the cumulative effect of high temperature; T0 is the maximum temperature of the day to be corrected; T1 is the maximum temperature of the day before the day to be corrected; T2 is the maximum temperature of the two days before the day to be corrected; a, b, c are coefficients, 0≤a, b, c≤1.

[0057] The coefficients a, b, and c are obtained by parameter optimization method, as follows: construct the Pearson correlation coefficient between the daily maximum temperature and the daily maximum power load, as shown in formula (4). Taking the maximum Pearson correlation coefficient as the objective function and 0≤a, b, c≤1 as the constraint conditions, a, b, and c parameters are globally optimized by genetic algorithm.

[0058]

[0059] Among them, r(T0′,P) is the Pearson correlation coefficient between the corrected daily maximum temperature T0′ and the daily maximum power load P; N is the number of samples.

[0060] S2: Mapping of rainfall;

[0061] The influence of rainfall on the load of residents in summer is different, as follows: when the rainfall is heavy, the load of residents decreases accordingly as the rainfall increases, while when the rainfall is light, the small rainfall will aggravate the air stuffiness and increase the load of residents. When selecting similar days, in order to consider the influence of rainfall factors on the load of residents, according to the rainfall level, the rainfall is mapped to the interval [0,1] through the segmented mapping method, as shown in formula (5).

[0062]

[0063] Among them, R is the rainfall mapping value.

[0064] S3: Selection of characteristic meteorological factors;

[0065] Based on historical residential load data and meteorological data, the maximum daily temperature, rainfall, maximum perceived temperature and average perceived temperature were selected as characteristic meteorological factors through the Pearson correlation coefficient method. The calculation formula for perceived temperature AT is as follows:

[0066] AT=1.07T+0.2E-0.65V-2.7 (6)

[0067]

[0068] Wherein, AT is the perceived temperature (°C), T is the air temperature (°C), E is the water vapor pressure (hPa), V is the wind speed (m / sec), and RH is the relative humidity (%).

[0069] S4: Euclidean distance calculation;

[0070] Based on the selected feature quantity, the Euclidean distance index is constructed, and the data of the day to be tested and the historical day data are substituted into the Euclidean distance index formula to obtain the Euclidean distance index value between the day to be tested and the historical day, and the date with the smaller Euclidean distance index value is selected as the similar day.

[0071] The specific formula of the Euclidean distance index is shown in formula (8).

[0072]

[0073] Among them, b1, b2, b3, and b4 are the normalized historical daily maximum temperature value, rainfall mapping value, maximum perceived temperature value, and average perceived temperature value, respectively; c1, c2, c3, and c4 are the normalized maximum temperature value, rainfall mapping value, maximum perceived temperature value, and average perceived temperature value, respectively.

[0074] Specifically, the method for determining the historical daily maximum temperature value b1 and the maximum temperature value c1 of the day to be measured is as follows: when the maximum temperature of the historical day or the day to be measured is greater than or equal to the critical temperature, it is determined as the corrected daily maximum temperature T0′; when the maximum temperature of the historical day or the day to be measured is less than the critical temperature, it is determined as the uncorrected daily maximum temperature T0.

[0075] Example:

[0076] In this embodiment, actual residential load data and meteorological data of a certain area in southwest China in July and August 2022 are selected as data sets. Among them, the residential load on August 17, 2022 is selected as the daily data set to be tested, and the residential load on other dates is selected as the historical daily data set. The specific implementation steps are as follows: Figure 1 shown.

[0077] The five dates with the smallest Euclidean distance index values ​​are selected as similar days through Euclidean distance calculation. The selected five similar days are August 22, 2022 (distance: 0.0566), August 21, 2022 (distance: 0.0649), August 16, 2022 (distance: 0.0920), August 18, 2022 (distance: 0.1001), and August 20, 2022 (distance: 0.1272). The specific selection results are as follows Figure 2 shown.

[0078] In order to verify the effectiveness of the method used in the present invention, the mean absolute error (MAE) is selected as the evaluation index, and formula (9) is used for calculation.

[0079]

[0080] Where n is the number of samples; y i For historical day resident loads; is the residential load on the day to be measured.

[0081] The evaluation index calculation results are as follows: Figure 3 As shown, according to the results, it can be seen that the average absolute errors of the selected 5 similar days are all less than 4%, and the smaller the Euclidean distance of the selected typical day, the smaller the average absolute error. The above results well verify the effectiveness of the method for selecting similar days for residential loads of the present invention.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A method for selecting similar days of residential load taking into account the cumulative effect of high temperature, characterized in that: The method specifically comprises the following steps: S1: Considering the influence of high temperature accumulation effect, based on the determination of critical temperature, the daily maximum temperature is corrected by using weighted function combined with parameter optimization method; S2: According to the rainfall level, the rainfall is mapped to the interval [0,1] through the segmented mapping method; S3: Based on historical residential load data and meteorological data, the daily maximum temperature, rainfall, maximum perceived temperature and average perceived temperature were selected as feature quantities through the Pearson correlation coefficient method; S4: Constructing the Euclidean distance index based on the selected feature quantity, substituting the data of the day to be tested and the historical day data into the Euclidean distance index formula, obtaining the Euclidean distance index value between the day to be tested and the historical day, and selecting the date with the smallest Euclidean distance index value as the similar day.

2. The method for selecting similar days of residential load according to claim 1, characterized in that: Step S1 specifically includes the following steps: S11: Determine the lowest critical temperature point of high temperature cumulative effect; Firstly, based on multi-day historical load data, the fitting curve L between the maximum temperature and the maximum power load in summer is obtained by polynomial fitting method. max , as shown in formula (1): Among them, L max is the daily maximum power load; T0 is the daily maximum temperature; a0, a1, a2 are polynomial parameters; Then, by fitting the curve L max The sensitivity curve S of the daily maximum power load relative to the daily maximum temperature is obtained by the derivation method, as shown in formula (2): Finally, the average sensitivity curve S is obtained from the sensitivity curve S avg The temperature at the intersection of the two curves is determined as the lowest critical temperature point T of the high temperature cumulative effect. low ; S12: Corrected daily maximum temperature; If the daily maximum temperature is higher than the minimum critical temperature point T for three consecutive days or more, low When the maximum temperature of the day is , the cumulative effect of high temperature is considered and the correction is made using the formula shown in formula (3); T0'=aT0+bT1+cT2 (3) Among them, T0′ is the daily maximum temperature after correction considering the cumulative effect of high temperature; T0 is the maximum temperature of the day to be corrected; T1 is the daily maximum temperature of the day before the correction; T2 is the maximum temperature of the two days before the correction; a, b, c are coefficients, 0≤a,b,c≤1.

3. The method for selecting similar days of residential load according to claim 2, characterized in that: In step S12, the coefficients a, b, and c are obtained by using a parameter optimization method, specifically as follows: construct the Pearson correlation coefficient between the daily maximum temperature and the daily maximum power load, as shown in formula (4); take the maximum Pearson correlation coefficient as the objective function, take 0≤a, b, c≤1 as the constraint condition, and use a genetic algorithm to globally optimize the parameters a, b, and c; Among them, r(T0′,P) is the Pearson correlation coefficient between the corrected daily maximum temperature T0′ and the daily maximum power load P; N is the number of samples.

4. The method for selecting similar days of residential load according to claim 1, characterized in that: In step S2, according to the rainfall level, the rainfall is mapped to the interval [0,1] by a segmented mapping method, as shown in formula (5); Among them, R is the rainfall mapping value.

5. The method for selecting similar days of residential load according to claim 1, characterized in that: In step S3, the calculation formula of the body temperature AT is as follows: AT=1.07T+0.2E-0.65V-2.7 (6) Among them, AT is the perceived temperature, T is the air temperature, E is the water vapor pressure, V is the wind speed, and RH is the relative humidity.

6. The method for selecting similar days of residential load according to claim 1, characterized in that: In step S4, the Euclidean distance indicator formula is as shown in formula (8): Among them, b1, b2, b3, b4 are respectively the normalized historical daily maximum temperature value, rainfall mapping value, maximum perceived temperature value and average perceived temperature value; c1, c2, c3, c4 are respectively the normalized maximum temperature value, rainfall mapping value, maximum perceived temperature value and average perceived temperature value of the day to be measured; Specifically, the method for determining the historical daily maximum temperature value b1 and the maximum temperature value c1 of the day to be measured is as follows: when the maximum temperature of the historical day or the day to be measured is greater than or equal to the critical temperature, it is determined as the corrected daily maximum temperature T0′; when the maximum temperature of the historical day or the day to be measured is less than the critical temperature, it is determined as the uncorrected daily maximum temperature T0.

Images