CN-122026298-A - New energy day-ahead predicted power correction method and system for load peak time

CN122026298ACN 122026298 ACN122026298 ACN 122026298ACN-122026298-A

Abstract

The invention provides a new energy day-ahead predicted power correction method and system for load peak time, which relate to the technical field of new energy power generation and comprise the steps of obtaining new energy initial day-ahead predicted power of a predicted object in a predicted day-full time period and historical electric power before a predicted day; and correcting the load peak time part in the initial day-ahead predicted power of the new energy according to the new energy output expected deviation value, and further combining the day-ahead predicted power of the daily load peak time with the non-load peak time part of the initial day-ahead predicted power of the new energy to obtain the new energy target day-ahead predicted power of the predicted day-full time. According to the method, the new energy initial day-ahead predicted power is corrected through the obtained new energy output expected deviation value, the reliability of the power prediction result of the prediction day is improved, and the risk of a power supply gap is reduced.

Inventors

WU JI
ZHANG HUIYU
WANG ZHENG
MA WENWEN
ZHOU HAI
WANG QINGLIANG
CHENG XU
ZHOU CHAOHUI
QIN FANG
YAN PENG

Assignees

中国电力科学研究院有限公司
国家电网有限公司
国网福建省电力有限公司

Dates

Publication Date: 20260512
Application Date: 20241104

Claims (20)

1. The new energy daily forecast power correction method for the load peak time is characterized by comprising the following steps of: acquiring the initial day-ahead predicted power of a new energy source of a predicted object in a full period of a predicted day and the historical electric power before the predicted day; constructing a new energy output deviation power set in a load peak time in a full time period according to the historical electric power; Calculating a new energy output expected deviation value by utilizing the output deviation power distribution characteristics of the new energy output deviation power set and combining an output expected deviation expression corresponding to the output deviation power distribution characteristics; Correcting a load peak time part in the initial daily predicted power of the new energy according to the expected deviation value of the new energy output to obtain the daily predicted power of the daily load peak time; and combining the daily predicted power of the daily load peak period with the non-load peak period part in the initial daily predicted power of the new energy to obtain the new energy target daily predicted power of the predicted daily total period.
2. The method of claim 1, wherein the historical power before the predicted day comprises daily load power, actual power output of the new energy source, and initial predicted power before the day for the new energy source at each time in a historical time period; the construction of the new energy output deviation power set in the load peak time in the full time period according to the historical electric power comprises the following steps: determining a load peak period according to the daily load power at each moment in the historical time period; And determining a new energy output deviation power set of the load peak time according to the new energy actual output power of the load peak time and the new energy initial day-ahead predicted power.
3. The method of claim 2, wherein said determining a load peak period from the daily load power at each instant in said historical period of time comprises: Determining daily load peak time in the historical time period according to daily load power of each time in the historical time period; Determining the load peak time of the historical time period according to the daily load peak time; and determining the load peak time and the set time before and after the load peak time in the historical time period as the load peak time.
4. A method according to claim 2 or 3, wherein said determining a new energy output bias power set for said load peak period from said new energy actual output power and a new energy initial day-ahead predicted power for said load peak period comprises: the method comprises the steps of screening the actual output power of the new energy and the initial daily predicted power of the new energy at each moment in a historical time period before a predicted day according to a load peak time period to obtain the actual output power of the new energy and the initial daily predicted power of the new energy at each moment in the load peak time period; And determining a new energy output deviation power set at each moment in the load peak time according to the actual output power of the new energy at each moment day by day in the load peak time and the corresponding initial day-ahead predicted power of the new energy.
5. The method according to claim 4, wherein the determining the new energy output deviation power set at each moment in the peak load period according to the actual output power of the new energy and the corresponding initial pre-day predicted power of the new energy at each moment in the peak load period comprises: The actual output power of the new energy at each moment day by day in the load peak time is differed from the corresponding initial day-ahead predicted power of the new energy, and the output deviation power of the new energy at each moment day by day in the load peak time is obtained; aiming at each moment, the daily new energy output deviation power is orderly arranged in order from small to large, and the new energy output deviation power set in the load peak time is determined.
6. The method of claim 1, wherein the determining of the desired deviation expression of the output force corresponding to the output force deviation power distribution characteristic comprises: Selecting the median of the new energy output deviation power set for each moment; When the output deviation power distribution characteristic is that the median of the new energy output deviation power set is smaller than or equal to a preset value, selecting a lognormal distribution new energy output expected deviation expression; And when the output deviation power distribution characteristic is that the median of the new energy output deviation power set is larger than a preset value, a normal distribution new energy output expected deviation expression is selected.
7. The method of claim 6, wherein calculating a new energy output desired deviation value using the output deviation power distribution characteristics of the new energy output deviation power set in combination with an output desired deviation expression corresponding to the output deviation power distribution characteristics comprises: when the output deviation power distribution characteristic is that the median of the new energy output deviation power set is smaller than or equal to a preset value, selecting the new energy output deviation power which is larger than the preset value in the new energy output deviation power set as a new energy output deviation calculation power set; Calculating standard deviation and average value of the new energy output deviation calculation power set; Substituting the standard deviation and the mean value into the lognormal distribution new energy output expected deviation expression, and calculating a new energy output expected deviation value; the expected deviation expression of the output of the log-normal distribution new energy is as follows: Wherein E (x n ) is the expected deviation of the new energy output at the time n, x n is the new energy output deviation calculation power set at the time n, mu n is the average value of the new energy output deviation calculation power set at the time n, and sigma n is the standard deviation of the new energy output deviation calculation power set at the time n.
8. The method of claim 6, wherein calculating a new energy output desired deviation value using the output deviation power distribution characteristics of the new energy output deviation power set in combination with an output desired deviation expression corresponding to the output deviation power distribution characteristics comprises: when the output deviation power distribution characteristic is that the median of the new energy output deviation power set is larger than a preset value, determining all new energy output deviation powers in the new energy output deviation power set as a new energy output deviation calculation power set; Calculating the average value of the new energy output deviation calculation power set; substituting the average value into the normal distribution new energy output expected deviation expression to calculate a new energy output expected deviation value; The expected deviation expression of the normal distribution new energy output is as follows: E(x n )＝μ n Wherein E (x n ) is the expected deviation of the new energy output at the time n, x n is the new energy output deviation calculation power set at the time n, and mu n is the average value of the new energy output deviation calculation power set at the time n.
9. The method of claim 1, wherein the correcting the load peak period in the initial daily predicted power of the new energy source according to the expected deviation value of the new energy source output to obtain the daily predicted power of the daily load peak period comprises: Obtaining the daily predicted power of the daily load peak period according to the daily predicted power of each moment of the daily load peak period according to the difference value of the new energy initial daily predicted power and the corresponding new energy output expected deviation value in the load peak period part of the new energy initial daily predicted power; the calculation formula of the daily predicted power at each moment in the daily load peak period part is as follows: P n correct ＝P n -E(x n ) Wherein, P n correct is the predicted power before day at the time n in the load peak period part, P n is the predicted power before day at the new energy source at the time n in the load peak period part, E (x n ) is the expected deviation value of the new energy source output at the time n in the load peak period part.
10. The method according to claim 1 or 9, wherein the combining the non-load peak period part of the day-ahead predicted power of the day-ahead load peak period and the initial day-ahead predicted power of the new energy source, to obtain the new energy source target day-ahead predicted power of the predicted day-ahead period, further comprises: Determining the starting and ending time of the load peak time in the predicted full-day target pre-day predicted power and the starting and ending time of the non-load peak time as the period to be adjusted of the new energy target pre-day predicted power; Smoothly adjusting the predicted power of the new energy target day-ahead in the period to be adjusted through a moving average calculation formula to obtain the predicted power of the new energy target day-ahead in the full period of the adjusted predicted day; wherein, the moving average calculation formula is as follows: P i ＝(P i-1 +P i+1 )/2 Wherein, P i is the new energy target day-ahead predicted power of the time i in the adjusted predicted day-to-adjust period, P i-1 is the new energy target day-ahead predicted power of the time i-1 before the predicted day time i, and P i+1 is the new energy target day-ahead predicted power of the time i+1 after the predicted day time i.
11. A new energy pre-day prediction power correction system for peak load time, comprising: the power acquisition module is used for acquiring the initial day-ahead predicted power of the new energy source of the predicted object in the full period of the predicted day and the historical electric power before the predicted day; the new energy output deviation power set construction module is used for constructing a new energy output deviation power set of a load peak time in a full time period according to the historical electric power; The new energy output expected deviation value calculation module is used for calculating a new energy output expected deviation value by utilizing the output deviation power distribution characteristics of the new energy output deviation power set and combining an output expected deviation expression corresponding to the output deviation power distribution characteristics; The load peak time day-ahead predicted power acquisition module is used for correcting the load peak time part in the initial day-ahead predicted power of the new energy according to the new energy output expected deviation value to obtain day-ahead predicted power of the day-ahead load peak time; and the full-period day-ahead predicted power acquisition module is used for combining the day-ahead predicted power of the day load peak time period with the non-load peak time period part in the new energy initial day-ahead predicted power to obtain the new energy target day-ahead predicted power of the predicted day full time period.
12. The system of claim 11, wherein the historical power before the predicted day comprises daily load power, actual power output of the new energy source, and initial predicted power before the day for the new energy source at each time in a historical time period; the new energy output deviation power set construction module comprises: a load peak period determining sub-module for determining a load peak period according to daily load power at each moment in the historical period; And the new energy output deviation power set determining submodule is used for determining the new energy output deviation power set of the load peak time according to the new energy actual output power of the load peak time and the new energy initial day-ahead predicted power.
13. The system of claim 12, wherein the peak load period determination submodule is specifically configured to: Determining daily load peak time in the historical time period according to daily load power of each time in the historical time period; Determining the load peak time of the historical time period according to the daily load peak time; and determining the load peak time and the set time before and after the load peak time in the historical time period as the load peak time.
14. The system of claim 12 or 13, wherein the new energy output deviation power set determination submodule includes: the load peak time power screening unit is used for screening the actual output power of the new energy and the initial daily predicted power of the new energy at each moment in the historical time period before the predicted day according to the load peak time to obtain the actual output power of the new energy and the initial daily predicted power of the new energy at each moment in the load peak time; And the new energy output deviation power set determining unit is used for determining the new energy output deviation power set at each moment in the load peak time according to the actual output power of the new energy at each moment day by day in the load peak time and the corresponding initial day-ahead predicted power of the new energy.
15. The system of claim 14, wherein the new energy output deviation power set determining unit is specifically configured to: The actual output power of the new energy at each moment day by day in the load peak time is differed from the corresponding initial day-ahead predicted power of the new energy, and the output deviation power of the new energy at each moment day by day in the load peak time is obtained; aiming at each moment, the daily new energy output deviation power is orderly arranged in order from small to large, and the new energy output deviation power set in the load peak time is determined.
16. The system of claim 11, wherein the new energy output expected deviation value calculation module is configured to: Selecting the median of the new energy output deviation power set for each moment; When the output deviation power distribution characteristic is that the median of the new energy output deviation power set is smaller than or equal to a preset value, selecting a lognormal distribution new energy output expected deviation expression; And when the output deviation power distribution characteristic is that the median of the new energy output deviation power set is larger than a preset value, a normal distribution new energy output expected deviation expression is selected.
17. The system of claim 16, wherein the new energy output expected deviation value calculation module is specifically configured to: when the output deviation power distribution characteristic is that the median of the new energy output deviation power set is smaller than or equal to a preset value, selecting the new energy output deviation power which is larger than the preset value in the new energy output deviation power set as a new energy output deviation calculation power set; Calculating standard deviation and average value of the new energy output deviation calculation power set; Substituting the standard deviation and the mean value into the lognormal distribution new energy output expected deviation expression, and calculating a new energy output expected deviation value; the expected deviation expression of the output of the log-normal distribution new energy is as follows: Wherein E (x n ) is the expected deviation of the new energy output at the time n, x n is the new energy output deviation calculation power set at the time n, mu n is the average value of the new energy output deviation calculation power set at the time n, and sigma n is the standard deviation of the new energy output deviation calculation power set at the time n.
18. The system of claim 16, wherein the new energy output expected deviation value calculation module is specifically configured to: when the output deviation power distribution characteristic is that the median of the new energy output deviation power set is larger than a preset value, determining all new energy output deviation powers in the new energy output deviation power set as a new energy output deviation calculation power set; Calculating the average value of the new energy output deviation calculation power set; substituting the average value into the normal distribution new energy output expected deviation expression to calculate a new energy output expected deviation value; The expected deviation expression of the normal distribution new energy output is as follows: E(x n )＝μ n Wherein E (x n ) is the expected deviation of the new energy output at the time n, x n is the new energy output deviation calculation power set at the time n, and mu n is the average value of the new energy output deviation calculation power set at the time n.
19. The system of claim 11, wherein the peak load period pre-day predicted power acquisition module is configured to: Obtaining the daily predicted power of the daily load peak period according to the daily predicted power of each moment of the daily load peak period according to the difference value of the new energy initial daily predicted power and the corresponding new energy output expected deviation value in the load peak period part of the new energy initial daily predicted power; the calculation formula of the daily predicted power at each moment in the daily load peak period part is as follows: P n correct ＝P n -E(x n ) Wherein, P n correct is the predicted power before day at the time n in the load peak period part, P n is the predicted power before day at the new energy source at the time n in the load peak period part, E (x n ) is the expected deviation value of the new energy source output at the time n in the load peak period part.
20. The system according to claim 11 or 19, wherein the system further comprises: The full-period day-ahead prediction power adjustment module is used for: Determining the starting and ending time of the load peak time in the predicted full-day target pre-day predicted power and the starting and ending time of the non-load peak time as the period to be adjusted of the new energy target pre-day predicted power; Smoothly adjusting the predicted power of the new energy target day-ahead in the period to be adjusted through a moving average calculation formula to obtain the predicted power of the new energy target day-ahead in the full period of the adjusted predicted day; wherein, the moving average calculation formula is as follows: P i ＝(P i-1 +P i+1 )/2 Wherein, P i is the new energy target day-ahead predicted power of the time i in the adjusted predicted day-to-adjust period, P i-1 is the new energy target day-ahead predicted power of the time i-1 before the predicted day time i, and P i+1 is the new energy target day-ahead predicted power of the time i+1 after the predicted day time i.

Description

New energy day-ahead predicted power correction method and system for load peak time Technical Field The invention relates to the technical field of new energy power generation, in particular to a new energy daily forecast power correction method and system for load peak time. Background The random fluctuation of the new energy power makes the power grid face the problems of power supply protection and power consumption promotion, and the accurate new energy power prediction is one of effective means for solving the problems, so that the core work of 'power supply protection and power consumption promotion' of a company can be effectively supported. In the period of high power consumption load such as winter and summer, cold tide, the accuracy of new energy power prediction is crucial to electric power and electric quantity balance. In the prior art, although many researches on a day-ahead power prediction algorithm exist, the existing prediction algorithm only performs day-ahead power prediction for all time periods, a special time period is not considered, and a power prediction correction algorithm for the special time period is lacked, so that the problem of inaccurate power prediction results is caused. For example, since the power supply demand is high in the load peak period, the power supply demand is a key attention period of power grid power and electric quantity balance, and if the predicted force value is much higher than the actual force value in the period, a large power supply gap will appear, so that smooth operation of the power grid is affected. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides a new energy daily forecast power correction method facing to a load peak time, which comprises the following steps: Acquiring the initial day-ahead predicted power of a new energy source of a predicted object in a full period of a predicted day and the historical electric power of the predicted object before the predicted day; constructing a new energy output deviation power set of a load peak time in a full time period according to the historical electric power; Calculating a new energy output expected deviation value by utilizing the output deviation power distribution characteristics of the new energy output deviation power set and combining an output expected deviation expression corresponding to the output deviation power distribution characteristics; Correcting a load peak time part in the initial daily predicted power of the new energy according to the expected deviation value of the new energy output to obtain the daily predicted power of the daily load peak time; And combining the daily predicted power of the daily load peak period with the non-load peak period part in the initial daily predicted power of the new energy to obtain the new energy target daily predicted power of the predicted daily total period. Preferably, the historical electric power before the forecast day comprises daily load power, actual output power of new energy and initial daily forecast power of new energy at each moment in a historical time period; constructing a new energy output deviation power set of a load peak time in a full time period according to historical electric power, comprising: determining a load peak period according to the daily load power at each moment in the historical period; And determining a new energy output deviation power set in the load peak period according to the new energy actual output power in the load peak period and the new energy initial day-ahead predicted power. Preferably, determining the load peak period from the daily load power at each time in the historical period includes: Determining daily load peak time in the historical time period according to daily load power of each time in the historical time period; determining the load peak time of the historical time period according to the daily load peak time; the load peak time and the set times before and after the load peak time in the history period are determined as the load peak time. Preferably, determining the new energy output deviation power set in the load peak period according to the new energy actual output power in the load peak period and the new energy initial day-ahead predicted power comprises: the method comprises the steps of screening the actual output power of the new energy and the initial daily predicted power of the new energy at each moment in a historical time period before a predicted day according to a load peak time period to obtain the actual output power of the new energy and the initial daily predicted power of the new energy at each moment in the load peak time period; And determining a new energy output deviation power set at each moment in the load peak time according to the actual new energy output power at each moment day by day in the load peak time and the corresponding new energy initial day-ahead predicted power. Preferably, determining t