CN-122026537-A - Real-time deviation correcting rolling scheduling method for wind-light complementary system coupling wind-light uncertainty

CN122026537ACN 122026537 ACN122026537 ACN 122026537ACN-122026537-A

Abstract

The invention discloses a real-time deviation rectifying and rolling scheduling method of a wind-light complementary system coupling wind-light uncertainty, which comprises the steps of determining output prediction deviation of a target period, determining model prediction frequency and the period length of a next prediction period adjacent to the target period, constructing a wind-light output uncertainty interval of the next prediction period according to the model prediction frequency, obtaining a conventional fluctuation interval and an extreme fluctuation interval, matching deviation rectifying strategies of the conventional fluctuation interval and the extreme fluctuation interval according to the period length of the next prediction period, determining a target fluctuation interval corresponding to each moment in the wind-light output uncertainty interval according to actual output of each moment in the next prediction period, and taking the deviation rectifying strategy of the target fluctuation interval corresponding to the moment as the deviation rectifying and rolling scheduling strategy of the moment. The invention belongs to the field of power system policy optimization. The invention improves the running stability and economy of the water-wind-solar complementary system.

Inventors

WANG YONGJIAN
TAN PINGXIANG
HE YUANYING
WU CAIYU
HUANG CONGXIANG
ZHOU MENGYANG
Peng Lequan
Zhang Dongao
YAO QIUYU

Assignees

华电四川发电有限公司宝珠寺水力发电厂
华电四川发电有限公司

Dates

Publication Date: 20260512
Application Date: 20260416

Claims (10)

1. The real-time deviation rectifying rolling scheduling method of the wind-solar complementary system for coupling the wind-solar uncertainty is characterized by comprising the following steps of: Acquiring actual output and model predicted output of the water and wind in a target period, and determining output prediction deviation of the target period according to the actual output and the model predicted output; Determining a model prediction frequency and a period length of a next prediction period adjacent to the target period according to the output prediction bias; According to the model prediction frequency, constructing a water-wind-light output uncertainty interval of the next prediction period, and obtaining a conventional fluctuation interval and an extreme fluctuation interval; According to the period length of the next prediction period, matching correction strategies of the conventional fluctuation interval and the extreme fluctuation interval, wherein the correction strategies comprise hydropower regulation correction, energy storage regulation correction or flexible load joint correction; when the next prediction period is entered, determining a target fluctuation interval corresponding to the moment from the water-wind-light output uncertainty interval according to the actual output of each moment of the next prediction period, and taking a deviation correcting strategy of the target fluctuation interval corresponding to the moment as a deviation correcting rolling scheduling strategy of the moment.
2. The method for real-time deskewing and rolling scheduling of a coupled wind-solar hybrid system according to claim 1, wherein determining the output prediction bias of the target period according to the actual output and the model predicted output comprises: Wherein, the The deviation is predicted for the output of the target period, For the actual output of the target period of time, The output is predicted for the model of the target period.
3. The method for real-time deskewing and rolling scheduling of a coupled wind-solar hybrid system of claim 2, wherein determining a model predictive frequency based on the output predictive bias comprises: Wherein, the The frequency is predicted for the model and, A minimum value is preset for the model predictive frequency, A maximum value is preset for the model predictive frequency, Is a preset empirical deviation threshold.
4. The method for real-time deskewing and rolling scheduling of a coupled wind-solar hybrid system according to claim 2, wherein determining a period length of a next predicted period adjacent to the target period according to the output prediction bias comprises: Wherein, the For the period length of the next predicted period, In order to allow for a minimum length of the time period, In order to allow for a maximum length of time period, Is a preset empirical deviation threshold.
5. The method for real-time deviation rectifying and rolling scheduling of the wind-light complementary system coupled with wind-light uncertainty according to claim 1, wherein the method for constructing the wind-light output uncertainty section of the next prediction period according to model prediction frequency and obtaining a conventional fluctuation section and an extreme fluctuation section comprises the following steps: Obtaining prediction error data from the historical data at the same frequency as the model prediction to construct a set of error samples, comprising: Wherein, the To predict frequency as in history data Is provided for a set of historical prediction error samples, Is the first The deviation of the predicted output of the individual history, Determining an error mean value and a standard deviation according to the error sample set; based on standard deviation, constructing a water-wind-light output uncertainty interval of a next prediction period, which comprises the following steps: Wherein, the , , For the lower boundary of the uncertainty interval of the wind-solar power generation, For the upper boundary of the uncertainty interval of the wind-solar power generation, The output average value is predicted for the model of the next prediction period, In order to pre-set the confidence coefficient, For predicting the frequency as Standard deviation of the historical output prediction deviation; dividing a regular fluctuation interval from an extreme fluctuation interval, comprising: 。
6. the method for real-time deskewing and rolling scheduling of a coupled wind-solar hybrid system according to claim 5, wherein determining an error mean value and determining a standard deviation according to the error sample set comprises: Wherein, the For predicting the frequency as Is a function of the historical average output predicted deviation of (1), For predicting the frequency as Standard deviation of the historical output prediction deviation.
7. The method for real-time deskewing and rolling scheduling of a coupled wind-solar hybrid system according to claim 1, wherein the matching of the deskewing strategies of the regular fluctuation interval and the extreme fluctuation interval according to the period length of the next prediction period comprises: setting a response time, comprising: Wherein, the For the response time of the energy storage system, For the response time of the hydroelectric generating set, Response time for flexible load; for the conventional fluctuation interval, if Matching the stored energy to adjust and rectify the deviation; If it is Matching hydropower adjustment deviation correction; If it is Matching the hydropower adjustment deviation correction and the flexible load combined deviation correction; For extreme fluctuation intervals, if Matching the energy storage adjustment deviation correction with the hydropower adjustment deviation correction; If it is Matching the energy storage adjustment deviation correction with the hydropower adjustment deviation correction; If it is And matching the energy storage adjustment deviation correction, the hydropower adjustment deviation correction and the flexible load joint deviation correction.
8. The method for real-time deskew rolling scheduling of a coupled wind-solar hybrid system of claim 7, wherein generating a deskew strategy matching function comprises: Wherein, the In order to implement the deviation-rectifying strategy, In order to match the function, For the period length of the next predicted period, Is a regular fluctuation interval or an extreme fluctuation interval.
9. The method for real-time deskewing and rolling scheduling of a wind-solar complementary system coupled with wind-solar uncertainty according to claim 1, comprising the following steps: Energy storage adjustment deviation correction: When the actual output is lower than the predicted output of the corresponding model, the energy storage system discharges; When the actual output is higher than the predicted output of the corresponding model, the energy storage system is charged; And (3) hydroelectric adjustment and deviation correction: When the actual output is lower than the predicted output of the corresponding model, the output of the hydroelectric generating set is improved; and when the actual output is higher than the predicted output of the corresponding model, reducing the output of the hydroelectric generating set.
10. The method for real-time deskewing and rolling scheduling of a wind-solar complementary system coupled with wind-solar uncertainty according to claim 1, comprising the following steps: and (3) flexible load combined deviation correction: When the actual output is lower than the predicted output of the corresponding model, the adjustable load is cut down or delayed to use electricity; and when the actual output is higher than the predicted output of the corresponding model, guiding the adjustable load to increase electricity consumption or executing a transferable electricity consumption task.

Description

Real-time deviation correcting rolling scheduling method for wind-light complementary system coupling wind-light uncertainty Technical Field The invention relates to the field of power system strategy optimization, in particular to a real-time deviation correcting rolling scheduling method for a water-wind-light complementary system of coupled wind-light uncertainty. Background With the rapid development of new energy power generation technology, the duty ratio of wind power and photovoltaic power generation in a power system is continuously improved. However, wind power and photovoltaic power generation are greatly influenced by natural factors such as wind speed, solar radiation intensity and the like, and the output has obvious randomness, fluctuation and uncertainty, so that the safety and stability of a power system are challenged. In order to improve the new energy consumption capability, a water-wind-solar complementary system with cooperative operation of hydropower, wind power and photovoltaic power generation is gradually widely applied. However, in the actual running process, the scheduling mode of the traditional water-wind-solar complementary system depends on a fixed prediction period and a day-ahead scheduling plan, and when the actual output of wind and light deviates greatly from a predicted value, the existing scheduling method is difficult to dynamically adjust in time, so that the problems of power unbalance, wind and light abandoning or adjustment efficiency reduction and the like are easily caused. Therefore, the invention provides a real-time deviation correcting rolling scheduling method for a wind-light complementary system of coupling wind-light uncertainty to solve the problems. Disclosure of Invention The invention solves the technical problem of lower adjustment efficiency of the scheduling method of the existing water-wind-light complementary system in the prior art by providing the real-time deviation rectifying rolling scheduling method of the water-wind-light complementary system with the uncertainty of the coupling wind-light, and achieves the technical effect of improving the adjustment efficiency of the water-wind-light complementary system. In a first aspect, the invention provides a real-time deviation correcting rolling scheduling method for a water-wind-solar complementary system of coupled wind-solar uncertainty, which comprises the following steps: Acquiring actual output and model predicted output of the water and wind in a target period, and determining output prediction deviation of the target period according to the actual output and the model predicted output; determining a model prediction frequency and a period length of a next prediction period adjacent to the target period according to the output prediction deviation; according to the model prediction frequency, constructing a water-wind-light output uncertainty interval of the next prediction period, and obtaining a conventional fluctuation interval and an extreme fluctuation interval; According to the period length of the next prediction period, matching correction strategies of a conventional fluctuation interval and an extreme fluctuation interval, wherein the correction strategies comprise hydropower adjustment correction, energy storage adjustment correction or flexible load joint correction; when the next prediction period is entered, determining a target fluctuation interval corresponding to the moment from the water-wind-light output uncertainty interval according to the actual output of each moment of the next prediction period, and taking a deviation correcting strategy of the target fluctuation interval corresponding to the moment as a deviation correcting rolling scheduling strategy of the moment. Further, determining an output prediction deviation of the target period according to the actual output and the model predicted output, including: Wherein, the The deviation is predicted for the output of the target period,For the actual output of the target period of time,The output is predicted for the model of the target period. Further, determining a model predictive frequency from the output predictive bias includes: Wherein, the The frequency is predicted for the model and,A minimum value is preset for the model predictive frequency,A maximum value is preset for the model predictive frequency,Is a preset empirical deviation threshold. Further, determining a period length of a next predicted period adjacent to the target period according to the output prediction bias includes: Wherein, the For the period length of the next predicted period,In order to allow for a minimum length of the time period,In order to allow for a maximum length of time period,Is a preset empirical deviation threshold. Further, according to the model prediction frequency, constructing a water-wind-light output uncertainty interval of a next prediction period, and obtaining a conventional fluctuation interval and an extre