Search

CN-117290701-B - Climbing event prediction method and device, electronic equipment and storage medium

CN117290701BCN 117290701 BCN117290701 BCN 117290701BCN-117290701-B

Abstract

The disclosure provides a climbing event prediction method, a climbing event prediction device, electronic equipment and a storage medium, relates to the technical field of computers, and particularly relates to the field of power detection. The method comprises the steps of sliding time windows on a wind power sequence one by one and intercepting sequences in the time windows to obtain a plurality of wind power subsequences, determining a plurality of first subsequences belonging to a climbing event in the plurality of wind power subsequences, respectively determining neighbor subsequences of each first subsequence based on Euclidean distances between each first subsequence and other subsequences in the plurality of first subsequences, determining an occurrence time interval of the climbing event based on sampling time of wind power in the neighbor subsequences of each first subsequence, and determining the next occurrence time of the climbing event based on the last occurrence time and the occurrence time interval of the climbing event. By adopting the technical scheme disclosed by the invention, the prediction accuracy of the power climbing event can be improved.

Inventors

  • HE YINGJING
  • Wang Cenfeng
  • ZHU KEPING
  • WANG LEI
  • SUN FEIFEI

Assignees

  • 国网浙江省电力有限公司经济技术研究院
  • 国网浙江省电力有限公司

Dates

Publication Date
20260505
Application Date
20230919

Claims (7)

  1. 1. A method for predicting a hill climbing event, the method being based on wind turbines in mountain and coastal microclimate scenes, comprising: acquiring output power data of a wind turbine generator, forming a wind power sequence arranged in time sequence, sliding a time window of the wind power sequence one by one in power, and intercepting sequences in the time window to obtain a plurality of wind power subsequences; Determining a plurality of first subsequences belonging to a climbing event in the plurality of wind power subsequences, wherein the climbing event meets the conditions that the difference value between the maximum value and the minimum value of the wind power in the plurality of first subsequences is larger than the power threshold value of rated installed capacity of the wind turbine, determining neighbor subsequences of the first subsequences respectively based on Euclidean distances between the first subsequences and other subsequences in the plurality of first subsequences, and determining other subsequences which are not overlapped with the first subsequences in time in the plurality of first subsequences as neighbor subsequences; determining an occurrence time interval of the climbing event based on sampling times of wind power in neighbor subsequences of each first subsequence; determining a next occurrence time of the climbing event based on a last occurrence time of the climbing event and the occurrence time interval; determining a distance representation of each of the first sub-sequences based on Euclidean distances between each of the first sub-sequences and other sub-sequences in the plurality of first sub-sequences; Selecting a first subsequence with the smallest distance picture from the first subsequences as a main mode of the climbing event; Based on the primary mode of the climbing event, adjusting a climbing duration and a climbing amplitude of the climbing event; The determining the occurrence time interval of the climbing event based on the sampling time of the wind power in the neighbor subsequence corresponding to each first subsequence includes: determining neighbor sampling time of each first sub-sequence based on sampling time of wind power in neighbor sub-sequence of each first sub-sequence; determining an occurrence time interval of the climbing event based on neighbor sampling time of each first sub-sequence; The determining, based on the neighbor sampling time of each of the first sub-sequences, an occurrence time interval of the climbing event includes: The neighbor sampling time of each first sub-sequence is arranged according to the time sequence, and a neighbor sampling time sequence is obtained; Starting from the initial value of s being 1, obtaining the (s+1) th neighbor sampling time and the(s) th time interval between the(s) th neighbor sampling time in the neighbor sampling time sequence one by one to obtain a time interval sequence, wherein s is a positive integer; fitting the third-order polynomial by taking any time interval in the time interval sequence as an independent variable and a dependent variable of the third-order polynomial to obtain coefficients of each item in the third-order polynomial; and solving the third-order polynomial by using coefficients of each item in the third-order polynomial to obtain the occurrence time interval of the climbing event.
  2. 2. The method of claim 1, wherein the determining the neighbor subsequences of each of the first subsequences based on euclidean distances between each of the first subsequences and other subsequences of the plurality of first subsequences, respectively, comprises: Determining, for any one of the first subsequences, a plurality of second subsequences that do not overlap with the first subsequence among the plurality of first subsequences; And selecting a second subsequence with the smallest Euclidean distance with the first subsequence from the plurality of second subsequences based on the Euclidean distance between the first subsequence and each second subsequence, and taking the second subsequence with the smallest Euclidean distance with the first subsequence as a neighbor subsequence of the first subsequence.
  3. 3. The method according to claim 1, wherein determining the neighbor sampling time of each of the first sub-sequences based on the sampling time of the wind power in the neighbor sub-sequence corresponding to each of the first sub-sequences, respectively, comprises one of: For any one of the first subsequences, under the condition that the first subsequence corresponds to one of the neighbor subsequences, taking the sampling time of the 1 st wind power in the neighbor subsequences as the neighbor sampling time of the first subsequence; And aiming at any first subsequence, under the condition of a plurality of neighbor subsequences corresponding to the first subsequence, acquiring the sampling time of the 1 st wind power in each neighbor subsequence to obtain a plurality of sampling times, and selecting the sampling time with the smallest time interval between the sampling time of the 1 st wind power in the first subsequence from the plurality of sampling times as the neighbor sampling time of the first subsequence.
  4. 4. A hill climbing event prediction device, characterized in that the device is based on wind turbines in mountain and coastal microclimate scenes, comprising: The sub-sequence intercepting module is used for acquiring output power data of the wind turbine generator to form wind power sequences arranged in time sequence, sliding time windows of the wind power sequences one by one in power mode, intercepting sequences in the time windows and obtaining a plurality of wind power sub-sequences; The system comprises a climbing event detection module, a wind turbine generator system control module and a wind turbine generator system control module, wherein the climbing event detection module is used for determining a plurality of first subsequences belonging to a climbing event in the plurality of wind power subsequences, and the condition that the climbing event meets comprises that the difference value between the maximum value and the minimum value of the wind power in the plurality of first subsequences is larger than the power threshold value of the rated installed capacity of the wind turbine generator system; a subsequence screening module, configured to determine, based on euclidean distances between each first subsequence and other subsequences in the plurality of first subsequences, neighbor subsequences of each first subsequence, respectively, including determining, for each first subsequence, other subsequences in the plurality of first subsequences that do not overlap with the first subsequence in time, and selecting a subsequence with a minimum euclidean distance as a neighbor subsequence; The time interval determining module is used for determining the occurrence time interval of the climbing event based on the sampling time of the wind power in the neighbor subsequence of each first subsequence; a climbing event prediction module, configured to determine a next occurrence time of the climbing event based on a last occurrence time of the climbing event and the occurrence time interval; A module for determining a distance portraits of the first sub-sequences based on euclidean distances between the first sub-sequences and other sub-sequences in the first sub-sequences, respectively, selecting a first sub-sequence with the smallest distance portraits from the first sub-sequences as a main mode of the climbing event, and adjusting the climbing duration and the climbing amplitude of the climbing event based on the main mode of the climbing event; The time interval determining module is further used for determining neighbor sampling time of each first sub-sequence based on sampling time of wind power in the neighbor sub-sequence of each first sub-sequence; The time interval determining module is further used for arranging neighbor sampling time of each first subsequence according to time sequence to obtain a neighbor sampling time sequence, obtaining the (s+1) th time interval between the(s) th neighbor sampling time and the(s) th neighbor sampling time in the neighbor sampling time sequence one by one from the initial value of s being 1 to obtain a time interval sequence, wherein s is a positive integer, fitting a third-order polynomial according to any time interval in the time interval sequence as an independent variable and a dependent variable of the third-order polynomial to obtain coefficients of each item in the third-order polynomial, and solving the third-order polynomial according to the coefficients of each item in the third-order polynomial to obtain the occurrence time interval of the climbing event.
  5. 5. The apparatus of claim 4, wherein the subsequence screening module comprises: a first screening unit configured to determine, for any one of the first subsequences, a plurality of second subsequences that do not overlap with the first subsequence, among the plurality of first subsequences; And the second screening unit is used for selecting a second subsequence with the smallest Euclidean distance with the first subsequence from the plurality of second subsequences based on the Euclidean distance between the first subsequence and each second subsequence, and taking the second subsequence with the smallest Euclidean distance with the first subsequence as a neighbor subsequence of the first subsequence.
  6. 6. An electronic device, comprising: at least one processor, and A memory communicatively coupled to the at least one processor; Wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-3.
  7. 7. A non-transitory computer readable storage medium storing computer instructions, wherein the computer instructions are for causing the computer to perform the method of any one of claims 1-3.

Description

Climbing event prediction method and device, electronic equipment and storage medium Technical Field The present disclosure relates to the field of computer technology, and in particular, to the field of power detection. The disclosure relates to a method, a device, an electronic device and a storage medium for predicting a climbing event. Background In order to respond to the development strategy of tightly implementing the carbon peak and the carbon neutralization, the development direction of 'full-force promotion of new energy development and promotion of clean energy supply' is provided. The renewable energy source power generation installed capacity reaches the proportion of the total power generation installed capacity to be larger and larger. However, due to the influence of the random and fluctuation of the clean energy source, the access of wind power and other renewable energy sources with high proportion to generate electricity brings higher requirements to the auxiliary service of the power system. According to statistics, the wind discarding power is higher and higher. The wind power climbing event has serious influence on the power system. Particularly, in microclimate scenes such as mountain areas, coasts and the like, the wind power output can be steep rise and fall, and a high-risk climbing event is formed, so that the supply and demand balance of a power system is affected. The climbing type of the climbing event may be classified into an ascending climbing type and a descending climbing type. A hill climb event has 3 important attributes, hill climb amplitude, hill climb slope and hill climb duration, and is typically characterized by a high amplitude, short duration. Therefore, the rapid change of the wind power climbing event is difficult to accurately predict by adopting the traditional power generation prediction technology, and the timely grid connection and the digestion of new energy cannot be ensured. How to accurately establish the statistical characteristics and the prediction model of the wind power climbing event so as to better provide data support for auxiliary services of a power system is an important subject of our research. Different from large-range and long-period wind power fluctuation caused by extreme weather, the actual requirements for identifying short-time and small-range climbing events are required to be met in microclimate scenes such as mountain areas, coasts and the like. However, the prior art has difficulty in solving the technical problem. Disclosure of Invention The present disclosure provides a method, an apparatus, an electronic device, and a storage medium for predicting a climbing event, which can solve the above-mentioned problems. According to an aspect of the present disclosure, there is provided a method for predicting a climbing event, including: For the wind power sequences arranged according to the sampling time sequence, sliding a time window power by power and intercepting the sequences in the time window to obtain a plurality of wind power subsequences; determining a plurality of first subsequences belonging to a climbing event in the plurality of wind power subsequences; Determining neighbor subsequences of each first subsequence based on Euclidean distances between each first subsequence and other subsequences in the plurality of first subsequences, respectively; Determining an occurrence time interval of the climbing event based on sampling time of wind power in a neighbor subsequence corresponding to each first subsequence; And determining the next occurrence time of the climbing event based on the last occurrence time of the climbing event and the occurrence time interval. According to another aspect of the present disclosure, there is provided a hill climbing event prediction apparatus including: The sub-sequence intercepting module is used for sliding time windows on a power-by-power basis for the wind power sequences arranged according to the sampling time sequence and intercepting sequences in the time windows to obtain a plurality of wind power sub-sequences; The climbing event detection module is used for determining a plurality of first subsequences belonging to the climbing event in the plurality of wind power subsequences; A subsequence screening module, configured to determine neighbor subsequences of each first subsequence based on euclidean distances between each first subsequence and other subsequences in the plurality of first subsequences, respectively; The time interval determining module is used for determining the occurrence time interval of the climbing event based on the sampling time of the wind power in the neighbor subsequence of each first subsequence; and the climbing event prediction module is used for determining the next occurrence time of the climbing event based on the latest occurrence time of the climbing event and the occurrence time interval. According to another aspect of the present disclosure, the