Search

CN-121997016-A - Average wind direction calculation method, average wind direction calculation device, computer equipment and storage medium

CN121997016ACN 121997016 ACN121997016 ACN 121997016ACN-121997016-A

Abstract

The invention discloses an average wind direction calculation method, an average wind direction calculation device, computer equipment and a storage medium. The method comprises the steps of obtaining original wind direction data, carrying out coordinate conversion on each original wind direction data to obtain conversion results, calculating cosine similarity between each conversion result and a north-positive unit vector, carrying out outlier detection on the cosine similarity by using a box diagram method, identifying outliers in the original wind direction data, removing the outliers from the conversion results to obtain intermediate data, calculating an average value of all the intermediate data to obtain an average vector representing the average wind direction, calculating an included angle between the average vector and the north-positive unit vector, adjusting according to quadrants to obtain a final wind direction angle, and outputting the final wind direction angle. By implementing the method, the accuracy of wind energy resource evaluation can be improved, and the technical problems that the accuracy of wind direction data analysis is challenged and the like due to the defects of the traditional box diagram method and arithmetic average method in wind direction data processing are solved.

Inventors

  • YAN FENGWEI

Assignees

  • 浙江同济科技职业学院

Dates

Publication Date
20260508
Application Date
20250530

Claims (10)

  1. 1. The average wind direction calculation method is characterized by comprising the following steps: Acquiring original wind direction data; performing coordinate transformation on each piece of original wind direction data to obtain a transformation result; calculating cosine similarity between each conversion result and the unit vector in the north-positive direction; detecting abnormal values of the cosine similarity by using a box diagram method, identifying the abnormal values in the original wind direction data, and removing the abnormal values from the conversion result to obtain intermediate data; Calculating the average value of all the intermediate data to obtain an average vector representing the average wind direction; Calculating an included angle between the average vector and the north direction, and adjusting according to quadrants to obtain a final wind direction angle; and outputting the final wind direction angle.
  2. 2. The method of calculating an average wind direction according to claim 1, wherein said performing coordinate transformation on each of said raw wind direction data to obtain a transformation result comprises: And converting the original wind direction data into a unit vector in a rectangular coordinate system to obtain a conversion result.
  3. 3. The method of calculating an average wind direction according to claim 1, wherein the detecting the cosine similarity by using a box-line graph method, identifying an outlier in the original wind direction data, and removing the outlier from the conversion result to obtain intermediate data, includes: And combining the cosine similarities to form a constituting sequence, detecting abnormal values of values in the sequence by using a box diagram method to obtain an abnormal value sequence, and removing the abnormal value sequence from the conversion result to obtain intermediate data.
  4. 4. The method of calculating an average wind direction according to claim 1, wherein said averaging all of said intermediate data to obtain an average vector representing an average wind direction comprises: and carrying out average value calculation on the horizontal coordinate and the vertical coordinate on the intermediate data to obtain an average vector representing the average wind direction.
  5. 5. The method of claim 1, wherein calculating the angle between the average vector and the north direction and adjusting according to quadrants to obtain the final wind direction angle comprises: And when the average vector is positioned in a third quadrant or a fourth quadrant, converting the included angle into a correct wind direction angle range so as to obtain a final wind direction angle.
  6. 6. An average wind direction calculation device, comprising: The data acquisition unit is used for acquiring original wind direction data; The coordinate conversion unit is used for carrying out coordinate conversion on each piece of original wind direction data so as to obtain a conversion result; a similarity calculation unit, configured to calculate cosine similarity between each conversion result and the north direction unit vector; The rejecting unit is used for detecting abnormal values of the cosine similarity by using a box diagram method, identifying abnormal values in the original wind direction data, and rejecting the abnormal values from the conversion result to obtain intermediate data; The average value calculation unit is used for calculating the average value of all the intermediate data to obtain an average vector representing the average wind direction; The angle determining unit is used for calculating the included angle between the average vector and the north direction and adjusting according to quadrants to obtain a final wind direction angle; and the output unit is used for outputting the final wind direction angle.
  7. 7. The apparatus according to claim 6, wherein the conversion unit is configured to convert the raw wind direction data into a unit vector in a rectangular coordinate system to obtain a conversion result.
  8. 8. The apparatus according to claim 6, wherein the rejecting unit is configured to combine the cosine similarities to form a sequence, and perform outlier detection on values in the sequence using a box-line graph method to obtain an outlier sequence, and reject the outlier sequence from the conversion result to obtain intermediate data.
  9. 9. A computer device, characterized in that it comprises a memory on which a computer program is stored and a processor which, when executing the computer program, implements the method according to any of claims 1-6.
  10. 10. A storage medium storing a computer program which, when executed by a processor, implements the method of any one of claims 1 to 6.

Description

Average wind direction calculation method, average wind direction calculation device, computer equipment and storage medium Technical Field The invention relates to a wind energy resource evaluation method of a wind power plant, in particular to an average wind direction calculation method, an average wind direction calculation device, computer equipment and a storage medium. Background In the early stage of wind farm construction, accurate assessment of wind energy resources around a site is a vital task. Detailed analysis of anemometry data is required in the evaluation process, wherein the detailed analysis comprises evaluation of a plurality of core indexes such as 10-minute average wind speed and 10-minute average wind direction. The accuracy of these data directly affects the reliability of key indicators such as subsequent correlation analysis, turbulence intensity calculation, etc. The wind speed data is collected as instantaneous wind speed in meters per second at each test time point, with data ranging typically between 0 and 40 meters per second. And the wind direction data record that each test time point is based on the clockwise rotation angle of the north direction, the unit is degree, and the data range is 0 to 359 degrees. In the conventional calculation method of wind speed and wind direction, wind speed data is usually calculated by arithmetic average, and is suitable for numerical processing. However, the wind direction data is different from the wind speed data, which is an angle data representing the direction of wind, and has a value in a circumferential range of 0 to 359 degrees. Because of the special nature of wind direction, directly calculating the average wind direction by adopting an arithmetic average method may cause larger deviation, and influence the subsequent analysis result. For example, when the wind direction data is distributed on both sides of the true north direction, the wind direction values like the north-northeast direction are 0 to 45 degrees, and the wind direction values in the north-northwest direction are 345 to 359 degrees, and although there is a large difference in these wind direction values, they represent the wind direction that is actually close to the true north direction. If the wind direction is calculated directly using arithmetic averaging, a result deviating from the actual wind direction will be obtained. For example, for a set of wind direction data (350,355,357,5,15,23,28,31,40,43), the dominant wind direction is north-northeast direction, and if the result obtained by directly using arithmetic average is 124.7 degrees, the dominant wind direction is east-southeast direction, and there is a large deviation from the actual dominant wind direction, so that significant errors occur in the analysis result. In addition, in the preprocessing process of wind direction data, abnormal value detection plays a crucial role. The conventional abnormal value detection method, such as a box diagram method, mainly judges the abnormal value in the data set through a statistical method. However, simply determining whether an outlier is based on the magnitude of the value may produce a large deviation due to the circumferential nature of the wind direction data. For example, the 15 degrees differ from 345 degrees by a large amount, but in practice they represent a wind direction very close to the north direction. Therefore, when the box-line diagram method is used to detect an abnormal value of wind direction data, some of the actual normal wind directions may be misjudged as the abnormal value, or the actual abnormal value may be missed. In one set of wind direction data (7,8,11,14,21,25,199,201,202,203,208,212,315,316,319,325,326,329,332,335,336,339,341,343,345,346,347,348,349,350,354,357), for example, the dominant wind direction is north west, the conventional box graph method may misjudge the north-northeast direction (7,8,11,14,21,25 degrees) as an outlier, and miss-judge the outlier (199,201,202,203,208,212 degrees) located in the south-southwest direction. Such misjudgment and missed judgment can seriously affect the accuracy of wind direction data, and further affect the calculation and subsequent analysis of the average wind direction. In combination, the shortcomings of the traditional box graph method and the arithmetic average method in wind direction data processing result in great challenges for the accuracy of wind direction data analysis. In the early stage of wind farm construction, in order to ensure the accuracy of wind data, especially the analysis accuracy of wind direction data, it is highly desirable to develop more accurate and efficient algorithms to solve these existing problems, thereby improving the accuracy of wind energy resource assessment and providing reliable data support for subsequent wind farm design and operation. Therefore, a new method is necessary to be designed, so that the accuracy of wind energy resource assessment