A method for identifying extreme coherent gusts based on equivalent vector wind speed
By using an identification method based on equivalent vector wind speed, which simultaneously considers gust amplitude, wind direction angle change, and rise time, the problem of discrepancies between the identification results and actual extreme operating conditions in existing technologies is solved, enabling more accurate wind turbine load assessment and design optimization.
Patent Information
- Application Number
- CN202611349939.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-02
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2046-09-02
AI Technical Summary
Existing methods for identifying extreme coherent gusts fail to simultaneously consider three major variables: gust amplitude, change in wind direction angle, and rise time. As a result, they cannot accurately reflect the synchronous abrupt changes in wind speed and direction within the wind turbine plane, leading to identification results that are seriously inconsistent with the actual load characteristics under extreme operating conditions.
An identification method based on equivalent vector wind speed is adopted. By using traditional mechanical/ultrasonic anemometers on remote sensing wind equipment or wind turbine towers and blades, wind speed and wind direction data at multiple height levels within the wind turbine sweep range are collected simultaneously. The equivalent wind speed, equivalent vector wind speed and wind direction of the X-axis and Y-axis components are calculated. Combined with high-pass filtering and continuous wavelet transform, the gust amplitude, rise time and wind direction change of extreme gust events are extracted.
The identified extreme coherent gusts can cover the entire rotor plane, and the identification results are more consistent with the extreme load characteristics actually borne by the wind turbine, improving the accuracy of load analysis and the sensitivity of the identification algorithm. This adapts to the complex wind field characteristics in my country and improves the design economy and operational safety of the wind turbine.
Smart Images

Figure CN122862891B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of wind power technology and relates to a method for identifying extreme coherent gusts based on equivalent vector wind speed. Background Technology
[0002] To ensure the safety and economy of wind turbine design, it is necessary to rationally determine the values of gust amplitude, wind direction angle change, and rise time in the extreme coherent gust (ECD) model during the wind turbine design process. Currently, the ECD model parameters in the IEC 61400-1 standard are mainly calibrated based on wind resource data from Europe and the United States. However, wind power development in my country faces complex and diverse application scenarios, including plains, offshore areas, and mountainous regions, with significant differences in wind field characteristics. Directly applying this standard model often leads to turbine designs that are either overly aggressive or overly conservative, making it difficult to balance safety and economy. Therefore, there is an urgent need for identification methods adapted to local wind conditions.
[0003] Currently, methods for identifying extreme coherent gusts mainly fall into two categories: gust identification based on the Gust Direction Index (GDI) and gust identification based on raw wind speed (Hannesdóttir's method). However, both have inherent flaws. The GDI-based gust identification method has gained widespread acceptance among domestic experts and scholars. However, this method predefines the rise time variable, rather than automatically identifying and filtering it through actual gusts. Therefore, the identified extreme coherent gusts may not cover the entire wind turbine plane; that is, the gust amplitude and wind direction angle change at all spatial points within the wind turbine plane cannot be consistent. The raw wind speed-based gust identification method also has some problems. First, because this method only considers the single variable of gust amplitude and does not consider the wind direction angle change and rise time variables, its standard deviation is large compared to the average value of the wind direction angle change and rise time. Therefore, it produces a large number of data points with wind direction angle changes of only a few degrees and rise times exceeding 600 seconds. Clearly, such data is insufficient to represent extreme coherent gusts. Secondly, this gust identification method selects low-frequency extreme gust events with large standard deviations during filtering, while high-frequency extreme gust events with large standard deviations that exhibit the same trend across the entire wind turbine plane are filtered out. Finally, this gust identification method does not consider whether the time points of the maximum and minimum values coincide with the time points of wind speed changes, nor does it consider the synchronous changes in wind speed and direction.
[0004] In summary, the existing methods generally fail to verify the temporal consistency between extreme wind speeds and wind direction changes, neglecting the synchronous coupling relationship between wind speed and wind direction. This leads to significant deviations between the identification results and the actual load characteristics of the wind turbine. Therefore, how to construct an identification method that can simultaneously consider three major variables—gust amplitude, wind direction angle change, and rise time—and truly reflect the coordinated abrupt changes across the entire wind turbine plane is a key issue that urgently needs to be addressed for accurate load assessment of wind turbines. Summary of the Invention
[0005] In view of this, in order to solve the problem that existing extreme coherent gust identification methods generally fail to simultaneously consider the three major variables of gust amplitude, wind direction angle change, and rise time, and cannot accurately reflect the synchronous change characteristics of wind speed and wind direction within the wind turbine plane, resulting in serious discrepancies between the identification results and the actual load characteristics under extreme operating conditions, this invention provides an extreme coherent gust identification method based on equivalent vector wind speed. This method can identify the gust amplitude, wind direction angle change, and rise time variables of extreme coherent gusts, and the identified extreme coherent gusts can cover the entire wind turbine plane.
[0006] To achieve the above objectives, the present invention provides the following technical solution:
[0007] An extreme coherent gust identification method based on equivalent vector wind speed includes the following steps:
[0008] S1. Obtain measured wind speed and wind direction angle data at multiple representative heights within the wind turbine plane;
[0009] Specifically, remote sensing wind sensing equipment (such as lidar) or traditional mechanical / ultrasonic anemometers installed on the wind turbine tower and blades are used to synchronously collect raw measured time-series data at at least three different height levels (e.g., hub height and multiple height points symmetrically distributed above and below) within the wind turbine's sweep range at a preset sampling frequency (preferably 1Hz). The measured data includes horizontal wind speed scalars at each height. and horizontal wind angle During the data acquisition process, missing and outlier values caused by equipment failure or environmental interference are removed to ensure the continuity and validity of the input data.
[0010] S2, Based on the wind speed at various heights within the wind turbine plane in step S1 The equivalent wind speed is calculated based on the area weight of the wind turbine plane. Equivalent vector wind speed of X-axis component Equivalent vector wind speed of Y-axis component And based on this, the equivalent vector wind direction can be calculated. ;
[0011] Specifically, equivalent wind speed The calculation formula is: Equivalent vector wind speed on the X-axis component The calculation formula is: Equivalent vector wind speed on the Y-axis component The calculation formula is: ; Let be the area of the entire wind turbine plane, where n is the radius of the wind turbine;h Let n be the number of wind speed data points at different measurement heights within the range of the wind turbine height, and n h ≥3; n h The measurement heights are arranged from low to high. The midpoint between two adjacent measurement heights serves as the boundary between adjacent wind turbine plane regions. The lower boundary of the region corresponding to the lowest measurement height is the lower boundary of the wind turbine sweep range, and the upper boundary of the region corresponding to the highest measurement height is the upper boundary of the wind turbine sweep range. i Let i be the area of the wind turbine plane region corresponding to the i-th measurement height, i=1,2,…,n h Each A i Wind speed at various altitudes One-to-one correspondence, and all A's. i The sum of the areas equals A; for the remaining areas above the highest measurement height and below the lowest measurement height, which are still within the sweep range of the wind turbine, their areas are respectively included in the area A corresponding to the highest and lowest measurement heights. i The wind speed at the corresponding boundary measurement height was measured respectively. Participate in equivalent wind speed calculation; and The first Wind speed measured at a certain altitude The components on the X and Y axes, among which , , The horizontal wind direction angle measured at the i-th altitude;
[0012] based on and The symbol, equivalent vector wind direction The equivalent wind direction obtained from the arctangent function is mapped to the 0~360° range in vector form, specifically through segmented calculations as follows:
[0013] when hour, ;
[0014] when hour, ;
[0015] when hour, ;
[0016] when hour, .
[0017] S3, Equivalent wind speed in step S2 The upward translation process is performed to obtain the equivalent vector wind speed of the X-axis component after the upward translation. Equivalent vector wind speed of Y-axis component X-axis reverse equivalent vector wind speed and Y-axis reverse equivalent vector wind speed ;
[0018] Specifically, the equivalent wind speed calculated based on step S2 and equivalent vector wind direction Calculate the equivalent vector wind speed on the X-axis component after upward translation. The calculation formula is: The equivalent vector wind speed on the Y-axis component after upward translation The calculation formula is: The reverse equivalent vector wind speed on the X-axis component after upward translation The calculation formula is: The reverse equivalent vector wind speed on the Y-axis component after upward translation The calculation formula is: .
[0019] S4. Equivalent vector wind speed after translation in step S3 is calculated for a preset duration. , , , The process is segmented, high-pass filtered, and the sum of the standard deviations of wind speed before and after filtering is calculated. Extreme gust events are extracted according to a preset screening ratio;
[0020] The high-pass filter uses a Butterworth second-order high-pass filter with a cutoff frequency of [missing information]. By turbulent integral scale Determined, that is ,in For turbulent integral scale, The measured average wind speed; extreme gust events were selected using... Sort by magnitude from largest to smallest, and take the top 1 / 4000 as extreme gust events. The standard deviation of the wind speed data before filtering. This represents the standard deviation of the filtered wind speed data.
[0021] S5. Perform continuous wavelet transform on the extreme gust events extracted in step S4, fit the error function curve, and extract the gust amplitude, rise time and wind direction angle change to complete the identification of extreme coherent gusts.
[0022] The continuous wavelet transform employs a mirrored first-order Gaussian wavelet: , Here is the normalization constant; the formula for fitting the error function curve based on the time and position parameters is: In the formula: The wind speed prior to the escalation of an extreme gust event; This refers to the wind speed following an extreme gust event; It is a time variable; τ is the local feature position parameter of the gust obtained by continuous wavelet transform, used to characterize the center time position of the gust's upward phase; τ is the scale parameter obtained by continuous wavelet transform, used to characterize the time scale of the gust's upward process. The error function is used; the extracted gust amplitude is... The rise time is from the wind speed. arrive The time elapsed; where 0.0125 and 0.9875 are the lower and upper limits of the normalized response of the error function, respectively. According to the mathematical properties of the error function, the rise time determined by the above two normalized response thresholds is approximately 3.17 times the scale parameter τ, which is consistent with the definition of using approximately 3 times the scale parameter to characterize the rise time of gusts; the change in wind direction angle is the change in the prevailing wind direction during the corresponding rise time period.
[0023] An extreme coherent gust identification system based on equivalent vector wind speed includes:
[0024] The data acquisition module is used to acquire measured wind speed and wind direction angle data at multiple representative heights within the wind turbine plane;
[0025] The equivalent calculation module is used to calculate the equivalent wind speed based on the wind speed at various heights within the wind turbine plane and the area weight of the wind turbine plane. Equivalent vector wind speed of X-axis component Equivalent vector wind speed of Y-axis component and equivalent vector wind direction ;
[0026] Translation filter module, used for equivalent vector wind speed The upward translation process is performed to obtain the equivalent vector wind speed of the X-axis component after the upward translation. Equivalent vector wind speed of Y-axis component X-axis reverse equivalent vector wind speed and Y-axis reverse equivalent vector wind speed After segmenting into segments with a preset duration, high-pass filtering is performed based on... Extracting extreme gust events;
[0027] The identification module is used to perform continuous wavelet transform on extreme gust events to extract gust amplitude, rise time, and wind direction angle change.
[0028] A computing device includes a processor and a memory, the memory storing computer program instructions, and the processor executing the computer program instructions to implement the above-mentioned extreme coherent gust identification method based on equivalent vector wind speed.
[0029] A computer-readable storage medium storing computer program instructions, which, when executed by a processor, implement the aforementioned method for identifying extreme coherent gusts based on equivalent vector wind speed.
[0030] The beneficial effects of this invention are as follows:
[0031] 1. The extreme coherent gust identification method based on equivalent vector wind speed disclosed in this invention solves the problem that traditional single-point wind speed identification cannot represent the characteristics of the entire field by equivalently synthesizing multi-height wind speed vectors according to the wind turbine plane area weight. Through wavelet transform and fitting verification, the gust events identified by this method can ensure that abrupt changes in wind speed and direction occur synchronously at each representative height layer within the wind turbine plane, making the identification results more consistent with the actual extreme load characteristics borne by the wind turbine and significantly improving the accuracy of load analysis.
[0032] 2. The extreme coherent gust identification method based on equivalent vector wind speed disclosed in this invention breaks through the limitation of existing methods that only focus on gust amplitude, and simultaneously realizes the automated identification and extraction of gust amplitude, wind direction angle change, and rise time. Compared with traditional methods, the gusts screened by this method have the characteristics of "larger amplitude, more drastic wind direction change, and shorter rise time" (as shown in the example, the average rise time is shortened from 232s to 66.7s, and the average wind direction angle change is increased from 15.8° to 91.7°), effectively eliminating a large number of unqualified data with weak wind direction changes or excessively long rise times, and is more in line with the physical definition of extreme coherent gusts in the IEC standard.
[0033] 3. The extreme coherent gust identification method based on equivalent vector wind speed disclosed in this invention adopts... The combined screening criteria and Butterworth high-pass filtering strategy overcome the shortcomings of traditional methods that tend to select low-frequency, large-standard-deviation events. This method can effectively retain high-frequency, large-standard-deviation gust events that reflect the overall coordinated change trend of the wind turbine plane, avoiding the misscreening of potentially destructive real extreme gusts and improving the sensitivity and reliability of the identification algorithm.
[0034] 4. The extreme coherent gust identification method based on equivalent vector wind speed disclosed in this invention, compared to directly applying IEC standard parameters based on European and American wind conditions, uses measured lidar data for localized identification, making it adaptable to the complex wind field characteristics of plains, offshore areas, and mountainous regions in my country. This method provides more accurate input conditions for wind turbine site load adaptability assessment and overall turbine structure optimization design, helping to avoid over-design while ensuring safety, and significantly improving the design economy and operational safety of wind turbines.
[0035] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description
[0036] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:
[0037] Figure 1 This is a flowchart of the extreme coherent gust identification method based on equivalent vector wind speed of the present invention;
[0038] Figure 2 The wind speed time history fitting curve and wavelet transform time-frequency heatmap are shown, among which... Figure 2 (a) is a graph showing the wind speed time history curve and the function fitting. Figure 2 (b) is a time-frequency heatmap of continuous wavelet transform;
[0039] Figure 3 For Case 1, the relative frequency distribution histograms of three variables—gust amplitude, change in wind direction angle, and rise time—are compared between the two methods. Figure 3 (a), (b), and (c) are histograms showing the relative frequency distribution of gust amplitude, wind direction angle change, and rise time under the two methods in Case 1, respectively. Red represents the gust identification method based on the original wind speed, and blue represents the gust identification method proposed in this invention.
[0040] Figure 4 For Case 2, the relative frequency distribution histograms of three variables—gust amplitude, change in wind direction angle, and rise time—are compared under the two methods. Figure 4 (a), (b), and (c) are histograms showing the relative frequency distribution of gust amplitude, wind direction angle change, and rise time under the two methods in Case 2, respectively. Red represents the gust identification method based on the original wind speed, and blue represents the gust identification method proposed in this invention.
[0041] Figure 5 Comparison of wind speed and direction at different altitudes at the site Figure 1 ,in Figure 5 (a) and (b) are three-dimensional time history comparison diagrams of wind speed and wind direction at different heights at the site.
[0042] Figure 6 Comparison of wind speed and direction at different altitudes at the site Figure 2 ,in Figure 6 (a) and (b) are time history comparison diagrams of two-dimensional sub-maps of wind speed and wind direction at different heights at the site.
[0043] Figure 7 The following is a time history broken line graph of equivalent wind speed and equivalent wind direction at site 1, where Figure 7 (a) and (b) are two-dimensional line graphs of the time histories of equivalent wind speed and equivalent wind direction at the site, respectively.
[0044] Figure 8 Comparison of wind speed and direction at different heights at Site 2 Figure 1 ,in Figure 8 (a) and (b) are three-dimensional time history comparison diagrams of wind speed and wind direction at different heights at Site 2, respectively;
[0045] Figure 9 Comparison of wind speed and direction at different heights at Site 2 Figure 2 ,in Figure 9 (a) and (b) are time history comparison diagrams of two-dimensional sub-graphs of wind speed and wind direction at different heights at Site 2, respectively;
[0046] Figure 10 The following is a time history broken line graph of the equivalent wind speed and equivalent wind direction at site 2, where Figure 10 (a) and (b) are two-dimensional line graphs of the time histories of equivalent wind speed and equivalent wind direction at site 2, respectively. Detailed Implementation
[0047] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.
[0048] like Figure 1 The extreme coherent gust identification method based on equivalent vector wind speed shown includes the following steps:
[0049] S1. Obtain the measured wind speed and wind direction angle data at multiple representative heights within the wind turbine plane and perform equivalent calculations.
[0050] Specifically, remote sensing wind sensing equipment (such as lidar) or traditional mechanical / ultrasonic anemometers installed on the wind turbine tower and blades are used to synchronously collect raw time-series data at at least three different height levels (e.g., hub height and multiple height points symmetrically distributed above and below) within the wind turbine's sweep range at a preset sampling frequency (preferably 1Hz). The measured data includes horizontal wind speed scalars at each height. and horizontal wind angle During the data acquisition process, missing and outlier values caused by equipment failure or environmental interference are removed to ensure the continuity and validity of the input data.
[0051] S2. Based on the wind speed at each height within the wind turbine plane and the area weight of the wind turbine plane, calculate the equivalent vector wind speed of the X-axis component. Equivalent vector wind speed of Y-axis component Based on this, the equivalent wind speed can be calculated. and equivalent vector wind direction ;
[0052] Specifically, S21, solve for the equivalent wind speed. ;
[0053] The IEC 61400-12-1:2022 standard introduces the concept of equivalent wind speed in the wind turbine plane, which is the wind speed corresponding to the kinetic energy flux passing through the swept area of the wind turbine, considering the variation of wind speed with height. Equivalent wind speed The calculation formula is shown in equation (1):
[0054] (1)
[0055] In the formula: n is the equivalent wind speed; h V represents the number of wind speed data points at different heights within the rotor height range. i Let A be the wind speed measured at the i-th height; A = πR² is the area of the entire wind turbine plane, where R is the radius of the wind turbine; A i Let i be the area of the wind turbine plane region corresponding to the i-th measurement height, i=1,2,…,n h n h Let n be the number of wind speed data points at different measurement heights within the range of the wind turbine height, and n h ≥3; n h The measurement heights are arranged from low to high. The midpoint between two adjacent measurement heights serves as the boundary between adjacent wind turbine plane regions. The lower boundary of the region corresponding to the lowest measurement height is the lower boundary of the wind turbine sweep range, and the upper boundary of the region corresponding to the highest measurement height is the upper boundary of the wind turbine sweep range. i Let i be the area of the wind turbine plane region corresponding to the i-th measurement height, i=1,2,…,n h Each A i Wind speed at various altitudes One-to-one correspondence, and all A's. i The sum of the areas equals A; for the remaining areas above the highest measurement height and below the lowest measurement height, which are still within the sweep range of the wind turbine, their areas are respectively included in the area A corresponding to the highest and lowest measurement heights. i The wind speed at the corresponding boundary measurement height was measured respectively. Involved in the calculation of equivalent wind speed; derived from equations (2) and (3):
[0056] (2)
[0057] (3)
[0058] In the formula: z i and z i+1 Let be the lower boundary height and upper boundary height of the i-th region, respectively; when i=1,2,…, At that time, the internal boundary z i+1 Take the midpoint between the i-th and (i+1)-th measured heights; the lower boundary of the first region is... The upper boundary of the last region is H+R. The integral result is obtained from equation (4):
[0059] (4)
[0060] S22. Solve for the equivalent vector wind speed of the X-axis component. Equivalent vector wind speed of Y-axis component and equivalent vector wind direction ;
[0061] The IEC 61400-12-1:2022 standard only describes a method for fitting equivalent wind speeds at multiple altitudes, but does not propose a method for fitting equivalent wind directions at multiple altitudes. This paper adopts multi-altitude vector wind speed to fit the equivalent wind direction, as shown in equations (5) and (6):
[0062] (5)
[0063] (6)
[0064] In the formula: For the first Actual wind speed at a certain height Component wind speed on the X-axis; For the first Actual wind speed at a certain height Wind speed component on the Y-axis; This is the wind direction angle.
[0065] Based on equations (5) and (6) and Using a similar method to calculate the equivalent wind speed The equivalent vector wind speed is calculated using the method described above. and As shown in equations (7) and (8):
[0066] (7)
[0067] (8)
[0068] In the formula: The equivalent vector wind speed on the X-axis component; This represents the equivalent vector wind speed along the Y-axis. Equivalent vector wind direction. The equivalent wind direction is obtained using the arctangent function. The arctangent function has a range of -90° to 90°, while the natural wind direction is between 0° and 360°. Therefore, in the process of obtaining the equivalent wind direction, a vector form is used to convert the equivalent wind direction obtained by the arctangent function to a range of 0° to 360°, as shown in equation (9):
[0069] (9)
[0070] S3, Equivalent wind speed in step S2 The upward translation process is performed to obtain the equivalent vector wind speed of the X-axis component after the upward translation. Equivalent vector wind speed of Y-axis component X-axis reverse equivalent vector wind speed and Y-axis reverse equivalent vector wind speed ;
[0071] Specifically, the equivalent wind speed calculated based on step S2 and equivalent vector wind direction Calculate the equivalent vector wind speed after upward translation. , , , As shown in equations (10), (11), (12), and (13):
[0072] (10)
[0073] (11)
[0074] (12)
[0075] (13)
[0076] In the formula: The equivalent vector wind speed on the X-axis component after upward translation; The equivalent vector wind speed on the Y-axis component after upward translation; The reverse equivalent vector wind speed on the X-axis component after upward translation; This is the reverse equivalent vector wind speed on the Y-axis component after upward translation.
[0077] S4. Equivalent vector wind speed after translation in step S3 is calculated for a preset duration. , , , The process is segmented, high-pass filtered, and the sum of the standard deviations of wind speed before and after filtering is calculated. Extreme gust events are extracted according to a preset screening ratio;
[0078] Specifically, the equivalent vector wind speed , , , The process is divided into segments with a step size of 10 minutes, and each segment is then subjected to high-pass filtering. This invention uses a Butterworth second-order high-pass filter for... , , , Perform filtering. The filter's cutoff frequency. It can be determined by the turbulence conditions at the actual site, as shown in equations (14), (15), and (16):
[0079] (14)
[0080] (15)
[0081] (16)
[0082] In the formula: For the turbulence integral time scale, For turbulent integral scale, This represents the measured average wind speed.
[0083] Calculate the standard deviation of the wind speed data before filtering. and the standard deviation of the filtered wind speed data As shown in equations (17) and (18):
[0084] (17)
[0085] (18)
[0086] In the formula: The wind speed before filtering; The average wind speed before filtering; The filtered wind speed; The average wind speed after filtering; This represents the number of sample points for each 10-minute wind speed segment.
[0087] Finally adopted The screening criteria are used to screen extreme gust events. The calculated results are then sorted from largest to smallest, and the top 1 / 4000 are selected as extreme gust events. Therefore, this paper uses the top 1 / 4000 ratio for gust screening as extreme gust events.
[0088] S5. Perform continuous wavelet transform on the extreme gust events extracted in step S4, fit the error function curve, and extract the gust amplitude, rise time and wind direction angle change to complete the identification of extreme coherent gusts.
[0089] After obtaining the extreme gust event, a wavelet transform is performed on it to obtain the extreme coherent gust, as shown in equation (19):
[0090] (19)
[0091] In the formula: The objective function is... For the mother wavelet, The normalization coefficient is used. The local data features on the slope upwards are extracted using a mirrored first-order Gaussian wavelet, as shown in equation (20):
[0092] (20)
[0093] In the formula: This is the normalization constant, and its value is equal to... .
[0094] The time and position parameters obtained from the wavelet transform above are used for fitting, as shown in equation (21):
[0095] (twenty one)
[0096] In the formula: The wind speed prior to the escalation of an extreme gust event; This refers to the wind speed following an extreme gust event; It is a time variable; τ is the local feature position parameter of the gust obtained by continuous wavelet transform, used to characterize the center time position of the gust's upward phase; τ is the scale parameter obtained by continuous wavelet transform, used to characterize the time scale of the gust's upward process. The error function is shown in equation (22):
[0097] (twenty two)
[0098] Based on the above processing principles, continuous wavelet transform is performed on extreme gust events to identify the specific locations and corresponding scales of abrupt wind speed changes within a 10-minute timeframe, which are then used for fitting, such as... Figure 2 As shown, where Figure 2(a) shows the wind speed time history curve and the function fitting graph. The black curve represents the curve of the measured original wind speed changing over time, reflecting the actual fluctuation of wind speed within a 600-second time span. The red curve represents the result of fitting the gust features (such as the uphill phase) extracted by wavelet transform, usually using an error function. Alternatively, linear / nonlinear fitting can be used to smooth the rising edge of gusts, thereby accurately extracting the gust amplitude. and rise time; Figure 2 (b) is a continuous wavelet transform time-frequency heat map, which can intuitively show the distribution of sudden wind speed changes (gusts) in the time-frequency domain, thus separating short-term gust events (high frequency, small scale) from long-term wind speed trends (low frequency, large scale).
[0099] The optimal parameters are calculated using the nonlinear least squares method. and ,use The gust amplitude is determined as an extreme coherent gust. The rise time is the optimal error function calculated using the nonlinear least squares method. The transition section is determined, that is, the wind speed from arrive The time elapsed. When calculating the change in wind direction angle during the corresponding rising time period, the wind speed should be from... The prevailing wind direction between the corresponding time point and the end of the fitted curve, and the wind speed from the beginning of the fitted curve to... The change in the prevailing wind direction between corresponding time points is determined as the change in wind direction angle for extreme coherent gusts.
[0100] Method validity verification
[0101] To verify the effectiveness of the proposed method, this invention uses a NERL_5WM wind turbine model and measured lidar data for subsequent numerical case analysis. First, site information is introduced; second, the results of the proposed method are compared and analyzed with those of traditional gust identification methods based on raw wind speed; finally, the two cases are discussed and analyzed.
[0102] This invention uses lidar data measured at two small turbulent sites for research, and the lidar sampling information is shown in Tables 1 and 2 below.
[0103] Table 1. LiDAR information for one site
[0104]
[0105] Table 2. LiDAR information at site two
[0106]
[0107] The NERL_5WM wind turbine model was used in this invention for subsequent research, and its parameters are shown in Table 3.
[0108] Table 3 Parameters of NERL_5MW Wind Turbine
[0109]
[0110] Based on the above lidar information and the 5 MW wind turbine model, the lidar height data and the area of the entire wind turbine plane occupied by each height layer were calculated according to the equivalent wind speed calculation formula, as shown in Tables 4 and 5 below.
[0111] Table 4 Site Measurement Height Points
[0112]
[0113] Table 5 Measurement height points at Site 2
[0114]
[0115] Case 1
[0116] Based on site-specific lidar data and on-site information, the ECDs selected using the gust identification method proposed in this invention and the traditional gust identification method based on raw wind speed were compared. The gust amplitude, wind direction angle change, and rise time under the two methods were also compared. Figure 3 As shown, blue represents the gust identification method based on the original wind speed (Hannesdóttir's method), and red represents the gust identification method proposed in this invention (Proposed method).
[0117] Figure 3 (a) The horizontal axis represents gust amplitude (m / s), and the vertical axis represents relative frequency (%). Figure 3(a) It can be seen that the data selected by the gust identification method proposed in this invention has the following distribution characteristics: In a single wind speed range, the 11-12 m / s range accounts for the highest proportion (19%); from multiple ranges, the gust amplitudes in the 5 m / s to 9 m / s, 9 m / s to 12 m / s, and above 12 m / s ranges account for 22%, 57%, and 21%, respectively. In contrast, in the data selected by the gust identification method based on the original wind speed, the 5 m / s to 6 m / s range accounts for the highest proportion (21%), and the gust amplitudes below 8 m / s account for a total of 84%. Thus, it can be seen that the ECD gust amplitudes selected by the gust identification method of this invention are generally greater than those of the gust identification method based on the original wind speed. However, although the gust identification method based on the original wind speed can ensure that the ECD gust amplitude is as large as possible, for example, it can select the highest wind speed range of 20 m / s to 22 m / s, further analysis reveals that the corresponding wind direction angle change is only 8.6°, and the rise time is as long as 476 s. In contrast, the method presented in this paper selected a maximum gust amplitude of 15 m / s, corresponding to a wind direction angle change of up to 84°, with an ascent time of only 107 s.
[0118] Figure 3 (b) The horizontal axis represents the change in wind direction (in degrees), and the vertical axis represents the relative frequency (%). Figure 3 (b) It can be seen that in the data filtered using the gust identification method proposed in this invention, the wind direction angle change accounts for the highest proportion in the 30° to 40° range, reaching 26%; the range less than 30° accounts for only 19% in total, while the range above 40° accounts for as high as 54% in total. In contrast, in the data filtered using the gust identification method based on the original wind speed, the 0° to 10° range accounts for as high as 84%, and the sum of all other ranges accounts for only 16%. It is evident that the ECDs filtered by the method of this invention generally have larger wind direction angle changes. In summary, the method proposed in this paper considers both gust amplitude and wind direction angle changes, and the identified gust events not only have higher gust amplitudes but also more significant wind direction angle changes, which is more consistent with the physical characteristics of ECDs.
[0119] Figure 3 (c) The horizontal axis represents rise time (in seconds), and the vertical axis represents relative frequency (%). Figure 3(c) It can be seen that, among the extreme coherent gust rise time data filtered using the gust identification method proposed in this invention, the rise time in the 0 s to 50 s interval accounts for the highest proportion at 47%, the rise time in the 0 s to 100 s interval accounts for 83%, and the data exceeding 100 s accounts for only 17%, with the highest wind speed rise time not exceeding 350 s. In contrast, among the data filtered using the gust identification method based on the original wind speed, the rise time in the 0 s to 50 s interval accounts for the highest proportion at 49%, the rise time in the 0 s to 100 s interval accounts for 61%, and the data exceeding 350 s accounts for a high proportion at 19%. Therefore, the ECD rise time filtered using the method proposed in this invention is generally short, with very few cases exceeding 200 s. Thus, the gust identification method proposed in this paper not only considers maximizing wind direction and speed but also takes into account the important factor of a relatively short rise time. In summary, the results obtained by using two methods to identify extreme coherent gusts at the same site show that the method proposed in this paper generally has larger gust amplitudes, larger changes in wind direction angles, and relatively shorter rise times.
[0120] Based on the results of the gust amplitude, wind direction angle change and rise time, the average value and standard deviation of the overall results are calculated, as shown in Table 6.
[0121] Table 6. Mean and Standard Deviation of ECD Variables at Site 1
[0122]
[0123] As shown in Table 6, the standard deviations of the rise time and wind direction angle change in the gust identification method based on the original wind speed (Method 2) are both greater than their mean values (rise time: 153.76 > 128.00; wind direction angle change: 25.06 > 12.47). Combined with... Figure 2 (a) It can be seen that the data distribution selected by this method exhibits a high right-skewed characteristic, while also containing a small amount of left-skewed data. The standard deviation is significantly affected by these extreme values, resulting in a few extreme values widening the overall standard deviation. In contrast, the gust identification method proposed in this paper (Method 1) has a mean gust amplitude of 10.02 m / s, which is greater than the original method's 6.39 m / s. The mean change in wind direction angle of 53.71° is also significantly greater than the original method's 12.47°, while the mean rise time of 72.11 s is shorter than the original method's 128.00 s. Judging from the average values, the gust events selected by the method in this paper generally have larger gust amplitudes, larger changes in wind direction angle, and shorter rise times, which are more consistent with the physical characteristics of extreme coherent gusts.
[0124] Case 2
[0125] Based on the lidar data from Site 2 and its on-site information, ECDs selected using the proposed gust identification method and the gust identification method based on raw wind speed were compared. The gust amplitude, wind direction angle change, and rise time under the two methods were also compared. Figure 4 As shown, blue represents the gust identification method based on the original wind speed (Hannesdóttir's method), and red represents the gust identification method proposed in this invention (Proposed method).
[0126] Figure 4 (a) The horizontal axis represents gust amplitude (m / s), and the vertical axis represents relative frequency (%). Figure 4 (a) It can be seen that in the data filtered using the gust identification method proposed in this invention, the three intervals of 2 m / s to 3 m / s, 8 m / s to 9 m / s, and 9 m / s to 10 m / s each account for 15%, ranking highest; the gust amplitude above 8 m / s accounts for 55% in total, and below 8 m / s accounts for 45%. Although below 8 m / s still accounts for a relatively high proportion, further analysis reveals that this is mainly because the available data at site 2 is much smaller than that at site 1, resulting in the gust amplitudes filtered by the method in this paper not being significantly better than those based on the original wind speed. In contrast, in the data filtered by the gust identification method based on the original wind speed, the 5 m / s to 7 m / s interval accounts for the highest proportion (48%), while above 8 m / s only accounts for 27%. Comparing the two methods, although they have different distribution characteristics, in terms of average value, the gust amplitudes filtered by the method in this invention are still slightly larger than those of the original method.
[0127] Figure 4 (b) The horizontal axis represents the change in wind direction (in degrees), and the vertical axis represents the relative frequency (%). Figure 4 (b) It can be seen that in the data filtered using the gust identification method proposed in this invention, the change in wind direction angle is highest in the 60° to 70° range, reaching 25%; the range greater than 70° accounts for 45% in total, while the range of 50° to 60° accounts for only 10%; the cumulative change in wind direction angle greater than 60° accounts for as high as 90%. In contrast, in the data filtered using the gust identification method based on the original wind speed, the range of 0° to 20° accounts for as high as 90%, while the sum of all other ranges accounts for only 10%.
[0128] Therefore, it is evident that the extreme coherent gusts screened by the method of this invention generally exhibit larger changes in wind direction angle, significantly outperforming the original method. In summary, the method proposed in this paper considers both gust amplitude and wind direction angle changes simultaneously, identifying gust events with not only higher gust amplitudes but also more significant changes in wind direction angle, thus better conforming to the physical characteristics of extreme coherent gusts.
[0129] Figure 4 (c) The horizontal axis represents rise time (in seconds), and the vertical axis represents relative frequency (%). Figure 4 (c) It can be seen that among the extreme coherent gusts screened using the gust identification method proposed in this invention, the highest proportion (60%) has an rise time in the range of 0 to 50 seconds; the range of 50 to 100 seconds accounts for 20%, the range of 100 to 150 seconds accounts for 5%, and the range of 150 to 200 seconds accounts for 15%, with no data points exceeding 200 seconds. In contrast, among the data screened by the gust identification method based on the original wind speed, the range of 0 to 50 seconds accounts for 27%, the range of 0 to 100 seconds accounts for only 41%, while the data exceeding 200 seconds accounts for as much as 60%, occupying the majority. Therefore, the gust events screened by the method of this invention generally have shorter rise times, with no cases exceeding 200 seconds. In summary, the method proposed in this paper effectively controls the rise time to better reflect extreme situations while considering the changes in gust amplitude and wind direction angle.
[0130] Based on the results of the gust amplitude, wind direction angle change and rise time, the average value and standard deviation of the overall results are calculated, as shown in Table 7.
[0131] Table 7. Mean and Standard Deviation of ECD Variables at Site 2
[0132]
[0133] As shown in Table 7, the average gust amplitude of the gust identification method proposed in this invention (Method 1) is 7.81 m / s, which is greater than the 7.34 m / s of the ECD identification method based on the original wind speed (Method 2). The average change in wind direction angle is also significantly greater than the 15.82° of the original method. However, the average rise time is 66.70 s, which is shorter than the 231.97 s of the original method. Judging from the average values, the gust events screened by the method in this paper generally have larger gust amplitudes, larger changes in wind direction angle, and shorter rise times, which are more consistent with the physical characteristics of extreme coherent gusts.
[0134] Case Result Analysis and Verification
[0135] This invention, when identifying wind speed, fits wind data at various altitudes to derive equivalent wind speed and equivalent wind direction. The changes in the equivalent wind speed represent the overall wind speed variation within the wind turbine plane. When a 10-minute segment of the equivalent wind speed is identified as the ECD (Electronic Wind Direction), it can be assumed that a sudden gust of wind occurred simultaneously and synchronously at all altitude levels during that time period. This invention will identify the largest... The value corresponds to the ECD and the field address 2, which identify the largest value. The corresponding ECD value is verified.
[0136] Figure 5 (a) and (b) are three-dimensional time history comparison diagrams of wind speed and wind direction at different altitudes at the site, respectively; Figure 6 (a) and (b) are time-history comparison diagrams of two-dimensional sub-plots of wind speed and wind direction at different altitudes at site 1, respectively; from Figure 5 , Figure 6 As can be seen from this, the largest one at the site The corresponding ECD values show that the wind speed time histories at various altitudes begin to change around 300 s, and reach their extreme values around 450 s. Simultaneously, the wind direction also shows an upward trend between 300 s and 450 s, with a wind direction change of approximately 90°.
[0137] Figure 7 (a) and (b) are two-dimensional line graphs of the time histories of equivalent wind speed and equivalent wind direction at the site, respectively; from Figure 7 As can be seen, the equivalent wind speed fitted at each altitude begins to change at 300 s, reaching its extreme around 450 s. The equivalent wind direction also shows an upward trend between 300 s and 450 s, with a wind direction change of approximately 100°. Furthermore, since the wind direction change is periodic from 0° to 360°, to present the continuous change in wind direction in the graph and to prevent jumps between 0° and 360°, wind direction angles greater than 360° will appear in the graph. This demonstrates that within the wind turbine plane, wind speed and direction at each altitude level experience simultaneous gusts and abrupt changes. Therefore, fitting equivalent wind speed and equivalent wind direction from multiple altitudes within the wind turbine plane can represent the changes throughout the entire wind turbine plane.
[0138] Figure 8 (a) and (b) are three-dimensional time history comparison diagrams of wind speed and wind direction at different heights at Site 2, respectively; Figure 9(a) and (b) are time-history comparison diagrams of two-dimensional sub-plots of wind speed and wind direction at various heights at Site 2, respectively. Figure 8 , Figure 9 As can be seen from this, the largest at site two The corresponding ECD values show that the wind speed time histories at various altitudes begin to change around 200 s, reaching their extreme values around 300 s. Simultaneously, the wind direction also shows an upward trend between 200 s and 300 s, with a wind direction change of approximately 70°.
[0139] Figure 10 (a) and (b) are two-dimensional line graphs of the time histories of equivalent wind speed and equivalent wind direction at site 2, respectively; from Figure 10 As can be seen, the equivalent wind speeds fitted at various heights begin to change at 200 s, reaching their extreme values around 300 s; the equivalent wind direction also shows an upward trend between 200 s and 300 s, with a wind direction change of approximately 50°. This indicates that within the wind turbine plane, wind speeds and directions at various height levels experience simultaneous gusts and abrupt changes. Therefore, using the equivalent wind speeds and directions fitted at multiple heights within the wind turbine plane can represent the changes throughout the entire wind turbine plane.
[0140] In summary, wind speed and direction change synchronously at various height levels within the wind turbine plane. Therefore, fitting the equivalent wind speed using multi-height wind speeds within the wind turbine plane is reasonable and effective. Under this variation, the load borne by the wind turbine is greater than the load caused by gusts identified at a single height level, and is closer to the actual load borne by the wind turbine.
[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.
Claims
1. A method for identifying extreme coherent gusts based on equivalent vector wind speed, characterized in that, Includes the following steps: S1. Obtain measured wind speed and wind direction angle data at multiple representative heights within the wind turbine plane; S2, Based on the wind speed at various heights within the wind turbine plane in step S1 The equivalent wind speed is calculated based on the area weight of the wind turbine plane. Equivalent vector wind speed of X-axis component Equivalent vector wind speed of Y-axis component And based on this, the equivalent vector wind direction can be calculated. ; S3, Equivalent wind speed in step S2 The upward translation process is performed to obtain the equivalent vector wind speed of the X-axis component after the upward translation. Equivalent vector wind speed of Y-axis component X-axis reverse equivalent vector wind speed and Y-axis reverse equivalent vector wind speed ; S4. Equivalent vector wind speed after translation in step S3 is calculated for a preset duration. , , , The process is segmented, high-pass filtered, and the sum of the standard deviations of wind speed before and after filtering is calculated. Extreme gust events are extracted according to a preset screening ratio; S5. Perform continuous wavelet transform on the extreme gust events extracted in step S4, fit the error function curve, and extract the gust amplitude, rise time, and wind direction angle change to complete the identification of extreme coherent gusts.
2. The extreme coherence gust identification method as described in claim 1, characterized in that, In step S1, remote sensing wind equipment or traditional mechanical / ultrasonic anemometers installed on the wind turbine tower and blades are used to simultaneously collect measured data at at least three different height levels within the wind turbine's sweep range. The measured data includes the horizontal wind speed scalar at each height. and horizontal wind angle .
3. The extreme coherence gust identification method as described in claim 2, characterized in that, Equivalent wind speed in step S2 The calculation formula is: Equivalent vector wind speed on the X-axis component The calculation formula is: Equivalent vector wind speed on the Y-axis component The calculation formula is: ; Let be the area of the entire wind turbine plane, where n is the radius of the wind turbine; h Let n be the number of wind speed data points at different measurement heights within the range of the wind turbine height, and n h ≥3; n h The measurement heights are arranged from low to high. The midpoint between two adjacent measurement heights serves as the boundary between adjacent wind turbine plane regions. The lower boundary of the region corresponding to the lowest measurement height is the lower boundary of the wind turbine sweep range, and the upper boundary of the region corresponding to the highest measurement height is the upper boundary of the wind turbine sweep range. i Let i be the area of the wind turbine plane region corresponding to the i-th measurement height, i=1,2,…,n h Each A i Wind speed at various altitudes One-to-one correspondence, and all A's. i The sum of the areas equals A; for the remaining areas above the highest measurement height and below the lowest measurement height, which are still within the sweep range of the wind turbine, their areas are respectively included in the area A corresponding to the highest and lowest measurement heights. i The wind speed at the corresponding boundary measurement height was measured respectively. Participate in equivalent wind speed calculation; and The first Wind speed measured at a certain altitude Components on the X and Y axes.
4. The extreme coherence gust identification method as described in claim 3, characterized in that, Equivalent vector wind direction in step S2 The equivalent wind direction obtained from the arctangent function is mapped to the 0~360° interval using vector form. Specifically, the calculation is performed piecewise as follows: when hour, ;when hour, ; when hour, ;when hour, .
5. The method for identifying extreme coherent gusts as described in claim 1, characterized in that, The equivalent vector wind speed on the X-axis component after upward translation in step S3 The calculation formula is: ; Equivalent vector wind speed on the Y-axis component after upward translation The calculation formula is: ; The reverse equivalent vector wind speed on the X-axis component after upward translation The calculation formula is: ; The reverse equivalent vector wind speed on the Y-axis component after upward translation The calculation formula is: .
6. The method for identifying extreme coherent gusts as described in claim 1, characterized in that, In step S4, the high-pass filter uses a Butterworth second-order high-pass filter with a cutoff frequency of [missing information]. ,in For turbulent integral scale, This represents the measured average wind speed. Extreme gust events screening adopted Sort by magnitude from largest to smallest, and take the top 1 / 4000 as extreme gust events. The standard deviation of the wind speed data before filtering. This represents the standard deviation of the filtered wind speed data.
7. The method for identifying extreme coherent gusts as described in claim 1, characterized in that, In step S5, the continuous wavelet transform uses a mirrored first-order Gaussian wavelet: , Here is the normalization constant; the formula for fitting the error function curve based on the time and position parameters is: In the formula: The wind speed prior to the escalation of an extreme gust event; This refers to the wind speed following an extreme gust event; It is a time variable; τ is the local feature position parameter of the gust obtained by continuous wavelet transform, used to characterize the center time position of the gust's upward phase; τ is the scale parameter obtained by continuous wavelet transform, used to characterize the time scale of the gust's upward process. The error function is used; the extracted gust amplitude is... The rise time is from the wind speed. arrive The time elapsed.
8. An extreme coherent gust identification system based on equivalent vector wind speed, characterized in that, include: The data acquisition module is used to acquire measured wind speed and wind direction angle data at multiple representative heights within the wind turbine plane; The equivalent calculation module is used to calculate the equivalent wind speed based on the wind speed at various heights within the wind turbine plane and the area weight of the wind turbine plane. Equivalent vector wind speed of X-axis component Equivalent vector wind speed of Y-axis component and equivalent vector wind direction ; Translation filter module, used for equivalent vector wind speed The upward translation process is performed to obtain the equivalent vector wind speed of the X-axis component after the upward translation. Equivalent vector wind speed of Y-axis component X-axis reverse equivalent vector wind speed and Y-axis reverse equivalent vector wind speed After segmenting into segments with a preset duration, high-pass filtering is performed based on... Extracting extreme gust events; The identification module is used to perform continuous wavelet transform on extreme gust events to extract gust amplitude, rise time, and wind direction angle change.
9. A computing device comprising a processor and a memory, the memory storing computer program instructions, characterized in that, The processor implements the extreme coherent gust identification method according to any one of claims 1 to 7 when executing computer program instructions.
10. A computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the extreme coherent gust identification method according to any one of claims 1 to 7.
Citation Information
Patent Citations
Method and device for identifying extreme coherent gust parameters
CN119025907A
Specific site extreme gust calculation method, device and equipment and storage medium
CN119179845A
