Wind farm wind measurement data cleaning method, device, equipment and medium

Abstract

The present application belongs to the technical field of wind measurement data cleaning, and particularly relates to a wind farm wind measurement data cleaning method, device, equipment and medium. The method firstly performs preliminary screening on the data through threshold test, wind speed increment and wind shear verification. Then, in combination with high-resolution reanalysis data and weather station data, the atmospheric stability is judged and the wind profile formula is corrected, and the data is cleaned according to the atmospheric stability and the theoretical wind speed. Finally, the present application further screens and fills the data based on the space-time distribution characteristics of the data, thereby improving the reliability of the data. The present application improves the accuracy of the wind farm wind measurement data, and provides strong data guarantee for wind power prediction.

Description

Technical Field

[0001] This invention belongs to the field of wind measurement data cleaning technology, specifically relating to a method, apparatus, equipment and medium for cleaning wind measurement data in wind farms. Background Technology

[0002] Wind power forecasting relies on historical power generation data from wind farms, meteorological conditions, and equipment information. It predicts future wind power output by establishing the relationship between wind speed and wind power output. However, actual wind measurement data is often inaccurate or incomplete due to sensor malfunctions, data transmission problems, and environmental interference, affecting the accuracy of the forecasting model. Therefore, ensuring data quality and effective data cleaning are crucial. Existing data restoration methods are mostly limited to single sites or short-term time series, lacking cross-site, long-term spatiotemporal correlation restoration, which affects restoration accuracy and model stability. Furthermore, wind farms in remote areas often lack sufficient wind measurement towers, making it difficult to obtain high-quality data. Reanalysis data (such as ERA5) and meteorological station data can effectively compensate for gaps and errors in wind measurement data, providing more accurate wind speed forecasts. Finally, wind speed variations at different altitudes are affected by atmospheric stability, but existing restoration methods ignore this factor, resulting in low restoration accuracy. Accurate atmospheric stability analysis and wind speed profile modeling can improve wind speed restoration accuracy, thereby enhancing the accuracy of wind power forecasting. Summary of the Invention

[0003] The purpose of this invention is to provide a method, apparatus, equipment, and medium for cleaning wind measurement data in wind farms, so as to solve the problem of cleaning wind measurement data in the prior art.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] In a first aspect, the present invention provides a method for cleaning wind measurement data from a wind farm, comprising the following steps:

[0006] Obtain the raw wind measurement data of the current wind farm, and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the raw wind measurement data;

[0007] The first wind measurement data is verified by wind speed increment and wind shear to obtain the second wind measurement data that passes the verification.

[0008] The Monin-Obukhov length is determined time-by-time; the stability of the wind profile formula is corrected based on the Monin-Obukhov length to obtain the modified wind profile formula; the theoretical wind speed of the current wind farm at different heights is determined time-by-time based on the modified wind profile formula; the data is filtered based on the wind speed deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data whose wind speed deviation is within the preset deviation range.

[0009] The wind direction deflection pattern of the third wind measurement data is verified based on the Monin-Obukhov length at each time step to obtain the fourth wind measurement data that meets the preset wind direction change threshold.

[0010] Calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within the given rolling window; calculate the deviation range based on the rolling mean and rolling standard deviation, and filter the fifth wind measurement data that meets the deviation range;

[0011] Obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values.

[0012] The missing data in the sixth wind measurement data were filled in to obtain the final wind measurement data after data cleaning.

[0013] Furthermore, the stability of the wind profile formula is corrected based on the Monin-Obukhov length, resulting in a revised wind profile formula. The theoretical wind speed at different altitudes of the current wind farm is then determined time-by-time using this revised formula, where:

[0014] The wind profile formulas include logarithmic wind profile formulas and power-law wind profile formulas. The logarithmic wind profile formula is modified according to the Monin-Obukhov length and the correction function to obtain the first modified formula. The power-law wind profile formula is modified according to the Monin-Obukhov length and the corrected power exponent to obtain the second modified formula.

[0015] The reference height and reference wind speed of the current wind farm are determined at each time step. Based on the reference height and reference wind speed, and the first or second revision, the theoretical wind speed of the current wind farm at different heights is calculated at each time step.

[0016] Furthermore, determine the reference height and reference wind speed of the current wind farm, including:

[0017] Determine the geographical location of the wind measurement towers in the current wind farm;

[0018] If a data reference point exists within a preset distance of the geographical location, the 10-meter wind speed data of the data reference point is acquired at each time step; it is determined whether the deviation between the data reference point and the 10-meter wind speed data of the wind measuring tower meets a preset deviation threshold; if the wind speed deviation does not meet the deviation threshold, the 10-meter wind speed data of the data reference point is used as the reference wind speed, and 10 meters is used as the reference height; wherein, the data reference point is a meteorological station, or the nearest grid point determined according to the reanalysis data resolution;

[0019] If there is no data reference point within the preset distance of the geographical location, or if the deviation between the data reference point and the 10-meter wind speed data of the meteorological tower meets the preset deviation threshold, then a height is selected from the current wind farm at each moment as the reference height, and the wind speed at the corresponding height is used as the reference wind speed.

[0020] Furthermore, the Monin-Obukhov length of the current wind farm is determined time-by-time, including:

[0021] The reference height and reference wind speed of the wind farm are determined at each time step, and the friction speed is calculated based on the reference height and reference wind speed.

[0022] Determine the current sensible heat flux of the wind farm at each moment;

[0023] The Monin-Obukhov length of the current wind farm is calculated hourly based on the friction velocity and sensible heat flux of the current wind farm.

[0024] Furthermore, obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm, including:

[0025] Identify multiple neighboring wind farms within a preset distance of the current wind farm;

[0026] The weight of each neighboring wind farm is determined based on the distance between each neighboring wind farm and the current wind farm.

[0027] Based on the weights of each neighboring wind power station, the weighted average wind speed and weighted average wind direction of each neighboring wind power station are calculated.

[0028] Furthermore, the missing data in the sixth wind measurement data were filled in to obtain the final wind measurement data after data cleaning, including:

[0029] If, at any given moment, wind measurement data exists at at least one height in the current wind farm, then the theoretical wind measurement data is calculated by revising the formula based on the wind profile, and the missing wind measurement data is filled in with the theoretical wind measurement data. If wind measurement data is missing at all heights in the current wind farm, then time-series data interpolation is performed on the wind measurement data at different times at the same height.

[0030] The wind measurement data after time series data interpolation is filled with a weighted average of neighboring wind power stations to obtain the final wind measurement data.

[0031] In a second aspect, the present invention provides a wind farm wind measurement data cleaning device, comprising:

[0032] The first cleaning module is used to acquire the original wind measurement data of the current wind farm and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the original wind measurement data.

[0033] The second cleaning module is used to perform wind speed increment and wind shear verification on the first wind measurement data to obtain the second wind measurement data that passes the verification.

[0034] The third cleaning module is used to determine the Monin-Obukhov length of the current wind farm at each time step; correct the wind profile formula based on the Monin-Obukhov length to obtain the corrected wind profile formula; determine the theoretical wind speed of the current wind farm at different heights based on the corrected wind profile formula; and filter the data based on the wind speed deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data whose wind speed deviation is within the preset deviation range.

[0035] The fourth cleaning module verifies the wind direction deflection pattern of the third wind measurement data time-by-time based on the Monin-Obukhov length, and obtains fourth wind measurement data that meets the preset wind direction change threshold.

[0036] The fifth cleaning module is used to calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within a given rolling window; and to calculate the deviation range based on the rolling mean and rolling standard deviation, and then filter the fifth wind measurement data that meets the deviation range.

[0037] The sixth cleaning module is used to obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values.

[0038] The seventh cleaning module is used to fill in the missing data in the sixth wind measurement data to obtain the final wind measurement data after data cleaning.

[0039] Furthermore, in the third cleaning module, the stability of the wind profile formula is corrected based on the Monin-Obukhov length to obtain the modified wind profile formula; based on the modified wind profile formula, the theoretical wind speed of the current wind farm at different heights is determined time-by-time, where:

[0040] The wind profile formulas include logarithmic wind profile formulas and power-law wind profile formulas. The logarithmic wind profile formula is modified according to the Monin-Obukhov length and the correction function to obtain the first modified formula. The power-law wind profile formula is modified according to the Monin-Obukhov length and the corrected power exponent to obtain the second modified formula.

[0041] The reference height and reference wind speed of the current wind farm are determined at each time step. Based on the reference height and reference wind speed, and the first or second revision, the theoretical wind speed of the current wind farm at different heights is calculated at each time step.

[0042] In a third aspect, the present invention provides an electronic device including a processor and a memory, the processor being configured to execute a computer program stored in the memory to implement the wind farm wind measurement data cleaning method described above.

[0043] In a fourth aspect, the present invention provides a computer-readable storage medium storing at least one instruction that, when executed by a processor, implements the wind farm wind measurement data cleaning method described above.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] This invention combines high-resolution reanalysis data and meteorological station data, based on theories related to atmospheric stability and boundary layer wind profiles. Through a multi-step data cleaning process, including preliminary screening, wind speed increment and wind shear verification, theoretical wind speed verification, rolling deviation verification, verification with data from nearby wind farms, and filling in abnormal and missing data, the accuracy of wind measurement data is improved. Each step addresses potential data anomalies or errors, effectively reducing the impact of inaccurate or missing data on wind power prediction. Improving the accuracy of wind power prediction is of great significance for the operation and maintenance of wind farms, the stable operation of the power grid, and the full utilization of renewable energy.

[0046] For abnormal or missing data, this invention proposes a method for filling theoretical wind measurement data based on wind profile models and interpolating time-series data, as well as a method for filling data by weighted average of adjacent wind power stations.

[0047] This invention improves the robustness of the method by incorporating data from neighboring wind farms for verification and data filling. It also enables the utilization of a wider range of data resources to optimize data quality in situations where sufficient local data reference points are lacking.

[0048] In summary, this invention solves the problem of inaccurate or missing wind measurement data in existing technologies by providing a wind farm wind measurement data cleaning method, thus providing more reliable data support for wind power prediction. The wind farm wind measurement data cleaning device, electronic equipment, and computer-readable storage medium provided by this invention also solve the problems raised in the background section. Attached Figure Description

[0049] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0050] Figure 1 This is a flowchart illustrating a wind farm wind measurement data cleaning method according to an embodiment of the present invention;

[0051] Figure 2 This is a structural block diagram of a wind farm wind measurement data cleaning device according to an embodiment of the present invention;

[0052] Figure 3 This is a structural block diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0053] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0054] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0055] Example 1

[0056] like Figure 1 As shown, a method for cleaning wind measurement data from a wind farm includes the following steps:

[0057] S1. Obtain the original wind measurement data of the current wind farm, and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the original wind measurement data.

[0058] Specifically, step S1 includes: acquiring raw wind measurement data of the wind farm at different heights, which is long-term time-series data. The raw wind measurement data is then verified against abnormal thresholds and abnormal jumps. Data that passes verification is recorded as the first wind measurement data, and data that fails verification is marked as NaN.

[0059] As an example, according to the "Regulations for Real-time Monitoring of Solar Resources in Photovoltaic Power Plants" (Energy Industry Standard, NB / T 32012-2013), the normal threshold ranges for wind speed and wind direction are 0-60. m / s And 0-360°. In plains areas without significant topographic obstruction, under normal weather conditions, the wind speed ranges at heights of 10 meters, 30 meters, 50 meters, and 70 meters are approximately 2-8... m / s 3-10 m / s 4-12 m / s and 5-14 m / s Under extreme weather conditions, they can reach 30 respectively. m / s 40 m / s 45 m / s and 50 m / sWind measurement data should be within the normal threshold range; otherwise, it is considered abnormal and exceeds the threshold. Furthermore, according to the "Regulations for Real-time Monitoring of Solar Resources in Photovoltaic Power Stations" (Energy Industry Standard, NB / T 32012-2013), the maximum change in wind speed over 5 minutes should not exceed 20. m / s The wind direction jump should not exceed 360°. The jump in continuous data should not exceed this standard; otherwise, it is considered abnormal. Data that passes this verification step will proceed to the next check.

[0060] S2. Perform wind speed increment and wind shear verification on the first wind measurement data to obtain the second wind measurement data that passes the verification.

[0061] Specifically, step S2 includes: using the wind speed increase law and horizontal wind shear rate (α) to quantify the wind speed change between different heights, obtaining the second wind measurement data that has passed the verification, and marking the data that has not passed the verification as NaN.

[0062] It should be noted that wind speed gradually increases with altitude. Therefore, the wind speed at different altitudes is checked step-by-step to ensure it follows this increasing pattern. If the wind speed at a certain altitude does not increase, it is marked as NaN.

[0063] In addition, adopt α This is used to quantify wind speed variations at different altitudes. The specific formula is:

[0064] (1)

[0065] Where z1 and z2 are adjacent heights, U z1 and U z2 This represents the wind speed at the corresponding altitude. Under normal atmospheric conditions, the wind shear rate near the ground is usually not too large. If the wind shear rate at a certain altitude exceeds the reasonable range (0.1~0.5), then... m / s / m If the result is negative, it indicates that the data is inaccurate. The data verified in this step will be further examined.

[0066] S3. Determine the Monin-Obukhov length at each time step; perform stability correction on the wind profile formula based on the Monin-Obukhov length to obtain the modified wind profile formula; determine the theoretical wind speed of the current wind farm at different heights at each time step based on the modified wind profile formula; filter the data based on the wind speed deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data where the wind speed deviation is within the preset deviation range.

[0067] Specifically, determining the Monin-Obukhov length of the current wind farm at each time step includes: determining the reference height and reference wind speed of the current wind farm at each time step, calculating the friction velocity based on the reference height and reference wind speed; determining the sensible heat flux at the corresponding time step; and calculating the Monin-Obukhov length based on the friction velocity and sensible heat flux at the corresponding time step.

[0068] Specifically, determining the reference height and reference wind speed of the current wind farm at each time step includes: determining the geographical location of the meteorological tower in the current wind farm; if a data reference point exists within a preset distance of the geographical location, acquiring the 10-meter wind speed data of the data reference point at the corresponding time; determining whether the deviation between the data reference point and the 10-meter wind speed data of the meteorological tower meets a preset deviation threshold; if the wind speed deviation does not meet the deviation threshold, using the 10-meter wind speed data of the data reference point at the corresponding time as the reference wind speed and 10 meters as the reference height; wherein, the data reference point is a meteorological station, or the nearest grid point determined according to the reanalysis data resolution; if no data reference point exists within a preset distance of the geographical location, or if the deviation between the data reference point and the 10-meter wind speed data of the meteorological tower meets the preset deviation threshold, then selecting a height at the corresponding time from the current wind farm as the reference height and the wind speed at the corresponding height as the reference wind speed.

[0069] As an example, using the geographical location information of weather stations, the nearest weather station to the current wind farm's anemometer tower is determined (the default maximum distance is 30 kilometers; beyond this distance, it is considered that no weather station meets the requirements). If no weather station meets the requirements, the latitude and longitude information of the reanalysis data is used to determine the nearest grid point to the wind farm's anemometer tower. A time window (e.g., 5 minutes) is set with the current time as the center to resolve potential time misalignment issues between different data sources. For the current wind farm, if a nearest weather station exists, the nearest 10-meter wind speed data within the specified time window is acquired hourly. If no weather station data exists, the 10-meter wind speed data from the reanalysis data is acquired in the same way. The 10-meter wind speed of the anemometer tower is compared with the difference between the weather station or the reanalysis data. If the wind speed difference is greater than a set threshold, the 10-meter wind speed from the weather station or the reanalysis data is selected as the reference wind speed. Otherwise, the 10-meter wind speed from the wind measurement tower is selected as the reference wind speed. At this point, the reference altitude... Set the height to 10 meters. If there is no weather station or reanalysis data, select a height from the wind farm's own data. (Prioritize lower altitudes), and use the wind speed at that altitude as... If both the reference altitude and reference wind speed are valid data, proceed to the next step.

[0070] Specifically, the friction speed is calculated based on the reference altitude and reference wind speed. (unit: m / s ):

[0071] (2)

[0072] Where H is the sensible heat flux and k is the von Kármán constant, with a value of approximately 0.4; The roughness length indicates the surface roughness. You can look up the table according to the surface type. See Table 1 for details.

[0073] Table 1. Reference Table of Land Use Types and Surface Roughness Length in China

[0074]

[0075] Specifically, the sensible heat flux of the wind farm is determined moment by moment, including: setting a time window (e.g., 5 minutes) centered on the current moment to resolve potential time misalignment issues between different data sources; dynamically selecting multi-source data to calculate the sensible heat flux H (unit: W / m 2 This refers to the amount of heat passing through a unit area per unit time, and its calculation formula is as follows:

[0076] (3)

[0077] Among them, C H The heat exchange coefficient is approximately 0.001; U represents the ground wind speed (unit: ...). m / s ); T s and T a These represent surface temperature and air temperature (unit: ). K ).

[0078] Optionally, for a given wind farm, if a nearest meteorological station exists (within 30 kilometers), the most recent relevant meteorological observation data within a specified time window is acquired hourly. If no meteorological station data exists, the nearest reanalysis grid point to the wind farm is determined, and H is calculated by acquiring the corresponding grid point data. If the reanalysis data itself already includes the variable H, the above formula is not used for calculation.

[0079] Specifically, the Monin-Obukhov length of the current wind farm is determined moment by moment, including: combining Calculate the Monin-Obukhov length L using H. The formula for calculating the Monin-Obukhov length L is as follows:

[0080] (4)

[0081] Where g is the acceleration due to gravity, approximately 9.81. m / s2 T represents atmospheric temperature (unit: ℃). K ); ρ represents air density, taken as 1.225. kg / m 3 ;c p This represents the specific heat capacity of air, approximately 1004. J / (kg⋅K) .

[0082] ①Unstable condition (L<0): Surface heating induces strong turbulence, with convection dominating.

[0083] ② Neutral conditions (L→∞): Surface heat flux is close to zero, turbulence is moderate, and it is mainly generated by wind speed shear.

[0084] ③ Stable conditions (L>0): Surface cooling induces weak turbulence, and stratification is dominant.

[0085] Specifically, wind profile formulas include logarithmic wind profile formulas and power-law wind profile formulas, which determine the theoretical wind speed of the current wind farm at different heights at each time step, including:

[0086] Based on the Monin-Obukhov length and correction function, the logarithmic wind profile formula is modified to obtain the first modified formula; based on the Monin-Obukhov length and corrected power exponent, the power-law wind profile formula is modified to obtain the second modified formula; the reference height and reference wind speed of the current wind farm are determined at each time step; based on the reference height and reference wind speed, and the first or second modified formula, the theoretical wind speed of the current wind farm at different heights is calculated at each time step.

[0087] As an example, using L, we introduce a correction function. The stability of the logarithmic wind profile formula is corrected.

[0088] (5)

[0089] Correction function Format:

[0090] (6)

[0091] Using L, the stability of the power law profile formula is corrected by adjusting the power exponent.

[0092] (7)

[0093] Where Λ is the wind speed profile index, the value of which is related to the stability condition:

[0094] (8)

[0095] As an example, choose either the logarithmic wind profile after stability correction or the power-law wind profile, and combine them moment by moment. and Calculate the theoretical wind speed at different altitudes. Verify the actual wind speed at each altitude at the corresponding time to ensure it is within the deviation range of the theoretical wind speed. Set the deviation percentage P (unit: %) and determine the deviation range. If the difference is not within the deviation range, the actual wind measurement data at the corresponding time and altitude is considered abnormal and marked as NaN.

[0096] (9)

[0097] in, V 理论 Theoretical wind speed, V 实际 This represents the actual wind speed.

[0098] S4. Based on the Monin-Obukhov length, the wind direction deflection pattern of the third wind measurement data is verified at each time step to obtain the fourth wind measurement data that meets the preset wind direction change threshold.

[0099] Specifically, the wind direction deflection pattern of the third wind measurement data is verified time-by-time based on the Monin-Obukhov length, including: checking the wind direction deflection pattern time-by-time in conjunction with the Monin-Obukhov length L. Under stable atmospheric conditions (L>0), the wind direction usually deflects slowly with increasing altitude. Drastic wind direction changes typically do not occur between 10 and 70 meters in altitude. If continuous wind direction changes exceed a reasonable wind direction change threshold, the verification fails.

[0100] S5. Calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within the given rolling window; calculate the deviation range based on the rolling mean and rolling standard deviation, and filter the fifth wind measurement data that meets the deviation range.

[0101] Specifically, step S5 includes:

[0102] (1) After setting a rolling window (e.g., 5 minutes), calculate the rolling mean and rolling standard deviation of wind speed within the window for each altitude time series data.

[0103] (2) Calculate the deviation range based on the rolling mean and rolling standard deviation of wind speed. Check the actual wind speed at each moment. If the actual wind speed exceeds the deviation range, the data is considered abnormal and marked as NaN.

[0104] (3) The method for handling wind direction is the same as that for wind speed.

[0105] Deviation range = rolling_mean ± multiple × rolling_std; the multiple of standard deviation defaults to 2.

[0106] S6. Obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values.

[0107] Specifically, the process involves obtaining the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance from the current wind farm. This includes: identifying multiple neighboring wind farms within a preset distance from the current wind farm; determining the weight of each neighboring wind farm based on its distance from the current wind farm; and calculating the weighted average wind speed and weighted average wind direction of each neighboring wind farm based on its weight.

[0108] As an example, S6 includes:

[0109] (1) Calculate the distance between the current wind farm and other wind farms (the default maximum is 30 kilometers; if the distance exceeds this, it is considered that there are no nearby wind farms that meet the requirements). Find all nearby wind farms within this range and extract their wind speed and direction data. If multiple nearby wind farms meet the conditions (the number of nearby wind farms is greater than 1), proceed to the next step.

[0110] (2) Calculate the weights based on the distance between the nearby wind farms and the current site, and then calculate the weighted average wind speed of the nearby wind farms. The weights are inversely proportional to the distance between the sites (the closer the distance, the greater the weight).

[0111] (3) Compare the current wind measurement data with the weighted average wind speed to check the deviation of the actual wind speed. Set the wind speed deviation threshold to ≤5. m / s If the actual wind speed deviation exceeds the threshold, it is marked as abnormal and recorded as a NaN value.

[0112] The wind direction is handled in the same way as the wind speed, with a wind direction deviation threshold of ≤30°.

[0113] S7. Fill in the missing data, i.e. NaN values, in the sixth wind measurement data to obtain the final wind measurement data after data cleaning.

[0114] Specifically, the missing data, i.e. NaN values, in the sixth wind measurement data are filled to obtain the final wind measurement data after data cleaning. This includes: for a given moment, if the current wind farm has wind measurement data at at least a certain height, then the theoretical wind measurement data is calculated by modifying the formula according to the wind profile, and the missing wind measurement data is filled with the theoretical wind measurement data; if the wind measurement data at all heights of the current wind farm is missing, then time series data interpolation is performed on the wind measurement data at different times at the same height; the wind measurement data after time series data interpolation is filled with a weighted average of the neighboring wind farms to obtain the final wind measurement data.

[0115] As an example, step S7 includes:

[0116] (1) Filling based on wind profile model. Fill the wind measurement data at different heights at the same time according to the calculation results of the modified wind profile formula. If the wind measurement data at all heights at a certain time of the current wind farm is missing, skip this step and proceed to the next step.

[0117] (2) Perform time-series data interpolation. For wind measurement data at different times at the same altitude, select different interpolation methods based on the data characteristics before and after the current time. Interpolation methods include linear interpolation, forward interpolation, backward interpolation, and cubic spline interpolation. If the interpolated data is still NaN, proceed to the next step. The wind direction is handled in the same way as the wind speed.

[0118] ① Linear interpolation requires at least one valid wind measurement data point at each time step.

[0119] Set at time and There is a valid wind speed measurement, and its corresponding value is and Regarding the current moment Interpolation is performed to obtain The formula is:

[0120] (10)

[0121] ② Forward interpolation requires valid wind measurement data from the previous time step.

[0122] Set at time There is a valid wind speed measurement, and its corresponding value is Regarding the current moment Interpolation is performed to obtain The formula is:

[0123] (11)

[0124] ③ Backward interpolation requires valid wind measurement data at the next time step.

[0125] Set at time There is a valid wind speed measurement, and its corresponding value is Regarding the current moment Interpolation is performed to obtain The formula is:

[0126] (12)

[0127] ④ Cubic spline interpolation requires at least one valid wind measurement data point before and after the time step. To ensure the continuity and smoothness of the interpolation function, ideally, there should be at least two valid wind measurement data points before and after the time step to be interpolated, which will make the spline interpolation more accurate.

[0128] For time The corresponding interpolation for:

[0129] (13)

[0130] in For time interval Defined cubic polynomial, , , and These are the polynomial coefficients for that interval, and they need to be calculated under the following conditions:

[0131] Define interval length :

[0132] (14)

[0133] By introducing the second derivative at the node Construct a tridiagonal system of equations:

[0134] (15)

[0135] , Determined based on boundary conditions.

[0136] Construct a tridiagonal system of equations to solve matrix m:

[0137] (16)

[0138] Where A is a tridiagonal matrix with dimensions n×n; is the second derivative to be solved; b is a vector constructed from the wind speed difference.

[0139] After solving matrix m, calculate the polynomial coefficients:

[0140] (17)

[0141] (3) Use the weighted average of neighboring wind farms to fill in the missing data. If the number of neighboring wind farms that meet the conditions is greater than 1, calculate the weighted average based on the wind measurement data of the neighboring wind farms and use the result to fill in the missing data of the current wind farm.

[0142] Example 2

[0143] like Figure 2 As shown, based on the same inventive concept as the above embodiments, the present invention also provides a wind farm wind measurement data cleaning device, comprising:

[0144] The first cleaning module is used to acquire the original wind measurement data of the current wind farm and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the original wind measurement data.

[0145] The second cleaning module is used to perform wind speed increment and wind shear verification on the first wind measurement data to obtain the second wind measurement data that passes the verification.

[0146] The third cleaning module is used to determine the Monin-Obukhov length of the current wind farm at each time step; correct the wind profile formula based on the Monin-Obukhov length to obtain the corrected wind profile formula; determine the theoretical wind speed of the current wind farm at different heights based on the corrected wind profile formula; and filter the data based on the wind speed deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data whose wind speed deviation is within the preset deviation range.

[0147] The fourth cleaning module is used to verify the wind direction deflection pattern of the third wind measurement data at each time step based on the Monin-Obukhov length, so as to obtain the fourth wind measurement data that meets the preset wind direction change threshold.

[0148] The fifth cleaning module is used to calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within a given rolling window; and to calculate the deviation range based on the rolling mean and rolling standard deviation, and then filter the fifth wind measurement data that meets the deviation range.

[0149] The sixth cleaning module is used to obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values.

[0150] The seventh cleaning module is used to fill in the missing data, i.e., NaN values, in the sixth wind measurement data to obtain the final wind measurement data after data cleaning.

[0151] Furthermore, in the third cleaning module, the stability of the wind profile formula is corrected based on the Monin-Obukhov length to obtain the modified wind profile formula; based on the modified wind profile formula, the theoretical wind speed of the current wind farm at different heights is determined time-by-time, where:

[0152] The stability of the wind profile formula is corrected based on the Monin-Obukhov length, resulting in a revised version of the wind profile formula. This revised version includes both logarithmic and power-law wind profile formulas. The logarithmic wind profile formula is corrected using the Monin-Obukhov length and a correction function, resulting in the first revised version. The power-law wind profile formula is then corrected using the Monin-Obukhov length and a corrected power exponent, resulting in the second revised version.

[0153] The reference height and reference wind speed of the current wind farm are determined at each time step. Based on the reference height and reference wind speed, and the first or second revision, the theoretical wind speed of the current wind farm at different heights is calculated at each time step.

[0154] Example 3

[0155] like Figure 3 As shown, the present invention also provides an electronic device 100 for implementing a wind farm wind measurement data cleaning method;

[0156] The electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on at least one processor 102, and at least one communication bus 104.

[0157] The memory 101 can be used to store computer program 103. The processor 102 implements the steps of the wind farm wind measurement data cleaning method of Embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101.

[0158] The memory 101 may primarily include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created based on the use of the electronic device 100 (such as audio data), etc. In addition, the memory 101 may include non-volatile memory, such as hard disk, RAM, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other non-volatile solid-state storage device.

[0159] At least one processor 102 may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Processor 102 may be a microprocessor or any conventional processor. Processor 102 is the control center of electronic device 100, connecting various parts of electronic device 100 via various interfaces and lines.

[0160] The memory 101 in the electronic device 100 stores multiple instructions to implement a wind farm wind measurement data cleaning method, and the processor 102 can execute multiple instructions to achieve the following:

[0161] Obtain the raw wind measurement data of the current wind farm, and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the raw wind measurement data;

[0162] The first wind measurement data is verified by wind speed increment and wind shear to obtain the second wind measurement data that passes the verification.

[0163] The Monin-Obukhov length is determined time-by-time; the stability of the wind profile formula is corrected based on the Monin-Obukhov length to obtain the modified wind profile formula; the theoretical wind speed of the current wind farm at different heights is determined time-by-time based on the modified wind profile formula; the data is filtered based on the wind speed deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data whose wind speed deviation is within the preset deviation range.

[0164] The wind direction deflection pattern of the third wind measurement data is verified based on the Monin-Obukhov length at each time step to obtain the fourth wind measurement data that meets the preset wind direction change threshold.

[0165] Calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within the given rolling window; calculate the deviation range based on the rolling mean and rolling standard deviation, and filter the fifth wind measurement data that meets the deviation range;

[0166] Obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values.

[0167] The missing data, i.e. NaN values, in the sixth wind measurement data are filled to obtain the final wind measurement data after data cleaning.

[0168] Example 4

[0169] If the modules / units integrated in the electronic device 100 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, and read-only memory (ROM).

[0170] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0171] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0172] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0173] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0174] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0175] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for cleaning wind measurement data from a wind farm, characterized in that, Includes the following steps: Obtain the raw wind measurement data of the current wind farm, and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the raw wind measurement data; The first wind measurement data is verified by wind speed increment and wind shear to obtain the second wind measurement data that passes the verification. The Monin-Obukhov length is determined time-by-time; the stability of the wind profile formula is corrected based on the Monin-Obukhov length to obtain the modified wind profile formula; the theoretical wind speed of the current wind farm at different heights is determined time-by-time based on the modified wind profile formula. Data is filtered based on the deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data whose wind speed deviation is within the preset deviation range; The wind direction deflection pattern of the third wind measurement data is verified based on the Monin-Obukhov length at each time step to obtain the fourth wind measurement data that meets the preset wind direction change threshold. Calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within the given rolling window; The deviation range is calculated based on the rolling mean and rolling standard deviation, and the fifth wind measurement data that meets the deviation range is selected. Obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values. Fill in the missing data in the sixth wind measurement data to obtain the final wind measurement data after data cleaning; Wind profile formulas include logarithmic wind profile formulas and power-law wind profile formulas; The stability of the wind profile formula is corrected based on the Monin-Obukhov length, resulting in the revised wind profile formula, which includes: Based on the Monin-Obukhov length and correction function, the logarithmic wind profile formula is modified to obtain the first revised form, which includes: Using the Monin-Obukhov length L, a correction function is introduced. Stability corrections were made to the logarithmic wind profile formula: Correction function The form is: Based on the Monin-Obukhov length and the modified power exponent, the power-law profile formula is modified to obtain the second revised formula, which includes: Using the Monin-Obukhov length L, the stability of the power law profile formula is corrected by adjusting the power exponent. in, The friction velocity is k; k is von Kármán's constant. It represents the roughness length of the earth's surface; Λ is the wind speed profile index; For reference wind speed; For reference height.

2. The wind farm wind measurement data cleaning method according to claim 1, characterized in that, The theoretical wind speed at different altitudes of the current wind farm is determined hourly using the modified wind profile formula, where: The reference height and reference wind speed of the current wind farm are determined at each time step. Based on the reference height and reference wind speed, and the first or second revision, the theoretical wind speed of the current wind farm at different heights is calculated at each time step.

3. The wind farm wind measurement data cleaning method according to claim 1, characterized in that, Determine the reference height and reference wind speed of the wind farm at each time step, including: Determine the geographical location of the wind measurement towers in the current wind farm; If a data reference point exists within a preset distance of the geographical location, the 10-meter wind speed data of the data reference point is acquired at each time step; it is determined whether the deviation between the data reference point and the 10-meter wind speed data of the wind measuring tower meets a preset deviation threshold; if the wind speed deviation does not meet the deviation threshold, the 10-meter wind speed data of the data reference point is used as the reference wind speed, and 10 meters is used as the reference height; wherein, the data reference point is a meteorological station, or the nearest grid point determined according to the reanalysis data resolution; If there is no data reference point within the preset distance of the geographical location, or if the deviation between the data reference point and the 10-meter wind speed data of the meteorological tower meets the preset deviation threshold, then a height is selected from the current wind farm at each moment as the reference height, and the wind speed at the corresponding height is used as the reference wind speed.

4. The wind farm wind measurement data cleaning method according to claim 1, characterized in that, Determining the Monin-Obukhov length of the current wind farm at each time step includes: The reference height and reference wind speed of the wind farm are determined at each time step, and the friction speed is calculated based on the reference height and reference wind speed. Determine the current sensible heat flux of the wind farm at each moment; The Monin-Obukhov length of the current wind farm is calculated hourly based on the friction velocity and sensible heat flux of the current wind farm.

5. The wind farm wind measurement data cleaning method according to claim 1, characterized in that, Obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm, including: Identify multiple neighboring wind farms within a preset distance of the current wind farm; The weight of each neighboring wind farm is determined based on the distance between each neighboring wind farm and the current wind farm. Based on the weights of each neighboring wind power station, the weighted average wind speed and weighted average wind direction of each neighboring wind power station are calculated.

6. The wind farm wind measurement data cleaning method according to claim 1, characterized in that, The missing data in the sixth wind measurement data were filled in to obtain the final wind measurement data after data cleaning, including: If, at any given moment, wind measurement data exists at at least one height in the current wind farm, then the theoretical wind measurement data is calculated by revising the formula based on the wind profile, and the missing wind measurement data is filled in with the theoretical wind measurement data. If wind measurement data is missing at all heights in the current wind farm, then time-series data interpolation is performed on the wind measurement data at different times at the same height. The wind measurement data after time series data interpolation is filled with a weighted average of neighboring wind power stations to obtain the final wind measurement data.

7. A wind farm wind measurement data cleaning device, characterized in that, include: The first cleaning module is used to acquire the original wind measurement data of the current wind farm and filter out the first wind measurement data that meets the first preset threshold and the first preset jump value from the original wind measurement data. The second cleaning module is used to perform wind speed increment and wind shear verification on the first wind measurement data to obtain the second wind measurement data that passes the verification. The third cleaning module is used to determine the Monin-Obukhov length of the current wind farm at each time step; correct the wind profile formula based on the Monin-Obukhov length to obtain the corrected wind profile formula; determine the theoretical wind speed of the current wind farm at different heights at each time step based on the corrected wind profile formula; and filter the data based on the wind speed deviation between the theoretical wind speed and the second wind measurement data to obtain the third wind measurement data whose wind speed deviation is within the preset deviation range. The fourth cleaning module verifies the wind direction deflection pattern of the third wind measurement data time-by-time based on the Monin-Obukhov length, and obtains fourth wind measurement data that meets the preset wind direction change threshold. The fifth cleaning module is used to calculate the rolling mean and rolling standard deviation of the fourth wind measurement data within a given rolling window. The deviation range is calculated based on the rolling mean and rolling standard deviation, and the fifth wind measurement data that meets the deviation range is selected. The sixth cleaning module is used to obtain the weighted average wind speed and weighted average wind direction of multiple neighboring wind farms within a preset distance of the current wind farm; calculate the wind speed deviation and wind direction deviation of the fifth wind measurement data based on the weighted average wind speed and weighted average wind direction; and filter out the sixth wind measurement data that meets the preset wind speed deviation threshold and wind direction deviation threshold based on the wind speed deviation and wind direction deviation values. The seventh cleaning module is used to fill in the missing data in the sixth wind measurement data to obtain the final wind measurement data after data cleaning. Wind profile formulas include logarithmic wind profile formulas and power-law wind profile formulas; The stability of the wind profile formula is corrected based on the Monin-Obukhov length, resulting in the revised wind profile formula, which includes: Based on the Monin-Obukhov length and correction function, the logarithmic wind profile formula is modified to obtain the first revised form, which includes: Using the Monin-Obukhov length L, a correction function is introduced. Stability corrections were made to the logarithmic wind profile formula: Correction function The form is: Based on the Monin-Obukhov length and the modified power exponent, the power-law profile formula is modified to obtain the second revised formula, which includes: Using the Monin-Obukhov length L, the stability of the power law profile formula is corrected by adjusting the power exponent. in, The friction velocity is k; k is von Kármán's constant. It represents the roughness length of the earth's surface; Λ is the wind speed profile index; For reference wind speed; For reference height.

8. The wind farm wind measurement data cleaning device according to claim 7, characterized in that, In the third cleaning module, the wind profile formula is corrected for stability based on the Monin-Obukhov length to obtain the revised wind profile formula. Based on the revised wind profile formula, the theoretical wind speed at different altitudes of the current wind farm is determined time-by-time, where: The wind profile formulas include logarithmic wind profile formulas and power-law wind profile formulas. The logarithmic wind profile formula is modified according to the Monin-Obukhov length and the correction function to obtain the first modified formula. The power-law wind profile formula is modified according to the Monin-Obukhov length and the corrected power exponent to obtain the second modified formula. The reference height and reference wind speed of the current wind farm are determined at each time step. Based on the reference height and reference wind speed, and the first or second revision, the theoretical wind speed of the current wind farm at different heights is calculated at each time step.

9. An electronic device, characterized in that, It includes a processor and a memory, the processor being used to execute a computer program stored in the memory to implement the wind farm wind measurement data cleaning method as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores at least one instruction, which, when executed by a processor, implements the wind farm wind measurement data cleaning method as described in any one of claims 1 to 6.

Images