Design method for calculating layer thickness based on high-altitude wind detection data of wind speed jitter rate

Abstract

The application relates to a high-altitude wind detection data calculation layer thickness design method based on a wind speed jitter rate, and belongs to the high-altitude wind detection field. The application obtains high-altitude wind second data based on navigation positioning data, calculates service high-altitude wind data, calculates a wind speed jitter rate, calculates a jitter rate change amount or a jitter rate change rate, obtains a fitted optimal pendulum period multiple, and obtains an optimal calculation layer thickness. The application fully considers factors such as navigation positioning errors, clock pendulum effects, side slip effects and the like, adopts a high-precision navigation positioning instrument, carries out high-altitude wind detection tests by means of balloons of various sizes and ropes of various lengths, adopts a comprehensive data processing method, sets an optimal smoothing time interval on the basis of balancing measurement errors and calculation layer thicknesses, determines an optimal calculation layer thickness, reduces detection errors, and retains more wind speed change details at the same time.

Description

Technical Field

[0001] This invention belongs to the field of high-altitude wind detection, specifically relating to a method for designing the thickness of a calculation layer for high-altitude wind detection data based on wind speed fluctuation rate. Background Technology

[0002] (1) Status of upper-air wind detection in my country and requirements for different altitude levels

[0003] Currently, my country's meteorological departments primarily use balloons as tracers for upper-air meteorological sounding operations. These balloons are often equipped with radiosondes, and wind direction and speed are calculated based on the position-time relationship. The horizontal distance traveled per unit time is the wind speed, which is typically obtained by dividing the horizontal distance of the balloon or radiosonde by the time interval.

[0004] Balloon-based wind measurement mainly includes L-band wind radar and GPS / BeiDou navigation wind measurement. Currently, my country's meteorological departments primarily use L-band radar for upper-air wind detection. L-band radar tracks tracers to obtain their location information and then calculates wind direction and speed. Its detection data is acquired at second intervals, meaning a set of data is acquired every 1 second.

[0005] In recent years, with the development of navigation technology, GPS / BeiDou navigation has been increasingly applied to upper-air meteorological detection, specifically for upper-air wind detection. It primarily uses navigation and positioning functions to locate the radiosonde mounted on a balloon, and then uses the positioning information to calculate wind speed and direction. Currently, the update frequency of domestic navigation satellite chips is typically 1Hz, meaning a set of data is acquired every 1 second; therefore, the satellite-second interval data acquired during navigation wind measurement is crucial.

[0006] Because L-band radar and navigation anemometers use second-interval data, the wind speed and direction data at these intervals fluctuate significantly over time, potentially leading to errors. Therefore, without processing, they cannot be directly used by operational departments. According to the "Operational Specifications for Regular Upper-Air Meteorological Observation" issued by the China Meteorological Administration in 2010, raw anemometer data cannot be directly provided to operational departments. It must first be processed to obtain wind patterns at specified altitudes, including isobaric surfaces, wind at designated heights, the maximum wind layer, upper-level wind characteristic layers, and the tropopause, before being used by operational departments.

[0007] According to the "Standard Operating Procedures for Routine Upper-Air Meteorological Observation," to ensure the accuracy of upper-air wind measurement data, the time intervals for calculating upper-air wind direction and speed used in meteorology and climatology are 1 minute, 2 minutes, and 4 minutes. Specifically, starting from the moment the balloon is released, for the first 20 minutes or less, each calculation layer is 1 minute; for 20-40 minutes, each calculation layer is 2 minutes; and after 40 minutes, each calculation layer is 4 minutes. Based on a standard 750g rubber-coated balloon's ascent speed of 400 m / min for routine meteorological operations, the calculation layer thicknesses for routine upper-air meteorological detection are 400 m, 800 m, and 1600 m, respectively.

[0008] However, the data obtained at time intervals of 1 min, 2 min, and 4 min have relatively long time intervals and a large calculation layer thickness. Although this can reduce the measurement error of upper-level winds to some extent, it also results in the loss of more wind details, which seriously affects the accuracy of wind measurement.

[0009] (2) Description of error

[0010] In actual upper-altitude wind detection, wind speed and direction exhibit significant random fluctuations with altitude, which characterize the measurement errors. Upper-altitude wind measurement errors mainly originate from equipment positioning errors (i.e., equipment positioning accuracy), localized small-scale atmospheric turbulence, pendulum effects, and sideslip effects.

[0011] a) Positioning error. Whether using L-band radar for wind measurement or navigation-based positioning, wind speed and direction are calculated based on the radiosonde's position. Therefore, the error in wind speed and direction depends on the positional error between the two positioning points, i.e., the radiosonde's detection accuracy. Increasing the time interval between data calculations can effectively reduce the positional error between the two points. For the same positioning equipment, the shorter the calculation time interval, the greater the error in wind speed and direction.

[0012] b) Pendulum effect. During ascent, the radiosonde carried by the weather balloon is pulled by the tether, causing it to exhibit a spiral "pendulum" motion, also known as the "pendulum effect." This pendulum effect will produce wind measurement errors, and the pendulum effect is related to the length of the tether. The longer the tether, the larger the pendulum period and the greater the error.

[0013] c) Local small-scale turbulence. The movement of the atmosphere on small scales in space and time is quite complex. Wind not only has direction and magnitude but also gusts, i.e., wind turbulence. Atmospheric motion on small scales is complex and variable. As a weather balloon ascends, it needs to traverse a series of turbulences, and its positioning will be affected by atmospheric turbulence, inevitably introducing errors in location information—this is atmospheric turbulence error. High-frequency fluctuations in wind speed and direction over short periods characterize wind measurement errors. According to the law of atmospheric turbulence variation with altitude, the greater the wind speed, the smaller the amplitude of wind speed and direction changes, and the amplitude of these changes tends to decrease with increasing altitude. That is, the scale of atmospheric turbulence tends to increase with increasing altitude, while the fluctuations caused by atmospheric turbulence decrease with increasing altitude. Therefore, when the scale of atmospheric turbulence is small, it will introduce certain wind measurement errors; when the scale of atmospheric turbulence is large, the fluctuations of turbulence decrease, and the errors caused by turbulence are weakened.

[0014] Due to factors such as equipment positioning errors, small-scale turbulence, and the pendulum effect, wind speed and direction exhibit high-frequency fluctuations over short periods, resulting in a jittery wind curve. Therefore, wind speed and direction calculated using second-interval data show significant fluctuations over time and cannot be directly applied to routine meteorological operations. In operational applications, smoothing is typically used to remove the high-frequency changes in wind speed and direction.

[0015] However, the measurement error of upper-level winds is directly related to the smoothing time interval and the thickness of the calculation layer. When smoothing wind speed and direction, different calculation time intervals, calculation layer thicknesses, or different smoothing scales will result in different processing results for the same measurement data. The shorter the time interval and the smaller the calculation layer thickness, the greater the wind measurement error, but it can still present the true details of wind speed changes to a certain extent. Figure 1 As shown on the left. The longer the time interval, the thicker the calculation layer, the greater the attenuation of wind speed fluctuations by the smoothing curve, and the smaller the wind measurement error. However, this masks more of the actual wind fluctuation characteristics, such as... Figure 1 As shown on the right. However, when the calculation time interval is too long or too short, or the calculation layer thickness is too large or too small, it cannot represent the true characteristics of upper-level winds. Therefore, based on actual operational needs and considering atmospheric turbulence characteristics and the measurement performance of positioning equipment, an optimal smoothing time interval and calculation layer thickness are designed to reduce detection errors while retaining more details of true wind speed changes.

[0016] (3) The introduction of jitter rate

[0017] Upper-altitude wind detection data is sampled at second intervals. Location information is acquired every second by an anemometer and then processed into wind speed and direction. However, due to equipment positioning errors, small-scale atmospheric turbulence, pendulum effects, and balloon sideslip effects, calculating wind speed and direction using data with very short intervals results in significant fluctuations over time, exhibiting a pronounced jitter characteristic. Figure 2 As shown. To correctly understand the characteristics of jitter, a proper jitter quantification method is required. Therefore, we define the concepts of jitter and jitter rate. Jitter refers to the short-term oscillations of data within a valid instantaneous time series, while the jitter rate is the amplitude of these oscillations per unit time.

[0018] Data jitter has two main types: deterministic jitter and random jitter. Deterministic jitter includes jitter caused by natural variations in wind speed with altitude and objective wind speed changes due to large-scale atmospheric turbulence. Random jitter, on the other hand, is caused by measurement errors such as positioning errors, pendulum effects, and small-scale atmospheric turbulence. In upper-air meteorological sounding, deterministic errors can be considered to be generated by the atmosphere itself and are not affected by human factors, while random errors are caused by measurements and can be modified manually. In practical operations, data smoothing can be used to remove high-frequency variations in wind speed and direction, thereby reducing jitter.

[0019] By observing changes in jitter rate, the fluctuation characteristics of the data can be visually displayed, characterizing the measurement error. A shorter smoothing time results in a smaller computational layer thickness, revealing more realistic details of wind speed changes, but also leading to greater measurement error and more pronounced jitter. Conversely, a longer smoothing time interval results in a thicker computational layer, causing greater attenuation of wind speed fluctuations by the smoothing curve, smaller measurement error, and smoother jitter, but also masking more of the actual wind fluctuation characteristics and causing phase lag. Therefore, in practical applications, an optimal smoothing time interval and optimal computational layer thickness should be set to reduce detection error while preserving more details of wind speed changes.

[0020] In view of this, the present invention designs an optimal calculation layer thickness design method based on high-altitude wind detection data with wind speed fluctuation rate. Summary of the Invention

[0021] (a) Technical problems to be solved

[0022] The technical problem to be solved by this invention is how to provide a design method for the calculation layer thickness of high-altitude wind detection data based on wind speed jitter rate, so as to solve the problems that when the calculation layer thickness is too large in the high-altitude wind detection calculation process, more wind details are lost, which seriously affects the accuracy of wind measurement, and when the calculation layer thickness is too small, the side slip effect of the balloon, the pendulum silence of the rope length and the positioning error of navigation are not taken into account, resulting in a large error in the calculated wind speed.

[0023] (II) Technical Solution

[0024] To address the aforementioned technical problems, this invention proposes a method for designing layer thickness based on high-altitude wind detection data with wind speed variability. This method includes the following steps:

[0025] S1. High-altitude wind detection is carried out by using a navigation and positioning device, with the help of balloons and ropes, and high-altitude wind second data is obtained based on navigation and positioning data;

[0026] S2, Calculate operational upper-level wind data

[0027] The least squares fitting method was used, and an integer multiple of the pendulum period was selected as the fitting time. The upper-level wind speed data within the fitting time was processed and calculated to obtain the average east-west wind speed u and north-south wind speed v within the fitting period.

[0028] S3, Calculate wind speed fluctuation rate

[0029] The fitted wind speed variability rate F is defined by the changes in north-south wind speed v and east-west wind speed u per unit time; the calculation formula is as follows:

[0030]

[0031] Where Fu is the east-west wind speed turbulence rate, and Fv is the north-south wind speed turbulence rate; u i v i t i represents the east-west wind speed, north-south wind speed, and corresponding time when the radiosonde is located at the i-th altitude layer; n is the number of upper-level wind fitting periods.

[0032] The north-south wind speed turbulence rate and the east-west wind speed turbulence rate are combined to form the wind speed turbulence rate:

[0033]

[0034] Wind speed fluctuations include deterministic fluctuations and measurement error fluctuations;

[0035] The measurement error jitter rate dF is expressed as:

[0036]

[0037] Where dFu is the measurement error jitter rate of east-west winds, dFv is the measurement error jitter rate of north-south winds, and Fuw and Fvw are the deterministic jitter rates caused by wind speed variations and atmospheric turbulence.

[0038] The combined jitter rate of the measurement error is:

[0039]

[0040] S4. Obtain the deterministic jitter rate of the atmospheric wind field by calculating the change in jitter rate.

[0041] The wind speed turbulence rate is iteratively calculated by using one pendulum period (1T), two pendulum periods (2T), ..., j pendulum periods (j*T) as the fitting time. As the fitting time interval gradually increases, the change in turbulence rate gradually decreases. When the change in turbulence rate is less than a certain threshold, the wind speed turbulence rate is used to replace the atmospheric wind field deterministic turbulence rate caused by the wind speed itself and atmospheric turbulence, thereby obtaining the atmospheric wind field deterministic turbulence rate.

[0042] The formula for calculating the change in wind speed fluctuation rate is:

[0043]

[0044] δu j δv j Fu represents the change in east-west and north-south wind speed turbulence rates corresponding to two adjacent fitting periods, where j is an integer multiple of the pendulum period used for fitting. j 、Fv j Let j be the east-west and north-south wind speed fluctuation rates under j pendulum cycles;

[0045] When the fitting time interval is long enough, the random fluctuations caused by the error are considered to be eliminated, and at this point, the wind speed fluctuations are considered deterministic fluctuations. Assuming the fitting time interval gradually increases, the change in the fluctuation rate gradually decreases; when δu j δv j When it is less than λ, i.e., δu j δv j When the time is sufficiently short, use Fu during the j-th period of the simple pendulum. j 、Fv j The deterministic jitter rate of the atmospheric wind field, Fuw and Fvw, is obtained instead of the deterministic jitter rate of the wind field itself and the atmospheric turbulence.

[0046] S5. Obtain the optimal multiple of the pendulum period by calculating the rate of change of the jitter rate.

[0047] By substituting the deterministic jitter rate of the atmospheric wind field into the calculation formulas for the measurement error jitter rate and the composite jitter rate of the measurement error, the composite jitter rate of the measurement error is obtained. The rate of change of the jitter rate is then calculated, and the optimal multiple of the pendulum period k is obtained.

[0048] S6. Obtain the optimal computational layer thickness

[0049] The optimal fitting time is calculated using the optimal multiple of the pendulum period, and the corresponding optimal calculation layer thickness is obtained based on the ascent velocity of the weather balloon.

[0050] h = w·k·T

[0051] Where h is the optimal calculated layer thickness, w is the balloon's ascent speed, k is the optimal pendulum period multiple obtained through S5, and T is the oscillation period of the lanyard attached to the balloon.

[0052] (III) Beneficial Effects

[0053] This invention proposes a method for designing the calculation layer thickness based on wind speed jitter rate for high-altitude wind detection data. This invention fully considers factors such as navigation and positioning errors, pendulum effect, and sideslip effect. By using a high-precision navigation and positioning instrument and conducting high-altitude wind detection experiments with balloons of various sizes and ropes of various lengths, and by employing a comprehensive data processing method, the optimal smoothing time interval is set to determine the optimal calculation layer thickness based on balancing measurement errors and calculation layer thickness, thereby reducing detection errors while retaining more details of wind speed changes. Attached Figure Description

[0054] Figure 1 The graph shows the characteristics of upper-level wind speed variation with altitude at different smoothing time intervals. The left side shows a smoothing time interval of 1 pendulum cycle, and the right side shows a smoothing time interval of 6 pendulum cycles.

[0055] Figure 2 Vertical distribution of high-altitude wind detection data at second intervals for a 750g rubber-coated balloon with a 30m long rope attached;

[0056] Figure 3 The example shows the evolution characteristics of upper-level winds corresponding to different smoothing time intervals. The example uses a 300g rubber-coated balloon with a 6m hanging rope and a pendulum period of 5s. The data smoothing time from left to right is 1-16 pendulum periods, which are 5s, 10s, 15s, 20s...80s respectively.

[0057] Figure 4 The graph shows the distribution of jitter rate variation with the fitting time interval. The horizontal axis represents an integer multiple of the pendulum period used for the fitting time, the main vertical axis represents the jitter rate, the secondary axis represents the calculation layer thickness, and the black solid line represents the calculation layer thickness.

[0058] Figure 5 A schematic diagram of the design method for the optimal calculation layer thickness based on wind speed fluctuation rate for high-altitude wind detection. Detailed Implementation

[0059] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0060] According to the "Operational Specifications for Regular Upper-Air Meteorological Observation" issued by the China Meteorological Administration in 2010, the raw data from radiosonde wind measurement cannot be directly provided to the operational departments. The raw data should first be processed to obtain the wind at the specified layers, such as isobaric surface wind, wind at a specified height, maximum wind layer, upper-air wind characteristic layer, and tropopause, before it can be used by the operational departments.

[0061] According to the "Standard Operating Procedures for Routine Upper-Air Meteorological Observation," to ensure the accuracy of upper-air wind measurement data, the time intervals for calculating upper-air wind direction and speed used in meteorology and climatology are 1 minute, 2 minutes, and 4 minutes. Specifically, starting from the moment the balloon is released, for the first 20 minutes or less, each calculation layer is 1 minute; for 20-40 minutes, each calculation layer is 2 minutes; and after 40 minutes, each calculation layer is 4 minutes. Based on a standard 750g rubber-coated balloon's ascent speed of 400 m / min for routine meteorological operations, the calculation layer thicknesses for routine upper-air meteorological detection are 400 m, 800 m, and 1600 m, respectively.

[0062] However, the data obtained at time intervals of 1 min, 2 min, and 4 min have relatively long time intervals and a large calculation layer thickness. Although this can reduce the measurement error of upper-level winds to some extent, it also results in the loss of more wind details, which seriously affects the accuracy of wind measurement.

[0063] With the development of navigation-based anemometer technology, navigation anemometers are gradually becoming the main method for upper-altitude wind detection. However, navigation anemometers in my country are still in their initial stages. Currently, navigation anemometers still use second-interval data, which cannot be directly provided to operational departments and requires smoothing. Internationally, the smoothing interval for navigation anemometers is 30 seconds, while in my country it is approximately 40 seconds. Furthermore, the differences in balloon and rope lengths are not considered when selecting the smoothing interval. Therefore, in the calculation of upper-altitude wind detection, the sideslip effect of the balloon, the pendulum-like noise reduction of the rope, and the positioning error of the navigation are not taken into account, resulting in a relatively large error in the calculated wind speed.

[0064] In the design process, this invention fully considers factors such as navigation and positioning errors, pendulum effect, and sideslip effect. By using a high-precision navigation and positioning instrument and conducting high-altitude wind detection experiments with balloons of various sizes and ropes of various lengths, and by adopting a comprehensive data processing method, a method for calculating the optimal layer thickness for high-altitude wind detection based on wind speed fluctuation rate is designed to balance measurement errors and calculate layer thickness.

[0065] S1. High-altitude wind detection is conducted using a navigation and positioning device, aided by balloons and ropes, and high-altitude wind data is obtained based on navigation and positioning data.

[0066] Upper-altitude wind detection based on navigation and positioning typically uses the WGS-84 geocentric coordinate system recommended in the "Upper-Air Meteorological Detection Data Processing Model" (GJB 6069-2007) for upper-altitude wind calculation. The location of the observation station and the balloon's position during BeiDou wind measurement are usually given in longitude, latitude, and altitude, using the WGS-84 geocentric coordinate system. When satellite navigation provides the balloon's latitude, longitude, and altitude for wind measurement, to calculate wind direction and speed, the longitude, latitude, and altitude parameters from the WGS-84 geocentric coordinate system must be converted to three-dimensional polar coordinate parameters in the station-centric coordinate system before calculating wind direction and speed, thus standardizing the positioning and wind measurement algorithm.

[0067] Currently, in upper-air meteorological detection services, upper-air meteorological detection systems based on BeiDou navigation and positioning typically use a sampling frequency of 1Hz, which means acquiring a set of latitude and longitude positioning information every second and calculating wind direction and speed at the second level based on the positioning information at 1-second intervals.

[0068] In its design, this invention uses a BeiDou navigation and positioning-based upper-air meteorological detection system with a sampling frequency of 1Hz. It employs the calculation method in the "Upper-Air Meteorological Detection Data Processing Model" to calculate upper-air wind second data.

[0069] S2, Calculate operational upper-level wind data

[0070] Since the high-altitude wind second data from navigation wind measurement cannot be directly applied to business operations, the data needs to be smoothed per unit time.

[0071] In the design process of this invention, the least squares method is used to fit the data of unit time intervals for smoothing the second data.

[0072] During ascent, the radiosonde aboard a weather balloon, pulled by its tether, undergoes a spiral "pendulum" motion, known as the "pendulum effect," which introduces wind measurement errors. For example, with a commonly used 30m tether, the pendulum effect can cause a positional error of nearly 10m, leading to significant inaccuracies in upper-level wind readings. Therefore, the "pendulum motion" of the radiosonde must be fully considered in calculations. Given the symmetry of the oscillation, the time interval is chosen as an integer multiple of the pendulum period, and compensation methods are used to mitigate this effect.

[0073] The period of a balloon's oscillation can be calculated using the formula for the period of a simple pendulum:

[0074]

[0075] Where L is the length of the hanging rope.

[0076] Depending on the rope length, the oscillation period varies. For example, the existing upper-air meteorological detection uses a 30m long rope, corresponding to an oscillation period of 11s. During smoothing, an interval of integer multiples of the oscillation period is selected for smoothing, such as 11s, 22s, 33s, etc., to reduce the deviation caused by the pendulum effect through data compensation.

[0077] Therefore, in the design process of this invention, the least squares fitting method is used, selecting an integer multiple of the pendulum period (1T, 2T, 3T, etc.) as the fitting time. The upper-level wind data within the fitting time is processed and calculated to obtain the average east-west wind speed u and north-south wind speed v during the fitting period. The upper-level wind data obtained through fitting and smoothing can then be applied to upper-air sounding operations.

[0078] S3, Calculate wind speed fluctuation rate

[0079] While the east-west wind speed *u* and north-south wind speed *v* obtained through the least squares fitting method can be directly applied to upper-air sounding operations, significant differences in upper-air wind data arise when different fitting periods are selected. A shorter fitting period reveals more accurate details of wind speed changes but results in larger measurement errors. Conversely, a longer fitting period leads to greater attenuation of wind speed fluctuations by the smoothing curve but also results in the loss of more wind field details, affecting the accuracy of upper-air wind measurements. In practical operations, an optimal smoothing time interval should be set.

[0080] The most intuitive characteristic of wind speed distribution with altitude is wind speed fluctuation, that is, the oscillation of wind speed with altitude. Therefore, wind speed fluctuation is used to characterize the change of wind speed with altitude. By analyzing the fluctuation of wind speed with altitude obtained from different fitting periods, the optimal fitting period is selected to reduce detection error while retaining more details of wind speed changes.

[0081] To quantitatively describe data jitter, the fitted wind speed jitter rate F is defined by the changes in north-south wind speed v and east-west wind speed u per unit time. The calculation formula is as follows:

[0082]

[0083] Where Fu is the east-west wind speed turbulence rate, and Fv is the north-south wind speed turbulence rate; u i v i t i represents the east-west wind speed, north-south wind speed, and corresponding time when the radiosonde is located at the i-th altitude level; n is the number of upper-level wind fitting periods.

[0084] The north-south wind speed turbulence rate and the east-west wind speed turbulence rate are combined to form the wind speed turbulence rate:

[0085]

[0086] Wind speed fluctuations include deterministic fluctuations and measurement error fluctuations. Deterministic fluctuations are caused by variations in wind speed and atmospheric turbulence, while measurement error fluctuations, also known as random fluctuations, are caused by measurement errors. Since random errors include those caused by the pendulum effect, random oscillations exhibit a certain periodicity. When the fitting time interval increases to an integer multiple of the oscillation period, the random error is partially eliminated. When the time interval increases sufficiently, the random measurement error can be considered essentially eliminated, and the fluctuation at this point is deterministic, primarily caused by variations in wind speed and atmospheric turbulence. In this invention, the fitting time is calculated as an integer multiple of the pendulum period.

[0087] Therefore, the measurement error jitter rate dF can be expressed as:

[0088]

[0089] Where dFu is the measurement error jitter rate of east-west winds, dFv is the measurement error jitter rate of north-south winds, and Fuw and Fvw are the deterministic jitter rates caused by wind speed variations and atmospheric turbulence.

[0090] The combined jitter rate of the measurement error is:

[0091]

[0092] S4. Obtain the deterministic jitter rate of the atmospheric wind field by calculating the change in jitter rate.

[0093] As discussed earlier, during data smoothing, shorter fitting time intervals result in smaller computational layer thicknesses, higher accuracy, and larger errors; conversely, longer fitting time intervals result in larger computational layers, lower accuracy, and smaller errors. Therefore, when the fitting time interval is sufficiently large, it can be assumed that random measurement errors have been largely eliminated, and the deterministic turbulence rates Fuw and Fvw of the atmospheric wind field can then be obtained.

[0094] like Figure 3 The diagram shows the evolution characteristics of upper-level winds corresponding to different smoothing time intervals. In designing this invention, we defined the concept of jitter rate change, which represents the magnitude of change in jitter rate between adjacent fitting periods. To intuitively display the characteristics of upper-level wind jitter rate change obtained from different fitting periods, we sequentially used 1 pendulum period (1T), 2 pendulum periods (2T), ..., j pendulum periods (j*T) as fitting times, iteratively calculating the wind speed jitter rate change. The calculation formula is as follows:

[0095]

[0096] δu j δv jFu represents the change in east-west and north-south wind speed turbulence rates corresponding to two adjacent fitting periods, where j is an integer multiple of the pendulum period used for fitting, such as j = 1, 2, 3... representing 1 pendulum period (1T), 2 pendulum periods (2T), 3 pendulum periods (3T), etc., respectively. j 、Fv j Let be the east-west and north-south wind speed fluctuation rates over j pendulum cycles.

[0097] When the fitting time interval is long enough, the random jitter caused by the error is considered to be eliminated, and the jitter can be considered deterministic jitter. Assuming the fitting time interval gradually increases, the change in jitter rate gradually decreases. When δu j δv j When λ is less than λ (as verified by experiments, this invention assumes λ to be between 0.001 and 0.002), i.e., δu j δv j When the time is sufficiently short, use Fu during the j-th period of the simple pendulum. j 、Fv j The deterministic jitter rate of the atmospheric wind field is obtained by replacing the deterministic jitter rates Fuw and Fvw generated by the wind speed itself and atmospheric turbulence.

[0098] S5. Obtain the optimal multiple of the pendulum period by calculating the rate of change of the jitter rate.

[0099] In the process of data smoothing, the shorter the fitting time interval, the smaller the computational layer thickness, the higher the accuracy, and the larger the error; the longer the fitting time interval, the larger the computational layer, the lower the accuracy, and the smaller the error, but it is easy to lose some of the true state of the wind. Therefore, it is necessary to find the optimal solution for the fitting time interval, computational layer thickness, and error.

[0100] In practical upper-air wind detection operations, the optimal fitting time interval can be found by analyzing the correspondence between the change in jitter rate and the fitting period. This minimizes measurement error while keeping the fitting time interval as small as possible, reducing the thickness of the upper-air wind calculation layer and improving upper-air wind accuracy. Therefore, this invention, in its design process, substitutes the deterministic jitter rate of the atmospheric wind field into the calculation formulas for the measurement error jitter rate dF and the composite jitter rate of the measurement error to obtain the composite jitter rate of the measurement error, thereby acquiring the rate of change of jitter rate and the optimal multiple of the pendulum period. Specifically, this includes:

[0101] S51. Substitute the deterministic jitter rate of the atmospheric wind field into the formula for calculating the measurement error jitter rate dF to obtain the measurement error jitter rate.

[0102]

[0103] Among them dFu kLet dFv be the jitter rate of the east-west wind measurement under k simple pendulum periods. k Let be the measurement error jitter rate of north-south winds under k simple pendulum cycles, and Fuw and Fvw be the deterministic jitter rates obtained by S4 caused by the wind speed itself and atmospheric turbulence;

[0104] S52, Obtain the composite jitter rate of measurement error

[0105] The combined jitter rate of the measurement error over k pendulum periods is:

[0106]

[0107] S53. Obtain the jitter rate change rate. The formula for calculating the jitter rate change rate is:

[0108] Fe k =[1-(dFs) k -dFs k+1 ) / (dFs k-1 -dFs k )]×100%

[0109] Fe k Fe represents the rate of change of jitter rate, indicating the difference in the change of jitter rate before and after the k-th pendulum cycle, and characterizing the degree to which the change of jitter rate decreases as k increases. k The larger the value of Fe, the more significant the decrease in the change in turbulence rate as k increases, indicating that there is still a large measurement error and random turbulence in the wind speed turbulence. k The smaller the value, the less significant the decrease in the change in turbulence rate as k increases, indicating a smaller degree of random turbulence in the measurement error of wind speed turbulence. In actual high-altitude wind detection operations, as k increases, Fe... k When the value is less than ε (ε is generally taken as 30% to 40%), it can be considered that increasing the fitting period contributes little to eliminating random fluctuations. At this time, k is the optimal multiple of the fitting pendulum period.

[0110] S6. Obtain the optimal computational layer thickness

[0111] As shown by the above method, the optimal fitting time can be calculated by using the optimal multiple of the pendulum period. Based on the ascent velocity of the weather balloon, the corresponding optimal calculation layer thickness can be obtained, which can be expressed by the following formula:

[0112] h = w·k·T

[0113] Where h is the optimal calculated layer thickness, w is the balloon's ascent speed, k is the optimal pendulum period multiple obtained through S5, and T is the oscillation period of the lanyard attached to the balloon.

[0114] Example 1:

[0115] During the design process of this invention, high-altitude wind detection experiments were conducted with 750g rubber-coated balloons equipped with 30m hanging ropes, 300g rubber-coated balloons equipped with 6m hanging ropes, and 200g rubber-coated balloons equipped with 4m hanging ropes. The relationship between the fitting time interval and the shaking rate and the calculated layer thickness during the detection experiments with different rope lengths was analyzed and studied.

[0116] Taking a 300g rubber-coated balloon with a 6m hanging rope as an example, Figure 4 As shown, the swing period of the 6m rope is approximately 5s, and the balloon's ascent speed is approximately 5±0.5m / s. The oscillation rates of east-west winds (dashed line), north-south winds (dotted line), and the combined total wind speed (solid line) during the experiment were analyzed.

[0117] The change in jitter rate with the increase of the fitting period shows that the jitter rate gradually decreases as the fitting period increases. The trend of the jitter rate decreasing is as follows: when the fitting period is small, the jitter decreases rapidly. As the fitting period continues to increase, the rate of decrease of the jitter rate gradually slows down. When the fitting period continues to increase to a certain time, the jitter rate tends to decrease to 0.

[0118] Taking the composite wind speed as an example, according to the formulas for the sway rate and the rate of change of sway rate, it can be seen that with a fitting time of 10s (2 times the pendulum period), the sway rate is 0.43, and the rate of change of sway rate is 69.1%; with a fitting time of 30s (6 times the pendulum period), the sway rate is 0.046, and the rate of change of sway rate drops to 40%. As can be seen from the figure, when the fitting time is selected between 30s (6 times the pendulum period) and 40s (8 times the pendulum period), the requirement of a 30%-40% rate of change of sway rate can be met. Therefore, based on the corresponding balloon ascent speed, the optimal calculation layer thickness should be 150-200m.

[0119] Taking the actual detection data from April 8, 2022 as an example, such as Figure 4 As shown, the fitting time should be designed to be 6 to 8 times the period, corresponding to a calculation layer thickness of 146-194m. Calculations indicate that, using a 300g balloon with a 6m lanyard, and controlling the balloon's ascent speed at approximately 5m / s, the optimal calculation layer thickness should be 146-194m to reduce detection errors, improve detection accuracy, and retain more accurate wind speed characteristics.

[0120] This invention fully considers factors such as navigation and positioning errors, pendulum effect, and sideslip effect. By using a high-precision navigation and positioning instrument and carrying out high-altitude wind detection experiments with balloons of various sizes and ropes of various lengths, it adopts a comprehensive data processing method to set the optimal smoothing time interval and determine the optimal calculation layer thickness based on balancing measurement errors and calculation layer thickness, thereby reducing detection errors while retaining more details of wind speed changes.

[0121] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for designing the thickness of a calculation layer based on high-altitude wind detection data using wind speed jitter rate, characterized in that, The method includes the following steps: S1. High-altitude wind detection is carried out by using a navigation and positioning device, with the help of balloons and ropes, and high-altitude wind second data is obtained based on navigation and positioning data; S2, Calculate operational upper-level wind data The least squares fitting method was used, and an integer multiple of the pendulum period was selected as the fitting time. The upper-level wind speed data within the fitting time was processed and calculated to obtain the average east-west wind speed u and north-south wind speed v within the fitting period. S3, Calculate wind speed fluctuation rate The fitted wind speed variability rate F is defined by the changes in north-south wind speed v and east-west wind speed u per unit time; the calculation formula is as follows: Where Fu is the east-west wind speed turbulence rate, and Fv is the north-south wind speed turbulence rate; u i v i t i represents the east-west wind speed, north-south wind speed, and corresponding time when the radiosonde is located at the i-th altitude layer; n is the number of upper-level wind fitting periods. The north-south wind speed turbulence rate and the east-west wind speed turbulence rate are combined to form the wind speed turbulence rate: Wind speed fluctuations include deterministic fluctuations and measurement error fluctuations; The measurement error jitter rate dF is expressed as: Where dFu is the measurement error jitter rate of east-west winds, dFv is the measurement error jitter rate of north-south winds, and Fuw and Fvw are the deterministic jitter rates caused by wind speed variations and atmospheric turbulence. The combined jitter rate of the measurement error is: S4. Obtain the deterministic jitter rate of the atmospheric wind field by calculating the change in jitter rate. The wind speed turbulence rate is iteratively calculated by using one pendulum period (1T), two pendulum periods (2T), ..., j pendulum periods (j*T) as the fitting time. As the fitting time interval gradually increases, the change in turbulence rate gradually decreases. When the change in turbulence rate is less than a certain threshold, the wind speed turbulence rate is used to replace the atmospheric wind field deterministic turbulence rate generated by the wind speed itself and atmospheric turbulence, thereby obtaining the atmospheric wind field deterministic turbulence rate. S5. Obtain the optimal multiple of the pendulum period by calculating the rate of change of the jitter rate. By substituting the deterministic jitter rate of the atmospheric wind field into the calculation formulas for the measurement error jitter rate and the composite jitter rate of the measurement error, the composite jitter rate of the measurement error is obtained, the rate of change of the jitter rate is calculated, and then the optimal pendulum period multiple k is obtained. S6. Obtain the optimal computational layer thickness The optimal fitting time is calculated using the optimal multiple of the pendulum period, and the corresponding optimal calculation layer thickness is obtained based on the ascent velocity of the weather balloon. h = w·k·T Where h is the optimal calculated layer thickness, w is the balloon's ascent speed, k is the optimal pendulum period multiple obtained through S5, and T is the oscillation period of the lanyard attached to the balloon.

2. The method for designing the thickness of the calculation layer based on wind speed fluctuation rate for high-altitude wind detection data as described in claim 1, characterized in that, In step S1, the location of the observation station and the location of the balloon during BeiDou wind measurement are given in terms of longitude, latitude and elevation, using the WGS-84 geocentric coordinate system.

3. The method for designing the thickness of the calculation layer based on wind speed fluctuation rate for high-altitude wind detection data as described in claim 2, characterized in that, When satellite navigation provides the latitude, longitude, and altitude of a balloon's location for wind measurement, if wind direction and speed are to be calculated, the longitude, latitude, and altitude parameters of the WGS-84 geocentric coordinate system must be converted into three-dimensional polar coordinate parameters of the station-centric coordinate system before wind direction and speed are calculated, in order to unify the positioning and wind measurement algorithm.

4. The method for designing the thickness of the upper-altitude wind detection data calculation layer based on wind speed jitter rate as described in claim 3, characterized in that, When detecting wind at high altitudes, a high-altitude meteorological detection system based on BeiDou navigation and positioning is used, with a sampling frequency of 1Hz, which means that a set of latitude and longitude positioning information is acquired every second. Based on the positioning information at 1-second intervals, the wind direction and speed are calculated in seconds.

5. The method for designing the thickness of the calculation layer based on wind speed fluctuation rate for high-altitude wind detection data as described in claim 1, characterized in that, In step S2, the period of the simple pendulum is calculated using the following formula: Where L is the length of the hanging rope.

6. The method for designing the thickness of the calculation layer based on wind speed fluctuation rate for high-altitude wind detection data as described in claim 1, characterized in that, In step S3, deterministic jitter is caused by changes in wind speed and atmospheric turbulence, while measurement error jitter is caused by random measurement errors and is also called random jitter. Random measurement errors include errors caused by the pendulum effect. Therefore, random oscillations have a certain periodicity. When the fitting time interval increases by an integer multiple of the oscillation period, the random measurement error will be partially eliminated. When the time interval increases to a sufficiently large value, it is considered that the random measurement error has been basically eliminated, and the jitter at this time is deterministic jitter.

7. The method for designing the layer thickness of high-altitude wind detection data based on wind speed variability rate as described in any one of claims 1-6, characterized in that, Step S4 specifically includes: Using one pendulum period (1T), two pendulum periods (2T), ..., j pendulum periods (j*T) as the fitting time, the change in wind speed turbulence rate is iteratively calculated using the following formula: δu j δv j Fu represents the change in east-west and north-south wind speed turbulence rates corresponding to two adjacent fitting periods, where j is an integer multiple of the pendulum period used for fitting. j 、Fv j Let j be the east-west and north-south wind speed fluctuation rates under j pendulum cycles; When the fitting time interval is long enough, the random fluctuations caused by the error are considered to be eliminated, and at this point, the wind speed fluctuations are considered deterministic fluctuations. Assuming the fitting time interval gradually increases, the change in the fluctuation rate gradually decreases; when δu j δv j When it is less than λ, i.e., δu j δv j When the time is sufficiently short, use Fu during the j-th period of the simple pendulum. j 、Fv j The deterministic jitter rate of the atmospheric wind field is obtained by replacing the deterministic jitter rates Fuw and Fvw generated by the wind speed itself and atmospheric turbulence.

8. The method for designing the thickness of the calculation layer based on wind speed fluctuation rate for high-altitude wind detection data as described in claim 7, characterized in that, In step S4, λ is between 0.001 and 0.

002.

9. The method for designing the thickness of the upper-altitude wind detection data calculation layer based on wind speed jitter rate as described in claim 7, characterized in that, Step S5 specifically includes the following steps: S51. Substitute the deterministic jitter rate of the atmospheric wind field into the formula for calculating the measurement error jitter rate dF to obtain the measurement error jitter rate. Among them dFu k Let dFv be the jitter rate of the east-west wind measurement under k simple pendulum periods. k Let be the measurement error jitter rate of north-south winds under k simple pendulum cycles, and Fuw and Fvw be the deterministic jitter rates obtained by S4 caused by the wind speed itself and atmospheric turbulence; S52, Obtain the composite jitter rate of measurement error The combined jitter rate of the measurement error over k pendulum periods is: S53. Obtain the jitter rate change rate. The formula for calculating the jitter rate change rate is: Fe k =[1-(dFs k -dFs k+1 ) / (dFs k-1 -dFs k )]×100% Fe k Let Fe be the rate of change of jitter, representing the difference in the change of jitter before and after the k-th pendulum cycle, characterizing the degree to which the change of jitter decreases as k increases; as k increases, Fe... k When the value is less than ε, it is considered that increasing the fitting period contributes little to eliminating random fluctuations. In this case, k is the optimal multiple of the fitting pendulum period.

10. The method for designing the thickness of the upper-altitude wind detection data calculation layer based on wind speed jitter rate as described in claim 9, characterized in that, ε is taken as 30% to 40%.

Images