A gust front identification method based on wind measurement laser radar

Abstract

The application discloses a gust front identification method based on wind measurement laser radar, and relates to the technical field of gust front detection. According to the included angle between the radial direction corresponding to the maximum shear value and the remaining radial directions in the radial wind speed field at the current scanning moment, the method determines the convergence zone identification threshold corresponding to each radial direction, and the greater the included angle, the smaller the corresponding convergence zone identification threshold; then, the convergence zone identification algorithm is used to perform detection according to the convergence zone identification threshold corresponding to each radial direction to obtain the convergence zone identification result at the current scanning moment; when the quality state of the convergence zone identification result is determined to be abnormal, the extrapolation result obtained based on the gust front identification result at the last scanning moment is fused to output the gust front identification result at the current scanning moment. The method can realize continuous detection and identification of the gust front from the far field to the near field in the detection area of the laser radar, can significantly improve the detection effect of the laser radar in the near field, and can greatly improve the overall identification accuracy of the gust front.

Description

Technical Field

[0001] This application relates to the field of gust front detection technology, and in particular to a gust front identification method based on wind-measuring lidar. Background Technology

[0002] Gust fronts typically generate significant low-level wind shear, posing a serious threat to human activities, especially in the aviation sector. Sudden changes in wind speed and direction caused by airflow disturbances can be extremely dangerous during takeoff and landing, as well as during flight. As a severe convective weather phenomenon, gust fronts often accompany mature storm cells or squall lines, driven by thunderstorm winds, and are characterized by their suddenness, destructive power, rapid development, and difficulty in resistance. Their formation mechanism is as follows: the descending airflow within a strong convective cell brings cold air from upper levels to the ground, forming a cold pool. This near-surface area generates a divergent wind field that spreads outwards. Influenced by the cell's movement direction, topography, and weather background, the divergent airflow forms a weak cold front—the gust front—at the forefront of the strong convection. Because the near-surface outflow speed of the descending airflow is often extremely high, gust fronts often manifest as sudden, strong gusts preceding thunderstorms, accompanied by significant wind shear, easily triggering severe meteorological disasters. When a gust front passes over an airport, it causes significant wind shear in the surrounding area, accompanied by a sudden increase in air pressure, a surge in wind speed, and a sharp drop in temperature. This poses a serious threat to aircraft takeoff and landing, potentially causing casualties and economic losses. Therefore, early identification and warning of gust fronts are crucial, allowing for targeted measures to mitigate dangers and reduce losses.

[0003] Currently, gust fronts are mainly identified using lidar based on convergence lines in the radial wind field. However, as the gust front approaches or passes directly above the lidar (overhead), its velocity convergence characteristics often weaken due to the lack of radial velocity components (i.e., the Doppler blind zone effect) and radial alignment with the radar. This means the algorithm's detection capability decreases due to changes in the radar's observation angle. Furthermore, the detection performance of close-range gust fronts can be further deteriorated by non-uniform intensity, different directions, and data anomalies caused by ground clutter. These factors contribute to the less-than-ideal accuracy and reliability of current gust front identification methods, especially when the gust front moves near or directly above the lidar, where the lidar is prone to missing detections. Summary of the Invention

[0004] To address the aforementioned problems and technical requirements, this application proposes a method for identifying gust fronts based on wind-measuring lidar. The technical solution of this application is as follows:

[0005] A method for identifying gust fronts based on wind-measuring lidar, the method comprising: The current scanning time is collected by lidar. The radial wind speed field at the current scan time The radial azimuth corresponding to the maximum shear value in the radial wind speed field is taken as the reference azimuth, and the angles between other radial azimuths and the reference azimuth are used as the basis for further analysis. Determine the convergence zone identification threshold corresponding to the radial orientation. included angle The larger the threshold, the higher the convergence zone identification threshold. The smaller, ; The convergence zone identification algorithm is used to detect convergence zones according to the corresponding convergence zone identification thresholds in each radial orientation, thus obtaining the current scanning time. The convergence zone identification results are GFD; When determining that the convergence zone identification results GFD have a risk of quality anomalies, based on the previous scan time... The gust front identification results are obtained at the current scanning time. The extrapolated results of GFW, and the current scan time The convergence zone identification results GFD and the extrapolated results GFW are fused to obtain the current scanning time. The results of gust front identification.

[0006] Its further technical solution is to, based on the previous scan time The gust front identification results are obtained at the current scanning time. The extrapolation results from GFW include: Using the Dynamic Time Warping (DTW) algorithm to update the previous scan time Gust front identification results and current scanning time The convergence zone identification results are correlated with GFD to determine the previous scan time. The gust front identification results show the movement speed at different front positions. The previous scan time is then analyzed based on the movement speed at different front positions. Extrapolate the gust front identification results to obtain the current scanning time. The extrapolated result is GFW.

[0007] Its further technical solution is to determine the previous scan time. The gust front identification results show the movement speed at different front positions, including: Previous scan time The gust front identification results are discretized into the set of the first front position. The current scan time The convergence zone identification result GFD is discretized into the set of second frontal positions. ; The first forward position set was analyzed using the Dynamic Time Warping (DTW) algorithm. Second forward position set The forward positions in the set are associated, for the first set of forward positions any forward position Determine the set of positions for the second forward line. Center and forward positions Several related forward positions ; When the second forward position is assembled There is only one position related to the forward line. Related forward positions At that time, calculate the position of the forward line. To the forward position The movement speed is used as the previous scan time. The gust front identification results are at the front line position Movement speed at that location; When the second forward position is assembled There are multiple positions related to the forward line. Related forward positions At that time, calculate the position of the forwards respectively. To each forward position The average moving speed is used as the average value of the previous scan time. The gust front identification results are at the front line position The speed of movement at that location.

[0008] Its further technical solution is to... (The sentence is incomplete and requires more context to be fully translated.) The convergence zone identification result GFD and the extrapolated result GFW are fused to obtain the current scanning time. The gust front identification results include: Filter the current scan time The convergence zone identification results in GFD are compared with the current scan time. The extrapolation results show that the distance between GFWs is less than the filtering threshold. The fragments were then compared with the selected convergence zone identification results from GFD and the current scan time. The extrapolated results are fused with GFW to obtain the current scan time. The results of gust front identification.

[0009] A further technical solution involves comparing the selected convergence zone identification results from the GFD with the current scanning time. The extrapolated results are fused with GFW to obtain the current scan time. The gust front identification results include: The discrete points on the GFD fragments of the selected convergence zone identification results are compared with the current scanning time. The extrapolated results are used to construct a new point set from the discrete points on the GFW and then curve fitting is performed to obtain the current scanning time. The results of gust front identification.

[0010] Its further technical solution is that the current scanning time The larger the extrapolation result, the greater the distance between the GFW and the lidar, the higher the screening threshold. The larger the threshold, the better. The value ranges from 0.5km to 2.5km, and is dynamically adjusted linearly with the extrapolation result of the distance between the GFW and the lidar.

[0011] Further technical solutions include: the gust front identification method also includes: At the time of detection of the current scan The convergence zone identification results show that the front length of the GFD is less than the length threshold, and / or, the scan time was detected. When the distance between the convergence zone identification result GFD and the lidar is less than the distance threshold, it is determined that the convergence zone identification result GFD has a quality anomaly risk, and the process is executed based on the previous scan time. The gust front identification results are obtained at the current scanning time. The extrapolation result of GFW is used in the following steps; otherwise, it is determined that the convergence zone identification result GFD does not have a quality anomaly risk, and the current scanning time is used. The convergence zone identification result of GFD is directly used as the current scanning time. The results of gust front identification.

[0012] A further technical solution is that the angle between the reference azimuth and the reference azimuth is... Radial orientation corresponding to the convergence zone identification threshold for:

[0013] in, As the baseline threshold, It is a decay function, and The value of varies with It decreases as it increases.

[0014] Its further technical solution is an attenuation function. or .

[0015] The further technical solution is to obtain the current scanning time. The convergence zone identification results of GFD include: Initialize and set the baseline threshold ; Based on the benchmark threshold The values ​​of are used to calculate the convergence zone identification threshold corresponding to each radial orientation, and the convergence zone identification algorithm is used to detect the convergence zone according to the convergence zone identification threshold corresponding to different radial orientations. When the current scan time is not detected When performing convergence zone identification (GFD) on the data, the baseline threshold is decreased by a fixed step size. It then recalculates the convergence zone identification thresholds for each radial orientation until the current scanning time is detected. The convergence zone identification result GFD or until it is determined that there is no gust front, when the baseline threshold is reached. When the number of winds decreases to the preset minimum threshold or the number of consecutive iterations reaches the preset threshold, it is determined that there is no gust front.

[0016] The beneficial technical effects of this application are: This application discloses a gust front identification method based on wind-measuring lidar. This method adds a convergence zone identification threshold that dynamically changes with radial azimuth to the existing convergence zone identification algorithm. The convergence zone identification threshold decreases as the radial azimuth angle between the radial azimuth and the maximum shear value increases. The extrapolated result GFW for the current scanning time is obtained based on the gust front identification result of the previous scanning time, and this extrapolated result is used as a reference for near-field gust front identification and fused with the convergence zone identification result GFD. This enables continuous detection and identification of far-field to near-field gust fronts in the lidar detection area, enhances the accurate capture capability of gust front identification, significantly improves the near-field detection effect of lidar, and greatly improves the overall accuracy of gust front identification.

[0017] When obtaining the extrapolation result GFW, the DTW algorithm is introduced to achieve precise point-to-point alignment of the gust front contour. This allows for the extraction of more reasonable and refined movement speed information, significantly improving the accuracy of gust front trajectory prediction. This, in turn, improves the accuracy of the extrapolation result GFW, thereby enhancing the accuracy of the fusion result. It can effectively adapt to gust fronts of different shapes and sizes, demonstrating strong adaptability and robustness.

[0018] When fusing GFD and GFW, GFD is first filtered within the GFW range by using a screening threshold before fitting. This can filter out GFD segments with large errors. Furthermore, a screening threshold that dynamically changes with distance can be introduced to further improve the accuracy and reliability of the final fusion result. In particular, for the identification of gust fronts in the near field region, it effectively makes up for the shortcomings of traditional methods and significantly improves the accuracy and completeness of the identification results. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating the effect of gust front recognition in an example.

[0020] Figure 2 This is a flowchart of a gust front identification method in one embodiment.

[0021] Figure 3 This is a schematic diagram illustrating the gust front recognition effect in another example. Detailed Implementation

[0022] The specific embodiments of this application will be further described below with reference to the accompanying drawings.

[0023] This application discloses a method for identifying gust fronts based on wind-measuring lidar, the method comprising: The current scanning time is collected by lidar. The radial wind speed field is calculated, and then the shear field of the entire region is calculated using existing methods. Existing convergence zone identification algorithms, after calculating the shear field of the entire region, use a fixed preset convergence zone identification threshold to detect convergence zones and thus identify gust fronts. The convergence zone identification algorithm can be the GFDA (Gust Front Detection Algorithm), the specific algorithm content of which is not described in this application. However, the identification effect of using a fixed convergence zone identification threshold is poor, especially when the shear intensity changes little, it is prone to missed detections. To solve this problem, this application adds a dynamic threshold adjustment strategy to the existing convergence zone identification algorithm to improve the accuracy and reliability of gust front identification, including: After obtaining the shear field of the entire region, at the current scan time The radial azimuth corresponding to the maximum shear value in the radial wind speed field is the reference azimuth, which is also the radial azimuth of the lidar where the maximum shear value is located. After determining this reference azimuth, the angles between other radial azimuths in the area and this reference azimuth can be calculated. The included angle For example, in one instance, the reference orientation is as follows: Figure 1 As shown in D0, the angle between any radial orientation D1 and the reference orientation D0 is... like Figure 1 As shown.

[0024] Further, based on the angle between each radial azimuth and the reference azimuth. Determine the convergence zone identification threshold corresponding to this radial orientation. included angle The larger the threshold, the higher the convergence zone identification threshold. The smaller the threshold, the better. The convergence zone identification threshold here refers to various core thresholds involved in the convergence zone identification algorithm, including shear peak threshold, shear segment wind speed difference threshold, shear segment azimuth spacing threshold, shear segment distance spacing threshold, characteristic length threshold, gust front length threshold, etc.

[0025] Determine the convergence zone identification thresholds corresponding to different radial orientations. Then, it can be applied to the convergence zone identification algorithm: When using the convergence zone identification algorithm for convergence zone detection, instead of detecting according to a fixed convergence zone identification threshold as in traditional methods, it detects different radial directions according to the convergence zone identification threshold corresponding to each radial direction, thus obtaining the current scanning time. The results of gust front identification.

[0026] This application adjusts the convergence zone identification threshold used in existing convergence zone identification algorithms from a fixed value to a dynamic value that varies with the radial azimuth. The core theoretical basis is based on the Doppler detection principle of lidar and the wind field structure characteristics of gust fronts: at the maximum shear value in the radial wind speed field (reference azimuth), the azimuth of the gust front line is perpendicular to the radar radial direction. At this position, the radial wind speed component is the largest and the convergence characteristics are the most significant. The larger the angle θ between the radial azimuth and the reference azimuth, the closer the front line is to being parallel to the radar radial direction, the stronger the Doppler blind zone effect, and the more obvious the attenuation of the radial wind speed component of the converged airflow. Traditional fixed thresholds are very prone to missed detection. Based on this, this application adopts a dynamic adjustment strategy of "the larger the included angle θ, the smaller the convergence zone identification threshold". In the reference azimuth where the shear signal is strong, a high threshold is set to control false alarms, and in the large included angle azimuth where the shear signal is weak, the threshold is reduced to improve the identification sensitivity. This effectively breaks through the limitations of traditional methods in terms of sensitivity, significantly enhances the accurate capture capability of gust front identification, greatly improves the identification rate of gust fronts in the near field and overhead stages, and realizes continuous detection and identification of far-field to near-field gust fronts in the lidar detection area.

[0027] Although incorporating a dynamic threshold adjustment strategy into the convergence zone identification algorithm can optimize the gust front identification performance, in some cases, especially in the near-field of gust fronts, the convergence zone identification algorithm based on the dynamic threshold adjustment strategy still struggles to identify very complete gust fronts. Therefore, another embodiment includes the following steps, please refer to... Figure 2 The flowchart shown: Step 210: Acquire the current scanning time using lidar. Radial wind speed field.

[0028] Step 220, at the current scan time The radial azimuth at the point of maximum shear in the radial wind speed field is taken as the reference azimuth, and the angles between other radial azimuths and the reference azimuth are used as the basis for further analysis. Determine the convergence zone identification threshold for each radial orientation. .

[0029] Convergence zone identification threshold and included angle The correspondence can be customized, and in one embodiment it can be represented as follows: .in, This is the baseline threshold. It is a decay function, and The value of varies with It decreases as it increases.

[0030] Based on this, a nonlinear decay function is selected. It can further adapt to the attenuation law of the radial wind speed component of lidar, taking into account both detection sensitivity and false alarm rate control.

[0031] In one embodiment, a cosine square form of the attenuation function is chosen. The decay function With the included angle The increase exhibits a non-linear, smooth decay. (Reference bearing) time Convergence zone identification threshold Compared to the baseline threshold No attenuation. hour Convergence zone identification threshold The attenuation is reduced to 0, perfectly matching the variation of the radar radial wind speed component with the included angle, thus avoiding threshold abrupt changes caused by linear attenuation.

[0032] In another embodiment, an exponential decay function is chosen. This function has a gentler decay rate, which can avoid false signals caused by the threshold dropping too quickly at large angles. It is suitable for complex detection scenarios such as airports with a lot of ground clutter and high background noise.

[0033] Step 230: Using a convergence zone identification algorithm, detect convergence zones according to the identification thresholds corresponding to different radial orientations to obtain the current scanning time. The convergence zone identification results are GFD.

[0034] From the above As can be seen from the expression, the convergence zone identification threshold corresponding to each radial azimuth is not only the angle between it and the reference azimuth. Related, and also related to the benchmark threshold. The value is related to the initial setting of the baseline threshold. In one embodiment, the baseline threshold is set during initialization. The initial benchmark threshold is initially selected based on the maximum shear value in the radial wind field at the current scanning moment. If the maximum shear value is lower than the industry-standard minimum shear value, then that minimum shear value is used as the initial benchmark threshold. The values ​​are used to calculate the convergence zone identification threshold for each radial orientation, and the convergence zone identification algorithm is used to detect convergence zones according to the corresponding radial orientation thresholds; when no convergence zone is detected at the current scanning time... When the convergence zone identification result is GFD, it may be due to the baseline threshold. The failure to detect gust fronts due to unreasonable values ​​could also be due to the absence of gust fronts in the region. Therefore, the baseline threshold should be decreased by a fixed step size. (For example, the decreasing step size is set to 10% of the initial baseline threshold), and the convergence zone identification threshold corresponding to each radial orientation is recalculated and detected again. This process is repeated until any of the following iteration termination conditions are met: ① The current scanning time is successfully detected. Upon obtaining the convergence zone identification result GFD, the iteration is immediately terminated; ② Benchmark threshold If the number of iterations is reduced to the preset minimum threshold and no effective GFD is detected, it is determined that there is no gust front in the detection area, and the iteration is terminated; if the number of iterations reaches the preset threshold (e.g., 5 times) and no effective GFD is detected, it is determined that there is no gust front in the detection area, and the iteration is terminated.

[0035] Step 240, obtain the current scan time The convergence zone identification results from GFD are not directly used as the current scan time. Instead of relying on the gust front identification results, it first detects the current scanning time. Does the convergence zone identification result GFD have any quality anomaly risk? This includes: At the time of detection of the current scan The convergence zone identification results show that the front length of the GFD is less than the length threshold, and / or, the scan time was detected. If the distance between the convergence zone identification result GFD and the lidar is less than the distance threshold, it is determined that the convergence zone identification result GFD has a quality abnormality risk and steps 250 and 260 are executed; otherwise, it is determined that the convergence zone identification result GFD does not have a quality abnormality risk and step 270 is executed.

[0036] In this step, the distance between the midpoint of the front line and the lidar is generally taken as the scanning time. The convergence zone identification result is the distance between GFD and LiDAR. The length threshold and distance threshold here can be customized. The distance threshold is generally set to 5km.

[0037] Step 250: When it is determined that the convergence zone identification result GFD has a risk of quality anomalies, based on the previous scan time... The gust front identification results are obtained at the current scanning time. The extrapolated results from GFW include: (1) Use the Dynamic Time Warping (DTW) algorithm to process the previous scan time. Gust front identification results and current scanning time The convergence zone identification results were correlated with GFD to determine the previous scan time. The gust front identification results show the movement speed at different frontal positions, including: Previous scan time The gust front identification results are discretized into the set of the first front position. and the current scan time The convergence zone identification result GFD is discretized into the set of second frontal positions. Each set of forward positions includes multiple discrete forward positions from the corresponding identification results.

[0038] Then, the Dynamic Time Warping (DTW) algorithm is used to analyze the first forward position set. Second forward position set By associating the forward positions in the data, we can obtain the first set of forward positions. Forward position Set up with the second forward position Forward position The resulting associated point pairs, due to the characteristics of the DTW algorithm, and It may not be a perfect one-to-one correspondence, or it may be one-to-many, but regardless of the relationship, for the set of first forward positions... any forward position They can determine the set of positions for the second forward line. Center and forward positions Several related forward positions .

[0039] When the second forward position is assembled There is only one position related to the forward line. Related forward positions At that time, calculate the position of the forward line. To the forward position The movement speed is used as the previous scan time. The gust front identification results are at the front line position Movement speed at location .

[0040] When the second forward position is assembled There are multiple positions related to the forward line. Related forward positions At that time, calculate the position of the forwards respectively. To each forward position The average moving speed is used as the average value of the previous scan time. The gust front identification results are at the front line position Movement speed at location .

[0041] (2) The previous scan time can be obtained from the above steps. The gust front identification results show the movement speed at different front positions, and then the previous scan time can be analyzed based on the movement speed at different front positions. Extrapolate the gust front identification results to obtain the current scanning time. The extrapolated result GFW includes determining the previous scan time. The gust front identification results are at any front position. Movement speed at location Then, the scan time was extrapolated. forward position At the scanning time Move to the extrapolation position Then, for each extrapolated position Curve fitting is performed to obtain the current scan time. The extrapolated result is GFW. Among them, It is the time interval between two scan times.

[0042] This method introduces the DTW algorithm to achieve precise point-to-point alignment of the gust front contour, which can extract more reasonable and refined moving speed information, greatly improving the accuracy of gust front trajectory prediction. It can effectively adapt to gust fronts of different shapes and sizes, showing strong adaptability and robustness, and can improve the accuracy of the extrapolation result GFW.

[0043] Step 260, for the current scan time The convergence zone identification result GFD and the extrapolated result GFW are fused to obtain the current scanning time. After identifying the gust front, the scanning process begins at the next scanning moment starting from step 210.

[0044] When the convergence zone identification result GFD has a risk of quality abnormalities, the convergence zone identification result GFD is often a series of discrete fragments. In order to improve the fusion accuracy, the current scanning time is used as the basis for fusion. Using the extrapolated results as a benchmark, the current scan time is selected. The convergence zone identification results in GFD are compared with the current scan time. The extrapolation results show that the distance between GFWs is less than the filtering threshold. The fragments selected from the convergence zone identification results of GFD are compared with the current scan time. The extrapolated result is GFW, that is, only the two sides of GFW. The GFD and GFW within the range are fused, and segments with large deviations from the GFW are filtered out to improve fusion accuracy. This includes: matching the discrete points on the GFD segments of the selected convergence zone identification results with the current scanning time. The extrapolated results are used to construct a new point set from the discrete points on the GFW and then curve fitting is performed to obtain the current scanning time. The results of gust front identification.

[0045] This step uses the highly accurate extrapolated result GFW as a reference and performs collaborative optimization identification with the convergence zone identification result GFD, which can further improve the overall accuracy of gust front identification, especially significantly improving the near-field detection effect of lidar. For example, in one embodiment, the obtained current scanning time... The convergence zone identification results of GFD are as follows Figure 3 The blue solid line in the image indicates the current scan time. Extrapolation results GFW Figure 3 As shown by the solid black line in the image, the final result obtained by fusion is the current scan time. The gust front identification results are as follows Figure 3 As shown by the solid red line in the image.

[0046] Furthermore, considering the high accuracy of GFD results in the far field but low accuracy in the near field, a screening threshold is set. With the current scanning time The extrapolated result shows that the distance between the GFW and the lidar increases as the distance increases, and decreases as the distance decreases. (Filtering threshold) Its core function is to define the effective screening range of credible GFD fragments, only when the distance between the GFD fragment and the GFW is less than... Only then will the fragment be included in the fusion calculation. The larger the value, the higher the confidence level of the GFD results, and the more dominant the fusion results are in GFD. The smaller the value, the lower the confidence level of the GFD results, and the fusion results are dominated by GFW.

[0047] Step 270: When it is determined that the convergence zone identification result GFD does not have any quality anomaly risk, the current scanning time is... The convergence zone identification result of GFD is directly used as the current scanning time. The results of gust front identification, for example, in Figure 1 In the example, the convergence zone identification result GFD is shown as the blue solid line, and it is directly used as the scanning time. The gust front identification results are then obtained. The scanning process then begins at step 210, proceeding to the next scanning moment.

[0048] The above descriptions are merely preferred embodiments of this application, and this application is not limited to the above embodiments. It is understood that other improvements and variations that can be directly derived or conceived by those skilled in the art without departing from the spirit and concept of this application should be considered to be included within the protection scope of this application.

Claims

1. A method for identifying gust fronts based on wind-measuring lidar, characterized in that, The method for identifying gust fronts includes: The current scanning time is collected by lidar. The radial wind speed field at the current scan time The radial azimuth corresponding to the maximum shear value in the radial wind speed field is taken as the reference azimuth, and the angles between other radial azimuths and the reference azimuth are used as the basis for further analysis. Determine the convergence zone identification threshold corresponding to the radial orientation. included angle The larger the threshold, the higher the convergence zone identification threshold. The smaller, ; The convergence zone identification algorithm is used to detect convergence zones according to the corresponding convergence zone identification thresholds in each radial orientation, thus obtaining the current scanning time. The convergence zone identification results are GFD; When determining that the convergence zone identification results GFD have a risk of quality anomalies, based on the previous scan time... The gust front identification results are obtained at the current scanning time. The extrapolated results of GFW, and the current scan time The convergence zone identification results GFD and the extrapolated results GFW are fused to obtain the current scanning time. The results of gust front identification.

2. The method for identifying gust fronts according to claim 1, characterized in that, Based on the previous scan time The gust front identification results are obtained at the current scanning time. The extrapolation results from GFW include: Using the Dynamic Time Warping (DTW) algorithm to update the previous scan time Gust front identification results and current scanning time The convergence zone identification results are correlated with GFD to determine the previous scan time. The gust front identification results show the movement speed at different front positions. The previous scan time is then analyzed based on the movement speed at different front positions. Extrapolate the gust front identification results to obtain the current scanning time. The extrapolated result is GFW.

3. The method for identifying gust fronts according to claim 2, characterized in that, Determine the previous scan time The gust front identification results show the movement speed at different front positions, including: Previous scan time The gust front identification results are discretized into the set of the first front position. , the current scan time The convergence zone identification result GFD is discretized into the set of second frontal positions. ; The first forward position set was analyzed using the Dynamic Time Warping (DTW) algorithm. Second forward position set The forward positions in the set are associated, for the first set of forward positions any forward position Determine the set of positions for the second forward line. Center and forward positions Several related forward positions ; When the second forward position is assembled There is only one position related to the forward line. Related forward positions At that time, calculate the position of the forward line. To the forward position The movement speed is used as the previous scan time. The gust front identification results are at the front line position Movement speed at that location; When the second forward position is assembled There are multiple positions related to the forward line. Related forward positions At that time, calculate the position of the forwards respectively. To each forward position The average moving speed is used as the average value of the previous scan time. The gust front identification results are at the front line position The speed of movement at that location.

4. The method for identifying gust fronts according to claim 1, characterized in that, The current scanning time The convergence zone identification result GFD and the extrapolated result GFW are fused to obtain the current scanning time. The gust front identification results include: Filter the current scan time The convergence zone identification results in GFD are compared with the current scan time. The extrapolation results show that the distance between GFWs is less than the filtering threshold. The fragments were then compared with the selected convergence zone identification results from GFD and the current scan time. The extrapolated results are fused with GFW to obtain the current scan time. The results of gust front identification.

5. The method for identifying gust fronts according to claim 4, characterized in that, The selected convergence zone identification results GFD fragments are compared with the current scanning time. The extrapolated results are fused with GFW to obtain the current scan time. The gust front identification results include: The discrete points on the GFD fragments of the selected convergence zone identification results are compared with the current scanning time. The extrapolated results are used to construct a new point set from the discrete points on the GFW and then curve fitting is performed to obtain the current scanning time. The results of gust front identification.

6. The method for identifying gust fronts according to claim 5, characterized in that, Current scan time The larger the extrapolation result, the greater the distance between the GFW and the lidar, the higher the screening threshold. The larger the threshold value, the higher the filtering threshold value. The value ranges from 0.5km to 2.5km, and is dynamically adjusted linearly with the extrapolation result of the distance between the GFW and the lidar.

7. The method for identifying gust fronts according to claim 1, characterized in that, The method for identifying gust fronts also includes: At the time of detection of the current scan The convergence zone identification results show that the front length of the GFD is less than the length threshold, and / or the scan time is detected. When the distance between the convergence zone identification result GFD and the lidar is less than the distance threshold, it is determined that the convergence zone identification result GFD has a quality anomaly risk, and the process is executed based on the previous scan time. The gust front identification results are obtained at the current scanning time. The extrapolation result of GFW is used in the following steps; otherwise, it is determined that the convergence zone identification result GFD does not have a quality anomaly risk, and the current scanning time is used. The convergence zone identification result of GFD is directly used as the current scanning time. The results of gust front identification.

8. The method for identifying gust fronts according to claim 1, characterized in that, The angle between the reference orientation and the reference orientation is Radial orientation corresponding to the convergence zone identification threshold for: in, As the baseline threshold, It is a decay function, and The value of varies with It decreases as it increases.

9. The method for identifying gust fronts according to claim 8, characterized in that, decay function or .

10. The method for identifying gust fronts according to claim 8, characterized in that, Get the current scan time The convergence zone identification results of GFD include: Initialize and set the baseline threshold ; Based on benchmark threshold The values ​​of are used to calculate the convergence zone identification threshold corresponding to each radial orientation, and the convergence zone identification algorithm is used to detect the convergence zone according to the convergence zone identification threshold corresponding to different radial orientations. When the current scan time is not detected When performing convergence zone identification (GFD) on the data, the baseline threshold is decreased by a fixed step size. It then recalculates the convergence zone identification thresholds for each radial orientation until the current scanning time is detected. The convergence zone identification result GFD or until it is determined that there is no gust front, when the baseline threshold is reached. When the number of winds decreases to the preset minimum threshold or the number of consecutive iterations reaches the preset threshold, it is determined that there is no gust front.

Images