A method for forecasting low-altitude wind shear at airports with complex terrain
By combining the forecast data of GFS and WRF modes, and using CALMET power downscale technology, high-precision forecasting of low-altitude wind shear at complex terrain airports is achieved, solving the problem of insufficient forecasting accuracy in the existing technology and improving flight safety.
Patent Information
- Application Number
- CN202410323684.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-21
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-03-21
AI Technical Summary
It is difficult to accurately predict low-altitude wind shear at airports in complex terrain. The existing methods are highly subjective, and the forecast accuracy and accuracy are insufficient, which cannot meet the needs of flight operation guarantee.
The forecast data of the GFS global forecast system mode is used as the initial field of the WRF mode, and the numerical forecast with a horizontal resolution of 1 km and a time resolution of 15 minutes is achieved through the WRF mesoscale numerical weather forecast mode, and the WRF forecast result is used as the initial CALMET The field is used to implement CALMET power downscale forecasting technology to achieve three-dimensional wind field forecasting at airports with horizontal resolution and vertical resolution of 100 meters, and calculate the horizontal shear and vertical shear of the runway reference surface and above 600 meters.
It improves the accuracy and advancement of low-altitude wind shear forecasts, provides objective forecast guidance, reduces flight return and reserve landing caused by low-altitude wind shear, and ensures the safety and normal flight operation.
Smart Images

Figure CN118330780B_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the technical field of aviation meteorological forecasting, and in particular, relates to a method for forecasting low-altitude wind shear at airports with complex terrain. Background Art
[0002] Air transportation is one of the most important forms of transportation in modern society, especially in areas with complex terrain. Since ground transportation construction is relatively difficult, the advantages of air transportation are more prominent. In recent years, with the development of social economy, a considerable number of transport airports have been built in these areas; however, due to the special geographical environment and topography of complex terrain airports, various meteorological elements have great unevenness, the weather is changeable, and the wind field is particularly complex, which seriously affects the safety and normality of flight.
[0003] The horizontal or vertical shear phenomenon of low-altitude wind has the characteristics of small temporal and spatial scale, strong suddenness, and difficulty in accurate detection and forecasting, which directly threatens flight safety. At domestic airports with high incidence of wind shear, the generation of wind shear is mostly related to the complex terrain of the airport. Therefore, forecasting low-altitude wind shear at airports with complex terrain is of great significance to improving the meteorological support capabilities of various airports and ensuring flight safety.
[0004] At present, airport meteorological stations mostly use real-time detection data from ground wind stations, weather radars, wind profiler radars and laser wind radars to identify low-altitude wind shear. This forecasting method is highly subjective, and the forecast precision and accuracy are insufficient. There is a lack of objective basis for the forecast of strong winds, wind shear and turbulence, and it cannot meet the needs of flight operation support. Summary of the invention
[0005] In order to solve the problems raised by the above-mentioned background technology, the present application provides a method for forecasting low-altitude wind shear at airports with complex terrain, which aims to meet the business needs of low-altitude wind shear forecasting at airports, improve the forecast accuracy, increase the forecast lead time, ensure the safety of flight operations, and make up for the insufficient forecasting ability of low-altitude wind shear under local complex terrain conditions, as well as the limitations of real-time detection data in the forecast timeliness of wind shear warning methods.
[0006] The technical solution adopted by the present invention is
[0007] A method for forecasting low-altitude wind shear at airports in complex terrain uses the forecast data of the GFS global forecast system model as the initial field of the WRF model, and realizes numerical forecasts with a horizontal resolution of 1 km and a temporal resolution of 15 minutes around the airport through the WRF mesoscale numerical weather forecast model; uses the WRF forecast results as the CALMET initial field, and applies the CALMET dynamic downscaling forecast technology to realize airport three-dimensional wind field forecast products with a horizontal and vertical resolution of 100 meters and a temporal resolution of 15 minutes; and calculates the horizontal shear and vertical shear within 600 meters of the airport runway datum and above for the predicted wind field.
[0008] This application provides a method for forecasting low-altitude wind shear at an airport with complex terrain. The processing steps are as follows:
[0009] Step 1: System initialization
[0010] The grid setting uses 1°×1°, 3-hour interval GFS forecast data as the large-scale meteorological initial field, and nests the triple nested grid on the mercator projection surface. The WRF version is WRF3.9.1.1, and the grid spacing from the outside to the inside is 25.0km, 5.0km, and 1.0km, respectively.
[0011] The microphysical process uses the VSM 3-type simple ice scheme, the atmospheric radiation scheme is the RRTM longwave and cloud shortwave radiation scheme, the near-surface layer scheme is the Monin-Obukhov scheme, the land surface process scheme is the Noah scheme, the boundary layer scheme is the YSU scheme, the cumulus parameterization scheme is the shallow convection Kain-Fritsch scheme, the ground flux is considered, the snow cover effect is not considered, the influence of clouds is considered, the source format of land use type and soil type data is 1, the number of soil layers in the land surface model is 4, and the urban physics scheme is none;
[0012] Dynamic frame vertical velocity is not suppressed, turbulence and mixing diffusion scheme, turbulence coefficient is horizontal Smagorinsky first-order closure, diff_6th_opt is default, diff_6th_factor is 0.12, basic sea level temperature is 290, top suction thickness is 5000 meters, top suction is set, suction coefficient is 0.2, horizontal diffusion coefficient is default, vertical diffusion coefficient is default, suction coefficient is 0.2, model dynamic frame parameters are non-hydrostatic, positive definite advection is 1, moist_adv_opt is 1;
[0013] Through the WRF mesoscale numerical weather forecast model, forecast data with a horizontal resolution of 1 km and a temporal resolution of 15 minutes are achieved around the airport. The forecast data includes wind field, pressure field and temperature field.
[0014] Step 2: Use GPS to measure the altitude of the airport and surrounding feature points, obtain the airport area terrain through interpolation method, and revise the terrain data used by CALMET, and finally produce terrain data with a spatial resolution of 100 meters.
[0015] The CALMET grid is a three-dimensional grid system consisting of X, Y, and Z directions. The X-axis and Y-axis represent the east-west and south-north directions respectively, which are consistent with the wind direction. The UTM coordinate system is established.
[0016] The terrain dynamics effect is parameterized. The vertical wind speed affected by the terrain is obtained by calculating the wind in the entire area and satisfying the atmospheric stability decreasing exponential function. The terrain dynamics effect on the horizontal wind component is obtained by minimizing the divergence of the initial guess field.
[0017] The wind component of the slope flow is adjusted into the wind field to adjust the aerodynamic impact. The thickness of the slope flow layer varies with the elevation of the slope top. The thermodynamic blocking effect of the terrain on the wind field is parameterized by the local Froude number.
[0018] The results of the third nesting of WRF (horizontal resolution of 1 km, time resolution of 15 minutes) are used as the initial guess field to input into CALMET for dynamic downscaling, and the three-dimensional wind field with a horizontal resolution of 100 meters is output. According to the grid design, the output wind field data has a horizontal resolution of 100 meters, 12 vertical layers, and the heights above the ground are 20 meters, 100 meters, 200 meters, 300 meters, 400 meters, 500 meters, 600 meters, 700 meters, 800 meters, 900 meters, 1600 meters, and 3200 meters.
[0019] Step 3: Calculate the projection of the u and v components of the air volume
[0020] First, the height of the forecast layer grid point from the ground is combined with the corresponding latitude and longitude ground altitude to convert its relative height from the ground into absolute height. Then, the forecast wind speed and direction of the two adjacent grid points above and below the absolute height are decomposed into westerly and southerly components u and v, and the u and v components are projected to the corresponding heights using the inverse distance weighted method respectively. Finally, the interpolated u and v components of the corresponding height of the latitude and longitude grid point are synthesized to obtain the wind direction and speed.
[0021] Step 4: Error analysis, calculate the mean square error, root mean square error and mean absolute error.
[0022] Step 5: Forecast of low-altitude wind shear, vertical shear β1 and horizontal shear β2. Calculate the vertical shear between the runway reference plane and 100 meters above it as the vertical shear at a height of 50 meters above the reference plane, and the vertical shear between 100 meters and 200 meters as the vertical shear of 150 meters, and so on. Calculate all vertical shear β1 respectively; calculate the horizontal shear of two adjacent grid points on the runway reference plane respectively, and take the middle longitude and latitude of the two adjacent grid points as the horizontal shear position, and so on to calculate the horizontal shear β2 of the full projection surface at each level.
[0023] In step 2, firstly, the grid point wind speed is used to obtain w,
[0024]
[0025] Then calculate and convert the vertical velocity W into the terrain following coordinate system,
[0026]
[0027] Where: W is the vertical velocity in the terrain tracking system; w is the vertical velocity in the Cartesian coordinate system;
[0028] V is the wind speed at the grid point, m / s; h t is the terrain height, m; u, v—horizontal wind component, m / s; k is the atmospheric stability power index; z is the vertical coordinate;
[0029] Where: N is the Brunt-Vaisala frequency; θ is the potential temperature; g is the acceleration of gravity; |V| is the wind speed at the grid point.
[0030] S=S e [1-exp(-x / L e )] 1 / 2 , S e =[h·g(Δθ / θ)sinα / (C D +k)] 1 / 2 , L e =h / (C D +k); where: S e is the equilibrium velocity of overland flow; L e is the equilibrium length scale; x is the distance from the top of the mountain; Δθ is the potential temperature difference of the surrounding environment; θ is the potential temperature of the surrounding environment; C D is the ground resistance coefficient; h is the slope flow height; α is the horizontal angle of the slope; k is the entrainment coefficient at the top of the slope flow layer; g is the gravitational acceleration.
[0031] In the method for predicting low-altitude wind shear at an airport with complex terrain, if the Fr value of the grid point is less than the critical Froude number and the wind has an uphill component, the wind direction is adjusted to a direction tangent to the terrain and the wind speed remains unchanged. If the Fr value exceeds the critical Froude number, no adjustment is required.
[0032] Δh t =(h max ) ij -(z) ijK ; Where: Fr is the local Froude number; V is the wind speed at the grid point, m / s; N is the Brunt-Vaisala frequency (1 / s); Δh t is the effective obstacle height, m; (h max ) ij is the maximum terrain height within the influence radius of the grid point (i, j), m; (z) ijK is the height of the grid point (i, j) in the upper layer k.
[0033] The method for predicting low-altitude wind shear at an airport with complex terrain, after component projection, obtains wind components u0 and v0, and synthesizes them into wind speed WS0 and wind direction WD0, the formula is as follows:
[0034]
[0035] The method for predicting low-altitude wind shear at a complex terrain airport is described as follows: θ1=(wd1-wd2)*π / 180; where ws1 and ws2 are the wind speeds of the upper and lower layers respectively, θ1 is the wind direction arc difference between the upper and lower layers wd1 and wd2, and the horizontal wind shear β2 calculation formula is: θ2=(wd3-wd4)*π / 180;
[0036] Among them, let ws3 and ws4 be the wind speeds of two adjacent grid points respectively, and θ2 be the wind direction arc difference between two adjacent grid points wd3 and wd4.
[0037] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:
[0038] 1) Using the forecast data of the GFS (Global Forecast System) model as the initial field of the WRF model, the WRF mesoscale numerical weather forecast model is used to achieve numerical forecasts with a horizontal resolution of 1 km and a temporal resolution of 15 minutes around the airport; using the WRF forecast results as the CALMET initial field, the CALMET dynamic downscaling forecast technology is applied to achieve a three-dimensional airport wind field forecast product with a horizontal and vertical resolution of 100 meters and a temporal resolution of 15 minutes; the horizontal and vertical shears within 600 meters of the runway datum and above are calculated for the predicted wind field, so as to effectively forecast the low-altitude wind shear in the area near the airport.
[0039] 2) The method of the present invention can predict low-altitude wind shear, provide an objective guidance basis for the forecast of low-altitude wind shear at airports with complex terrain, effectively improve the forecast accuracy, increase the forecast lead time, reduce the number of flights turning back or making alternate landings due to low-altitude wind shear, and ensure the safety and normal operation of flights. The method of the present invention will surely produce huge economic and social benefits and has good application prospects. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.
[0041] Figure 1 is a method flow chart of an embodiment of the present application;
[0042] Figure 2 It is a grid setting diagram of the nested scheme of the WRF mesoscale numerical model provided in the embodiment of the present application;
[0043] Figure 3 It is a wind field projection schematic diagram provided in an embodiment of the present application;
[0044] Figure 4 is a comparison diagram of terrain modification provided by an embodiment of the present application;
[0045] Figure 5 This is a comparison chart of actual measurement and forecast provided in the embodiment of the present application;
[0046] The solid line is the measured average wind speed, and the dotted line is the predicted average wind speed: (a) spring, (b) summer, (c) autumn, (d) winter
[0047] Figure 6 It is a frequency distribution diagram of the forecast wind direction and the measured wind direction provided in the embodiment of the present application;
[0048] Figure 7is an example diagram of horizontal shear forecast (at an altitude of 2250 meters) provided in an embodiment of the present application.
[0049] Wind rose diagrams for wind direction observation and forecast at different heights:
[0050] (a) 300 m, (b) 400 m, (c) 500 m, (d) 600 m, (e) 700 m, (f) 800 m, (g) 900 m, (h) 1600 m, (i) 3200 m;
[0051] Figure 8 This is an example diagram of vertical shear forecast (at an altitude of 2200 meters) provided in an embodiment of the present application. DETAILED DESCRIPTION
[0052] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.
[0053] like Figure 1 As shown, step (1), WRF mesoscale numerical weather forecast:
[0054] Step (1.1), data and nested grid solution
[0055] This implementation uses 1°×1°, 3-hour interval GFS (Global Forecast System data) / NCEP (National Centers for Environmental Prediction) forecast data as the large-scale meteorological initial field. The nesting method uses a triple nested grid scheme on the mercator projection surface. The WRF version is WRF3.9.1.1. The grid spacing from the outside to the inside is 25.0km, 5.0km, and 1.0km respectively. The grid settings are as follows: Figure 2 shown.
[0056] Step (1.2), physical solution setup
[0057] The microphysical process scheme adopts the VSM 3 type simple ice scheme, the atmospheric radiation scheme is the RRTM longwave and cloud (Dudhia) shortwave radiation scheme, the near-surface layer scheme is Monin-Obukhov, the land surface process scheme is the Noah scheme, the boundary layer scheme is the YSU scheme, the cumulus parameterization scheme is the shallow convection Kain-Fritsch (new Eta) scheme, the ground flux is considered, the snow cover effect is not considered, the influence of clouds is considered, the source format of land use type and soil type data is 1 (WPS / geogridbut with dominant categories recomputed in real), the number of soil layers in the land surface model is 4, and the urban physics scheme is none.
[0058] Step (1.3), power frame setting
[0059] The vertical velocity is not suppressed, the turbulence and mixing effect diffusion scheme is used, the turbulence coefficient is the horizontal Smagorinsky first-order closure, diff_6th_opt is the default, diff_6th_factor is 0.12, the basic sea level temperature is 290, the top suction thickness is 5000 meters, the top suction is set, the suction coefficient is 0.2, the horizontal diffusion coefficient is the default, the vertical diffusion coefficient is the default, the suction coefficient is 0.2, the model dynamic frame parameters are non-hydrostatic, the positive advection is 1, and moist_adv_opt is 1.
[0060] Step (1.4), WRF mesoscale numerical weather forecast
[0061] Through the WRF mesoscale numerical weather forecast model, the wind field, pressure field, temperature field and other forecast data with a horizontal resolution of 1 km and a time resolution of 15 minutes around the airport are achieved, thus achieving the first step. However, for complex terrain airports with a runway length of only about 3 kilometers, the forecast directly used for the airport's low-level wind field is still too rough, and the CALMET dynamic downscaling forecast technology is also needed to further improve the spatial resolution.
[0062] Step (2), CALMET dynamic downscaling:
[0063] CALMET adjusts the initial field by using terrain dynamics, overland flow, closure effect and divergence minimization principle for the objective analysis field (WRF forecast output meteorological elements). In the three-dimensional wind field forecast process, the dynamic influence of terrain, inclined airflow and blocking effect are considered in detail. The specific steps of CALMET dynamic downscaling are as follows:
[0064] Step (2.1), terrain and underlying surface correction
[0065] The CALMET terrain data used in this application is SRTM3, but since SRTM3 data often has small data gaps and blank areas in water areas, high mountain areas and canyons, most of which are caused by radar echo quality problems, the missing terrain data in the area is checked and corrected.
[0066] In addition, the detection time of SRTM3 data is often earlier than the construction time of the airport. Therefore, GPS is used to measure the altitude of the airport and surrounding feature points, and the airport area terrain is obtained through interpolation method. The terrain data used by CALMET is corrected, and finally the terrain data with a spatial resolution of 100 meters is produced. This application uses 1km resolution land use type, albedo and other data, and uses bilinear interpolation method to obtain 100m resolution data of the airport area.
[0067] Step (2.2), Grid System
[0068] The CALMET grid is a three-dimensional grid system consisting of X, Y, and Z directions. The directions represented by the X-axis and Y-axis are east-west and south-north, respectively, which are consistent with the direction of wind direction. The coordinate system is the UTM coordinate system.
[0069] The calculation formula of the Z-axis (varies with terrain) coordinate in the CALMET model is as follows:
[0070] Z=zh t
[0071] Among them: Z—coordinate system that changes with terrain; z—Cartesian coordinate; h t —Terrain altitude.
[0072] In a coordinate system that changes with terrain, the vertical rate of wind speed is calculated as:
[0073]
[0074] Where: w is the vertical velocity in the Cartesian coordinate system; u, v are the horizontal components of the wind, m / s.
[0075] Step (2.3), terrain dynamics effect
[0076] CALMET uses the scheme proposed by Liu and Yocke to parameterize the terrain dynamic effects. The vertical wind speed affected by the terrain is obtained by calculating the wind in the entire area and satisfying the atmospheric stability decreasing exponential function. The terrain dynamic effects on the horizontal wind component are obtained by minimizing the divergence of the initial guess field.
[0077] First, the grid point wind speed is used to obtain w, and then the vertical velocity W is calculated and converted into the terrain following coordinate system.
[0078]
[0079] Where: W-vertical velocity in terrain tracking system; w-vertical velocity in Cartesian coordinate system; V-grid point wind speed, m / s; h t - Terrain height, m; u, v- horizontal wind components, m / s; k- atmospheric stability power exponent; z- vertical coordinate.
[0080]
[0081] Where: N-Brunt-Vaisala frequency (1 / s); θ-potential temperature, K; g-gravitational acceleration, m / s 2 ; |V|-grid point wind speed.
[0082] Step (2.4), overland flow
[0083] CALMET uses an empirical method to consider the scale of overland flow in complex terrain. Overland flow is calculated using parameters such as terrain slope, slope height, and time. The wind component of the overland flow is adjusted into the wind field to adjust the aerodynamic impact. The thickness of the overland flow layer changes with the elevation of the slope top.
[0084] S=S e [1-exp(-x / L e )] 1 / 2
[0085] S e =[h·g(Δθ / θ)sinα / (C D +k)] 1 / 2
[0086] L e =h / (C D +k)
[0087] Where: S e -Equilibrium velocity of overland flow; L e - equilibrium length scale; x - distance from the top of the mountain; Δθ - potential temperature difference of the surrounding environment; θ - potential temperature of the surrounding environment; C D -ground drag coefficient; h-slope flow height; α-horizontal angle of slope; k-entrainment coefficient at the top of the slope flow layer; g-gravitational acceleration.
[0088] Step (2.5), blocking effect
[0089] The thermodynamic blocking effect of terrain on the wind field is parameterized by the local Froude number. If the Fr value of the grid point is less than the critical Froude number and the wind has an uphill component, the wind direction is adjusted to the direction tangent to the terrain and the wind speed remains unchanged. If it exceeds the critical Froude number, no adjustment is required.
[0090]
[0091] Δh t =(h max ) ij -(z) ijK
[0092] Where: Fr is the local Froude number; V is the grid point wind speed, m / s; N is the Brunt-Vaisala frequency (1 / s); Δh t —Effective obstacle height, m; (h max ) ij —The maximum terrain height within the influence radius of the grid point (i, j), m; (z) ijK —Height of grid point (i,j) in upper layer k, m.
[0093] Step (2.6), CALMET dynamic downscaling
[0094] The results of the third nesting of WRF (horizontal resolution of 1 km, time resolution of 15 minutes) are used as the initial guess field to input into CALMET for dynamic downscaling, and the three-dimensional wind field with a horizontal resolution of 100 meters is output. According to the grid design, the output wind field data has a horizontal resolution of 100 meters, 12 vertical layers, and the heights above the ground are 20 meters, 100 meters, 200 meters, 300 meters, 400 meters, 500 meters, 600 meters, 700 meters, 800 meters, 900 meters, 1600 meters, and 3200 meters.
[0095] Step (3), wind field projection:
[0096] Due to the complex terrain and large altitude difference of the airport and its surrounding areas, the wind field forecast is the height layer above the ground, and the terrain is undulating. Therefore, the actual application effect of the wind field at the absolute height layer is better based on the airport needs. Here, the forecasted equidistant wind field above the ground is vertically interpolated to the corresponding absolute height layer.
[0097] Step (3.1), u, v component projection
[0098] First, the height of the forecast layer grid point from the ground is combined with the corresponding latitude and longitude ground altitude to convert its relative height from the ground into absolute height. Then, the forecast wind speed and direction of the two adjacent grid points above and below the absolute height are decomposed into westerly and southerly components u and v, and the u and v components are projected to the corresponding heights using the inverse distance weighted method respectively. Finally, the interpolated u and v components of the corresponding height of the latitude and longitude grid point are synthesized to obtain the wind direction and speed.
[0099] Step (3.2), calculation method
[0100] Calculation method: Figure 3 As shown, the forecast wind grid point corresponding to a certain longitude and latitude with a distance from the ground of g1 is recorded as A, the wind speed of this point is ws1, and the wind direction is recorded as wd1. The forecast wind grid point with a distance from the ground of g2 is recorded as B, the wind speed of this point is recorded as ws2, and the wind direction is recorded as wd2. The ground altitude of this longitude and latitude is recorded as h. The heights of points A and B from the ground are converted into absolute heights h1 and h2, and h1=g1+h, h2=g2+h;
[0101] The wind speed and direction are decomposed into the westerly and southerly components u and v, with the wind speed in m / s and the wind direction in degrees, u = WS × cos (270-WD), v = WS × sin (270-WD), and the grid point to be interpolated at the same longitude and latitude is denoted as R, and the absolute height is denoted as h (h2 < h < h1). The absolute value of the absolute height difference between point A and point R is denoted as The absolute value of the absolute height difference between point B and point R is recorded as have
[0102] The inverse distance weighted method is applied to project the u and v components of points A and B onto point R. The inverse distance weighted method mainly relies on the power value of the inverse distance. The power parameter can control the influence of the known points on the interpolation value based on the distance from the output point. The distance between the interpolation point and the sample point is used as the weight for weighted averaging. The distance of the sample point from the interpolation point is inversely proportional to the assigned weight. The formula for inverse distance weighted interpolation is: Among them, u0 and v0 are the wind components obtained by projection at the grid point R, N is the number of sample points used in the calculation process (N = 2), α i is the weight of each sample point used in the calculation process, u i and v i is the wind component value at points A and B.
[0103] The calculation formula for determining the weight is: Wherein, P is the exponential value. In this application, the relationship between the weight coefficients is considered to be linear, and the P value here is 1.
[0104] After obtaining the wind components u0 and v0 projected at point R, they are synthesized into wind speed WS0 and wind direction WD0.
[0105] The formula is as follows:
[0106]
[0107] Taking the airport runway plane as the reference plane, the predicted equidistant wind field from the ground is vertically projected to the corresponding absolute altitude layer. The focus of this application is to predict the low-level wind field of the airport. Therefore, the projected wind field is the airport runway reference plane and 100 meters, 200 meters, 300 meters, 400 meters, 500 meters, and 600 meters above it.
[0108] Step (4), error analysis:
[0109] MSE: Mean Squared Error, mean square error refers to the expected value of the square of the difference between the estimated value of the parameter and the true value of the parameter; MSE can evaluate the degree of change of the data. The smaller the value of MSE, the better the accuracy of the prediction model in describing the experimental data.
[0110] RMSE Mean Squared Error: Root mean square error is the arithmetic square root of the mean square error.
[0111]
[0112] MSE: Mean Absolute Error, the mean absolute error is the average of the absolute errors. The mean absolute error can better reflect the actual situation of the prediction value error.
[0113] f i Represents the predicted value, y i Represents the true value.
[0114] Step (5), low-level wind shear forecast:
[0115] Based on the wind field projections at each layer, the vertical and horizontal shears at and above the runway datum plane are calculated respectively at 100m, 200m, 300m, 400m, 500m and 600m.
[0116] Step (5.1), vertical shear
[0117] Assume that ws1 and ws2 are the wind speeds of the upper and lower layers respectively, θ1 is the wind direction arc difference between the upper and lower layers wd1 and wd2, and the vertical wind shear value β1 is calculated as:
[0118] θ1=(wd1-wd2)*π / 180
[0119]
[0120] Step (5.2), horizontal shear
[0121] Assume that ws3 and ws4 are the wind speeds at two adjacent grid points, θ2 is the wind direction arc difference between two adjacent grid points wd3 and wd4, and the calculation formula for horizontal wind shear β2 is:
[0122] θ2=(wd3-wd4)*π / 180
[0123]
[0124] The vertical shear between the runway reference plane and 100 meters above is calculated as the vertical shear at a height of 50 meters above the reference plane, the vertical shear between 100 meters and 200 meters is calculated as the vertical shear of 150 meters, and so on, to calculate all vertical shears β1. The horizontal shear of two adjacent grid points on the runway reference plane is calculated separately, and the middle longitude and latitude of the two adjacent grid points is used as the horizontal shear position, and the horizontal shear β2 of the full projection surface at each level is calculated by analogy.
[0125] Example 1
[0126] The Hengduan Mountains are a typical plateau mountain terrain. Affected by the complex terrain, the low-altitude wind field is extremely chaotic, and wind shear and turbulence are frequent. Among the 19 special airports in China, 10 airports are located in this area. To further illustrate the advantages of the present invention, the Dali Airport in Dali City, Yunnan Province is taken as an example to briefly explain the low-altitude wind shear forecasting method for complex terrain airports.
[0127] The forecast data comes from the Global Forecast System of the National Centers for Environmental Prediction of the United States. The GFS data from January 2016 to December 2018 are used as the initial field of WRF, with a spatial resolution of 1°×1° and a temporal resolution of 3 hours.
[0128] The wind profiler radar data include horizontal wind speed, horizontal wind direction, vertical speed and other data of the Dali Airport wind profiler radar from January 2016 to December 2018.
[0129] WRF Mesoscale Numerical Weather Forecast
[0130] The WRF daily forecast is carried out according to the method in step (1). The GFS data is UTC12:00 of the previous day. The range of the third-level nesting is 131 grid points in the north-south direction and 81 grid points in the east-west direction. The daily forecast duration is 42 hours, and the forecast period is from January 1, 2014 to December 31, 2018.
[0131] CALMET dynamic downscaling
[0132] First, the terrain and underlying surface of the Dali Airport area were modified and corrected according to the method in step (2.1), the existing terrain was measured using GPS, and the horizontal 100-meter resolution terrain of the airport area was obtained through bilinear interpolation.
[0133] Using 1km resolution land use type, albedo and other data, bilinear interpolation method is applied to obtain 100m resolution data of the airport area. The terrain modification is as follows: Figure 4 shown.
[0134] Then, the result of the third nesting of WRF is used as the initial field input into CALMET for dynamic downscaling, and a three-dimensional wind field with a horizontal resolution of 100 meters is output. The number of vertical layers of the output wind field is 12, and the heights above the ground are 20 meters, 100 meters, 200 meters, 300 meters, 400 meters, 500 meters, 600 meters, 700 meters, 800 meters, 900 meters, 1600 meters, and 3200 meters.
[0135] Wind field projection
[0136] According to the method of step (3), with the airport runway plane at an altitude of 2150 meters as the reference plane, the predicted equidistant wind field from the ground is vertically projected to the corresponding absolute height layer, and the wind fields at altitudes of 2150 meters, 2250 meters, 2350 meters, 2450 meters, 2550 meters, 2650 meters, and 2750 meters are projected respectively.
[0137] Forecast verification
[0138] The wind profile observation data and CALMET forecast data were used for evaluation and verification. Figure 5 The test data are shown in Table 1. All the correlation coefficients of the heights passed the 99.9% confidence test, with MAE between 2.70-2.99 and RMSE between 2.59-4.30.
[0139] Table 1 Error analysis between measured horizontal wind speed in wind field and CALMET wind speed WS
[0140] Height / m 300 400 500 600 700 800 900 Number of samples 16228 21910 22018 22089 22042 22024 22103 Correlation coefficient r 0.16 0.33 0.36 0.40 0.44 0.45 0.49 MAE 2.99 2.70 2.71 2.70 2.73 2.79 2.85 RMSE 4.30 3.59 3.63 3.65 3.70 3.77 3.82
[0141] The wind directions are divided into sixteen directions, and the frequency distribution of CALMET forecast wind direction and measured wind direction are statistically analyzed. The results are as follows: Figure 6 shown.
[0142] Low-level wind shear forecast
[0143] Horizontal shear at 2150m, 2250m, 2350m, 2450m, 2550m, 2650m, 2750m. Horizontal shear forecast as follows: Figure 7 As shown; and 2200m, 2300m, 2400m, 2500m, 2600m, 2700m vertical shear, vertical shear forecast as follows Figure 8 shown.
[0144] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.
[0145] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.
Claims
1. A method for predicting low-altitude wind shear at an airport with complex terrain, characterized in that: The forecast data of the GFS global forecast system model is used as the initial field of the WRF model, and the WRF mesoscale numerical weather forecast model is used to achieve a numerical forecast with a horizontal resolution of 1 km and a temporal resolution of 15 minutes around the airport. Using the WRF forecast results as the CALMET initial field, the CALMET dynamic downscaling forecast technology is applied to achieve a three-dimensional airport wind field forecast product with a horizontal and vertical resolution of 100 meters and a time resolution of 15 minutes; the horizontal shear and vertical shear within 600 meters of the airport runway reference plane and above are calculated for the predicted wind field; The processing steps are as follows: Step 1: System initialization The grid setting uses 1°×1°, 3-hour interval GFS forecast data as the large-scale meteorological initial field, and nests the triple nested grid on the mercator projection surface. The WRF version is WRF3.9.1.1, and the grid spacing from the outside to the inside is 25.0km, 5.0km, and 1.0km, respectively. The microphysical process uses the VSM 3-type simple ice scheme, the atmospheric radiation scheme is the RRTM longwave and cloud shortwave radiation scheme, the near-surface layer scheme is the Monin-Obukhov scheme, the land surface process scheme is the Noah scheme, the boundary layer scheme is the YSU scheme, the cumulus parameterization scheme is the shallow convection Kain-Fritsch scheme, the ground flux is considered, the snow cover effect is not considered, the influence of clouds is considered, the source format of land use type and soil type data is 1, the number of soil layers in the land surface model is 4, and the urban physics scheme is none; Dynamic frame vertical velocity is not suppressed, turbulence and mixing diffusion scheme, turbulence coefficient is horizontal Smagorinsky first-order closure, diff_6th_opt is default, diff_6th_factor is 0.12, basic sea level temperature is 290, top suction thickness is 5000 meters, top suction is set, suction coefficient is 0.2, horizontal diffusion coefficient is default, vertical diffusion coefficient is default, suction coefficient is 0.2, model dynamic frame parameters are non-hydrostatic, positive definite advection is 1, moist_adv_opt is 1; Through the WRF mesoscale numerical weather forecast model, forecast data with a horizontal resolution of 1 km and a temporal resolution of 15 minutes around the airport is achieved. The forecast data includes wind field, pressure field and temperature field. Step 2: Use GPS to measure the altitude of the airport and surrounding feature points, obtain the airport area terrain through interpolation method, and revise the terrain data used by CALMET, and finally produce terrain data with a spatial resolution of 100 meters. The CALMET grid is a three-dimensional grid system consisting of X, Y, and Z directions. The X-axis and Y-axis represent the east-west and south-north directions respectively, which are consistent with the wind direction. The UTM coordinate system is established. The terrain dynamics effect is parameterized. The vertical wind speed affected by the terrain is obtained by calculating the wind in the entire area and satisfying the atmospheric stability decreasing exponential function. The terrain dynamics effect on the horizontal wind component is obtained by minimizing the divergence of the initial guess field. The wind component of the slope flow is adjusted into the wind field to adjust the aerodynamic impact. The thickness of the slope flow layer varies with the elevation of the slope top. The thermodynamic blocking effect of the terrain on the wind field is parameterized by the local Froude number. The results of the third nesting of WRF (horizontal resolution of 1 km, time resolution of 15 minutes) are used as the initial guess field to input into CALMET for dynamic downscaling, and the three-dimensional wind field with a horizontal resolution of 100 meters is output; according to the grid design, the output wind field data has a horizontal resolution of 100 meters, 12 vertical layers, and the heights above the ground are 20 meters, 100 meters, 200 meters, 300 meters, 400 meters, 500 meters, 600 meters, 700 meters, 800 meters, 900 meters, 1600 meters, and 3200 meters; Step 3: Calculate the projection of the u and v components of the air volume First, the height of the forecast layer grid point from the ground is combined with the corresponding latitude and longitude ground altitude to convert its relative height from the ground into absolute height. Then, the forecast wind speed and direction of the two adjacent grid points above and below the absolute height are decomposed into westerly and southerly wind components u and v. The u and v components are projected to the corresponding heights using the inverse distance weighted method respectively. Then, the interpolated u and v components of the corresponding height of the latitude and longitude grid point are synthesized to obtain the wind direction and wind speed. Step 4: Error analysis, calculate the mean square error, root mean square error and mean absolute error; Step 5: Forecast of low-altitude wind shear, vertical shear β1 and horizontal shear β2. Calculate the vertical shear between the runway reference plane and 100 meters above it as the vertical shear at a height of 50 meters above the reference plane, and the vertical shear between 100 meters and 200 meters as the vertical shear of 150 meters, and so on. Calculate all vertical shear β1 respectively; calculate the horizontal shear of two adjacent grid points on the runway reference plane respectively, and take the middle longitude and latitude of the two adjacent grid points as the horizontal shear position, and so on to calculate the horizontal shear β2 of the full projection surface at each level.
2. The method for predicting low-altitude wind shear at an airport in complex terrain according to claim 1, characterized in that: In step 2, firstly, the grid point wind speed is used to obtain w, Then calculate and convert the vertical velocity W into the terrain following coordinate system, Where: W is the vertical velocity in the terrain tracking system; w is the vertical velocity in the Cartesian coordinate system; V is the wind speed at the grid point, m / s; h t is the terrain height, m; u, v—horizontal wind component, m / s; k is the atmospheric stability power index; z is the vertical coordinate; Where: N is the Brunt-Vaisala frequency; θ is the potential temperature; g is the gravitational acceleration.
3. The method for predicting low-altitude wind shear at an airport in complex terrain according to claim 1, characterized in that: S=S e [1-exp(-x / L e )] 1 / 2 , S e =[h·g(Δθ / θ)sinα / (C D +k)] 1 / 2 , L e =h / (C D +k); where: S e is the equilibrium velocity of overland flow; L e is the equilibrium length scale; x is the distance from the top of the mountain; Δθ is the potential temperature difference of the surrounding environment; θ is the potential temperature of the surrounding environment; C D is the ground resistance coefficient; h is the slope flow height; α is the horizontal angle of the slope; k is the entrainment coefficient at the top of the slope flow layer; g is the gravitational acceleration.
4. The method for predicting low-altitude wind shear at an airport in complex terrain according to claim 1, characterized in that: If the Fr value of the grid point is less than the critical Froude number and the wind has an uphill component, the wind direction is adjusted to the direction tangent to the terrain and the wind speed remains unchanged. If it exceeds the critical Froude number, no adjustment is required; Δh t =(h max ) ij -(z) ijK ; Where: Fr is the local Froude number; V is the wind speed at the grid point, m / s; N is the Brunt-Vaisala frequency (1 / s); Δh t is the effective obstacle height, m; (h max ) ij is the maximum terrain height within the influence radius of the grid point (i, j), m; (z) ijK is the height of the grid point (i, j) in the upper layer k.
5. The method for predicting low-altitude wind shear at an airport in complex terrain according to claim 1, characterized in that: After the wind components u0 and v0 are projected, they are synthesized into wind speed WS0 and wind direction WD0, and the formula is as follows: The calculation formula of vertical wind shear value β1 is: θ1=(wd-wd2)*π / 180; where ws1 and ws2 are the wind speeds of the upper and lower layers respectively, and θ1 is the wind speed of the upper and lower layers respectively. The difference in wind direction curvature between the lower two layers wd1 and wd2, The calculation formula for horizontal wind shear β2 is: θ2=(wd3-wd4)*π / 180; Among them, let ws3 and ws4 be the wind speeds of two adjacent grid points respectively, and θ2 be the wind direction arc difference between two adjacent grid points wd3 and wd4.
Citation Information
Patent Citations
Refined meteorological element forecasting system of near-ground layer and forecasting method thereof
CN110275224A
Wind field dynamic downscaling method and storage medium
CN110298114A
