A high-altitude wind field hourly prediction cycle assimilation correction method for ensuring aircraft route safety

Abstract

The application discloses a high-altitude wind field hourly prediction circulation assimilation correction method for ensuring aircraft route safety, belongs to the field of space weather, and solves the problem of how to improve the vertical layering precision and resolution of WRF mode prediction results; the method comprises the following steps: observation data preprocessing and quality control, setting a background field error covariance and an observation error covariance; constructing a detection data assimilation scheme; judging a weather condition at an initial time, inputting observation data bound at a corresponding time to perform assimilation to obtain an observation field; assimilating the observation field and the background field to generate an analysis field, starting a mode numerical prediction by using the analysis field, and obtaining a high-altitude wind field hourly prediction result; comparing with the observation data, correcting and evaluating the high-altitude wind field hourly prediction result, adjusting the detection data assimilation scheme to perform circulation assimilation correction, and until a prediction result meeting preset standards of an aircraft route and an optimal detection data assimilation scheme are obtained; and the application improves the accuracy of high-altitude wind prediction and ensures the safety of the aircraft route.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of aerospace meteorology, and relates to a high-altitude wind field hourly prediction circulation assimilation correction method for ensuring the safety of an aircraft flight path, in particular, a high-altitude wind hourly prediction circulation assimilation correction method for 20km below 250m stratification. BACKGROUND

[0002] Numerical weather prediction plays a very important role in the safety of an aircraft flight path, and accurate flight path meteorological information is crucial for the safety and trajectory accuracy of an aircraft, especially when a carrier rocket is passing through the atmosphere, the wind load caused by high-altitude strong winds will affect the flight attitude of the rocket and have a certain impact on the structural reliability of the rocket.

[0003] The current mainstream regional model WRF (The Weather Research and Forecasting Model) can give a relatively accurate aviation meteorological prediction, however, with the continuous improvement of flight accuracy, the influence of flight trajectory meteorological factors is also increasingly significant, and China's new generation of carrier rockets have put forward higher requirements for high-altitude wind meteorological support, requiring the entire layer wind field in a specific region above the sky based on height distribution at a specific time, and the vertical resolution requirement is high.

[0004] The high-altitude wind prediction result of the WRF model has the problems of insufficient vertical stratification and insufficient resolution, and cannot accurately describe the detailed information of the high-altitude wind distribution with height, and the vertical stratification cannot meet the requirements of the aircraft flight path support. SUMMARY

[0005] In order to solve the problems of insufficient vertical stratification and insufficient resolution in the high-altitude wind prediction result of the WRF model in a specific region above the sky at a specific time, and the great difference in the assimilation effect of different assimilation methods and different observation data under different weather conditions, the application provides a high-altitude wind field hourly prediction circulation assimilation correction method for ensuring the safety of an aircraft flight path, which pre-processes and quality controls the observation data, estimates the background field error, judges the weather condition at the initial time of the WRF model through the ground observation data in the observation data, performs three-dimensional variational assimilation on three types of multi-source heterogeneous data including microwave radiometer data, wind profile radar data and Beidou navigation detection high-altitude wind data, builds a numerical prediction system with the WRF model as the core, and realizes the 20km below 250m stratification high-altitude wind prediction in a specific region above the sky at a specific time under different weather conditions by using multi-source heterogeneous data in different combination modes, which improves the accuracy of high-altitude wind prediction and ensures the safety of an aircraft flight path.

[0006] The purpose of the application is achieved by the following technical solutions:

[0007] The application discloses a high-altitude wind field hourly prediction circulation assimilation correction method for ensuring flight path safety of an aircraft, and the method comprises the following steps:

[0008] Step one, collecting observation data in a required time period of a task area, wherein the observation data is measured multi-source heterogeneous data and ground observation data; after quality control of historical extreme values of the observation data and a background field deviation, the observation data is converted into a format processable by an obsproc module in a WRF mode; meanwhile, numerical prediction mode product data in the same time period is downloaded to obtain a background field;

[0009] Step two, constructing a background field error covariance estimation model by using default pre-generated background field error information of the WRF mode and a global background error covariance matrix CV3 provided by NCEP and self-provided by the WRF mode;

[0010] Step three, binding multi-source heterogeneous data with the highest accuracy rate of historical high-altitude wind field hourly prediction results under different weather conditions to construct a detection data assimilation scheme; estimating observation error covariance and setting observation error covariance parameters by using the obsproc module; estimating background field error covariance and setting background field error covariance parameters by using the background field error covariance estimation model; judging weather conditions of an initial time of the WRF mode by using measured ground observation data corresponding to the task area, inputting multi-source heterogeneous data bound at a corresponding time of the weather conditions of the initial time into the WRFDA based on the detection data assimilation scheme, and using a three-dimensional variation assimilation method 3DVAR to perform assimilation, so as to obtain an observation field required by the WRF mode;

[0011] Step four, using the three-dimensional variation assimilation method 3DVAR to assimilate the observation field and the background field to generate an analysis field, using the analysis field to start numerical prediction of the WRF mode, and obtaining high-altitude wind field hourly prediction results;

[0012] Step five, comparing the high-altitude wind field hourly prediction results with the observation data to obtain an accuracy rate of the high-altitude wind field hourly prediction results; if the accuracy rate does not meet preset standards of an aircraft flight path, a double-factor rolling MOS correction method is used to correct and evaluate the high-altitude wind field hourly prediction results, the combination mode of the multi-source heterogeneous data is adjusted, the detection data assimilation scheme is updated, and step three is returned to perform circulation assimilation correction; until the accuracy rate of the high-altitude wind field hourly prediction results meeting the preset standards of the aircraft flight path is obtained, at this time, the corresponding detection data assimilation scheme is an optimal detection data assimilation scheme; if the accuracy rate of the high-altitude wind field hourly prediction results meeting the preset standards of the aircraft flight path cannot be obtained, the detection data assimilation scheme corresponding to the highest accuracy rate of the high-altitude wind field hourly prediction results is selected as the optimal detection data assimilation scheme.

[0013] In step two, the constructed background field error covariance estimation model is as follows:

[0014] B = (x background -x true )(x background -x true ) T ;

[0015] B 1 / 2 = U p U v U h ;

[0016] In the formula, B is the background field error covariance, which is a symmetric semi-definite matrix, x background is the background field, x true is the true field, T is the transpose of the matrix; U p U v U h The WRF model default pre-generated background field error information is used, and the global background error covariance matrix CV3 provided by NCEP is used to realize the optimal fusion of the background field and the observation data, wherein U p is a physical transformation or balance operator, U v is a vertical transformation, and U h is a horizontal transformation.

[0017] In step three, the constructed detection data assimilation scheme is,

[0018] Different weather conditions are: sunny and cloudy, cloudy, overcast, and thunderstorm gale four weather conditions;

[0019] Multi-source heterogeneous data are: Beidou sounding data, wind profile radar data and microwave radiometer data; wherein the combination mode of multi-source heterogeneous data to be bound under different weather conditions includes: no observation data, Beidou sounding data, Beidou sounding data combined with wind profile radar data, Beidou sounding data combined with microwave radiometer data, wind profile radar data combined with microwave radiometer data, and Beidou sounding data combined with wind profile radar data and microwave radiometer data six combination modes.

[0020] In step three, the constructed detection data assimilation scheme is:

[0021] When the weather condition is sunny and cloudy, it is bound with Beidou sounding data;

[0022] When the weather condition is cloudy, it is bound with Beidou sounding data combined with wind profile radar data;

[0023] When the weather condition is overcast, it is bound with Beidou sounding data combined with microwave radiometer data;

[0024] When the weather condition is thunderstorm gale, it is bound with Beidou sounding data combined with microwave radiometer data.

[0025] In step 3, the method for determining the initial weather conditions of the WRF model based on the measured ground observation data corresponding to the mission area is as follows:

[0026] The cloud cover in the measured ground observation data corresponding to the mission area is ≤30%, which is considered to be sunny with few clouds;

[0027] 30%<cloud cover<80%, judged as cloudy;

[0028] If the cloud cover is ≥ 80%, or the rainfall in one hour is ≤ 2.5 mm, it is considered cloudy with light rain;

[0029] If the rainfall in one hour is greater than 2.5 mm, or the average wind speed is greater than 8 m / s, it is judged as a thunderstorm with strong winds.

[0030] The objective function of the three-dimensional variational assimilation method 3DVAR is:

[0031]

[0032] Where J(x) is the solution of the cost function, x represents the analysis variable on the WRF model grid, and x background is the background field, B represents the background field error covariance, y is the observation data, y o represents the observation operator, R represents the observation error covariance, and T represents the transpose of the matrix.

[0033] In step 5, the hourly forecast results of the upper-altitude wind field are compared with the observation data to obtain the accuracy of the hourly forecast results of the upper-altitude wind field:

[0034]

[0035] in,

[0036]

[0037]

[0038] Where, f w represents the accuracy of hourly high-altitude wind forecast results obtained from multi-source heterogeneous data bound to weather condition w; where w is the weather condition; a is the weight coefficient of east-west wind, b is the weight coefficient of north-south wind; f mu represents the accuracy of the east-west wind in the hourly forecast results of the upper-air wind field under m, where m represents the detection data assimilation scheme; f mv represents the accuracy of the north-south wind in the hourly forecast results of the high-altitude wind field under m; N represents the judgment result of whether the forecast result meets the wind corresponding to the preset standard of the aircraft route; i represents the height data of the i-th layer; n represents the total number of layers of the forecast wind product; u mi represents the east-west wind at the height of layer i in the hourly forecast results of the upper-altitude wind field under m, uoi represents the i-th layer upper air west-east wind of the observation data; v mi represents the i-th layer height of the upper air wind field hourly prediction result under m south-north wind; v oi represents the i-th layer upper air south-north wind of the observation data.

[0039] In step five, the method for correcting the evaluated upper air wind field hourly prediction result by using the double-factor rolling MOS correction method is:

[0040] y' t+h = b1y t+h + b2(O t -Y t ) + b0

[0041] In the formula, y' t+h is the corrected value at t+h, y t+h is the prediction value at t+h, O t is the grid real value at t, Y t is the prediction value at t; b0, b1 and b2 are the first model coefficient, the second model coefficient and the third model coefficient automatically established by using the least square method, t is the observation data time, h is the difference between the correction time and the observation data time, h = 1, 2,..., 24.

[0042] In step five, if the upper air wind field hourly prediction result accuracy rate that meets the preset standard of the aircraft route cannot be obtained, the method for selecting the detection data assimilation scheme corresponding to the highest upper air wind field hourly prediction result accuracy rate as the optimal detection data assimilation scheme is:

[0043]

[0044] If u mi -u oi ≤ u s , then f is 1; if u mi -u oi > u s , then f is 0.

[0045] If v mi -v oi ≤ v s , then f is 1; if v mi -v oi > v s , then f is 0.

[0046] In the formula, f' wrepresents the maximum value of the accuracy rate of the upper air wind field hourly forecast result under the weather condition w; f mu represents the average accuracy rate of the upper air east-west wind under the detection data assimilation scheme m; f mv represents the average accuracy rate of the upper air north-south wind under the detection data assimilation scheme m; u s is the east-west wind judgment standard in the preset standard of the aircraft route; v s is the north-south wind judgment standard in the preset standard of the aircraft route; f s is the percentage standard in the preset standard of the aircraft route; K 总 represents the total number of samples under the weather condition w; k represents the sample number under the detection data assimilation scheme m.

[0047] In step five, the preset standard of the aircraft route is:

[0048] Under different weather conditions, the overall assimilated data forecast accuracy rate of the upper air wind field hourly forecast of 20km below each 250m layer is: the east-west wind component, the north-south wind component and the wind speed are less than the judgment standard, and the accuracy rate Accuracy is greater than the percentage standard.

[0049] The beneficial effects of the present application are:

[0050] The technical scheme disclosed by the present application solves the problem of insufficient vertical layering and insufficient resolution in the upper air wind forecast result of the WRF model in a specific region at a specific time, and the present application uses the measured Beidou sounding data, wind profile radar data and microwave radiometer data in the multi-source heterogeneous data, which has the characteristics of stable quality, reliability and high vertical detection accuracy, can provide a complete description of the three-dimensional structure of the atmosphere, is one of the most basic data for numerical model prediction, and is a specimen for verifying the reliability of numerical model prediction and other observation data. The data is input into the numerical prediction model for assimilation, improves the prediction accuracy of the atmospheric temperature, humidity and wind profile, and improves the accuracy of weather prediction.

[0051] The application also solves the problem that different assimilation methods and different observation data have great differences in the effect of assimilation under different weather conditions; in view of the prediction and guarantee demand of high-altitude wind with high resolution in aerospace meteorological guarantee, a numerical prediction system taking WRF mode as the core is built by using the combination mode of multiple sources of heterogeneous data of the measured Beidou sounding data, wind profile radar data and microwave radiometer data in the observation data, through observation data preprocessing and quality control, background field error estimation, three-dimensional variation assimilation, combining with local meteorological data, that is, the observation data with the highest accuracy rate of different weather conditions and historical high-altitude wind field hourly prediction results are bound, the detection data assimilation scheme under different weather conditions is designed, and the initial time weather condition of the WRF mode is judged by the measured ground observation data in the observation data, the observation data bound at the corresponding time of the initial time weather condition is input into the WRFDA for assimilation, and the analysis field is generated by assimilating the observation field and the background field, the numerical prediction of the WRF mode is started by using the analysis field, the high-altitude wind field hourly prediction result is obtained, if the accuracy rate of the prediction result does not meet the preset standard of the aircraft route, the double-factor rolling MOS correction method is used to correct and evaluate the high-altitude wind field hourly prediction result, and the combination mode of the multiple sources of heterogeneous data in the detection data assimilation scheme is adjusted, the detection data assimilation scheme is updated and circulates, through the preset standard of the aircraft route of distinguishing the east and west winds and the north and south winds, the 20km below 250m layered high-altitude wind prediction above the specific region at the specific time is realized, the accuracy of the whole layer prediction of the high-altitude wind at the specific time and the specific region in the task is improved, the vertical resolution requirement is high, and it is suitable for high-resolution high-altitude wind hourly prediction, through experiments, the application can meet the standard that the east and west wind components and the north and south wind components are less than or equal to 5m / s (95%), the details of the high-altitude wind distribution with height are accurately described, the vertical layering meets the aircraft route guarantee requirement, and the aircraft route safety is guaranteed. BRIEF DESCRIPTION OF DRAWINGS

[0052] The application will be further described in detail below according to the drawings and embodiments.

[0053] Figure 1 It is a detection data assimilation scheme design flowchart under different weather conditions.

[0054] Figure 2 It is a numerical mode hourly prediction circulation assimilation flowchart.

[0055] Figure 3 It is the wind field prediction accuracy rate from March to April in 2024; wherein each point is the accuracy rate of the comparison between the prediction value and the observation value on that day, the red point is the eastward wind accuracy rate, the blue point is the northward wind accuracy rate, and the average eastward wind and northward wind accuracy rates of 60 days are 95.48% and 95.14%.

[0056] Figure 4is the wind field forecast accuracy rate on March 4, 2024; wherein, the red line is the WRF model prediction result, the green line is the observation value, the U wind accuracy rate is 94.05%, and the V wind accuracy rate is 96.43%.

[0057] Figure 5 is the east-west wind average accuracy rate under different weather conditions from March to April, 2024, and under different assimilation schemes.

[0058] Figure 6 is the north-south wind average accuracy rate under different weather conditions from March to April, 2024, and under different assimilation schemes. DETAILED DESCRIPTION

[0059] In view of the prediction support demand of high-resolution high-altitude wind in aerospace meteorological support, the high-altitude wind field hourly prediction circulation assimilation revision method for supporting aircraft route safety provided by the embodiment of the application mainly comprises: observation data preprocessing and quality control, background field error covariance estimation, different weather condition assimilation scheme design, prediction process design and result revision; the method specifically comprises:

[0060] Step one, observation data preprocessing and quality control: collecting observation data required in a time period of a task area, the observation data being measured multi-source heterogeneous data and ground observation data; after quality control of historical extreme values and background field deviations of the observation data, converting into a format processable by an obsproc module in a WRF model; simultaneously downloading numerical prediction model product data in the same time period to obtain a background field;

[0061] The multi-source heterogeneous data are measured Beidou sounding data, wind profile radar data and microwave radiometer data; the numerical prediction model product data are GFS data, ECMWF data and GRAPES data downloaded from a global model data provider.

[0062] In order to make the observation data effectively enter the WRFDA assimilation system, the observation data need to be preprocessed, the historical extreme values of the observation data are quality controlled together with background field deviations, and are converted into an ASCII format processable by the obsproc in the WRF model;

[0063] The quality control of the historical extreme values of the observation data and the background field deviations is used to identify and eliminate abnormal values in the observation data, and improve the quality and reliability of the observation data; in step one, the method for quality control of the historical extreme values of the observation data and the background field deviations comprises:

[0064] a, marking the observation data exceeding the historical extreme value range as suspicious values;

[0065] b, calculating the difference between the observation data and the background field value to obtain a deviation; marking the observation data with a deviation exceeding a threshold value as abnormal values;

[0066] c. Outlier rejection or correction according to background field value.

[0067] Mainly includes:

[0068] 1. Historical extreme value check: If the observation value of the observation data exceeds the historical extreme value range, it is marked as a suspicious value.

[0069] 2. Background field deviation check: Compare the observation value with the background field value, i.e. observation value-background field value=deviation; if the deviation exceeds the threshold value, it is marked as an outlier.

[0070] 3. Processing outliers: After confirming the outliers, you can choose to remove the data point or correct it according to the background field value.

[0071] Through the above steps, outliers in observation data can be effectively identified and processed, improving the quality and reliability of the data.

[0072] Observation data usually contains errors and needs to be quality controlled during assimilation. The obsproc module estimates the observation error covariance and sets the observation error covariance parameters. The obsproc module is compiled using FORTRAN and mainly includes the following steps:

[0073] (1) Remove observation values outside the specified time and spatial domain;

[0074] (2) Reorder and merge duplicate data reports;

[0075] (3) Calculate the air pressure or altitude according to the static relationship and observation information;

[0076] (4) Check the vertical consistency and superadiabatic condition of multi-layer observations;

[0077] (5) Estimate the observation error according to the pre-defined error file;

[0078] (6) Write the upper air meteorological sounding data file into the ASCII format recognized by the assimilation system;

[0079] The selected multi-source heterogeneous data of the application is stable and reliable, has high vertical detection precision, can provide complete description of three-dimensional structure of atmosphere, is one of the most basic data for numerical mode prediction, and is a specimen for credibility test of numerical mode prediction effect and other observation data. The assimilation of upper air meteorological observation data in the numerical prediction mode can improve the prediction accuracy of atmospheric temperature, humidity and wind profile, and improve the accuracy of weather prediction. However, different assimilation methods and different multi-source heterogeneous data have great differences in the assimilation effect under different weather conditions, and the evaluation of the detection data can establish the data assimilation scheme under different weather conditions. Taking the observation data of the local meteorological observation station on March 3, 2024 as an example, taking the GFS data, ECMWF data and GRAPES data downloaded from the global mode data provider as the numerical prediction mode product data, and taking the GFS data as an example, the technical scheme of the application is described.

[0080] Step two, background field error covariance estimation: the background field error information pre-generated by the WRF mode by default and the global background error covariance matrix CV3 provided by NCEP are used to construct a background field error estimation model, the background field error covariance is estimated by using a simplified model, and the background field error covariance parameters are set;

[0081] There is a correlation between the variables of the WRF mode due to various dynamic constraints, and this mutual relationship also exists between their background errors, and the autocorrelation between the same variables makes the matrix dimension reach 107x107, and the structure is extremely complex. It is extremely difficult to inverse such a super large matrix, and in actual operation, pre-conditioning is solved by controlling variable conversion.

[0082] The constructed background field error estimation model is:

[0083] B=(x background -x true )(x background -x true ) T ;

[0084] The method for estimating the background field error covariance by using a simplified model is:

[0085] B 1 / 2 =U p U v U h ;

[0086] In the formula, B is the background field error covariance, which is a symmetric semi-definite matrix, x background is the background field, x true is the true field, T is the transpose of the matrix; U p U v U hThe WRF model default pre-generated background field error information is used, and the global background error covariance matrix CV3 provided by NCEP is used to realize the optimal fusion of the background field and the observation data.

[0087] U p is a physical transformation or balance operator, the control variables are not correlated with each other, and the control variables are restored to physical variables through physical transformation, and the physical transformation actually reflects the cross relationship between the physical variables, mainly the balance relationship;

[0088] U v is a vertical transformation, and the vertical autocorrelation relationship of each control variable is simulated by isotropic empirical orthogonal functions (EOFs) or by applying a recursive iterative filter;

[0089] U h is a horizontal transformation, which defines the horizontal autocorrelation relationship of the control variables, and is simulated by continuously applying a horizontal recursive filter (approximating a diffusion operator, as affordable), alternative methods include diffusion operators, spectral operators and wavelet operators (diagonal).

[0090] For the research area, the WRF model default pre-generated background field error information and the global background error covariance matrix CV3 provided by NCEP are used, the vertical recursive filter is used to simulate the vertical covariance, and the initial background field is adjusted to approach the observation value according to the input preprocessed upper air wind detection data and microwave radiometer observation data on March 3, 2024, this process is controlled by a weight parameter, and the influence range and degree of the observation data in space and variable dimension are effectively controlled, the optimal fusion of the background field and the observation data is realized, and the accuracy of the model initial field is improved.

[0091] Step three, bind the multi-source heterogeneous data with the highest accuracy of different weather conditions and historical upper air wind field hourly prediction results, construct a detection data assimilation scheme as an initial detection data assimilation scheme; estimate the observation error covariance through the obsproc module and set the observation error covariance parameters; estimate the background field error covariance through the background field error covariance estimation model and set the background field error covariance parameters; determine the initial time weather condition of the WRF model by the measured ground observation data corresponding to the task area, input the multi-source heterogeneous data bound at the corresponding time of the initial time weather condition into WRFDA based on the detection data assimilation scheme, and use the three-dimensional variation assimilation method 3DVAR for assimilation to obtain the observation field required by the WRF model;

[0092] The weather condition at the initial time of the WRF model is determined by local actual ground observation data, and four types of weather are distinguished, including sunny and few clouds, cloudy, overcast and light rain, and thunderstorm and gale. The combination modes of multi-source heterogeneous data include: no observation data, Beidou sounding data, Beidou sounding data combined with wind profile radar data, Beidou sounding data combined with microwave radiometer data, wind profile radar data combined with microwave radiometer data, and Beidou sounding data combined with wind profile radar data and microwave radiometer data.

[0093] The method for determining the weather condition at the initial time of the WRF model by the ground observation data corresponding to the task area is as follows:

[0094] If the cloud cover in the ground observation data corresponding to the task area is less than or equal to 3 / 10, it is determined to be sunny and few clouds.

[0095] If the cloud cover is between 3 / 10 and 8 / 10, it is determined to be cloudy.

[0096] If the cloud cover is greater than or equal to 8 / 10, or the 1-hour rainfall is less than or equal to 2.5 mm, it is determined to be overcast and light rain.

[0097] If the 1-hour rainfall is greater than 2.5 mm, or the average wind speed is greater than 8 m / s, it is determined to be thunderstorm and gale.

[0098] The definitions of the four types of weather conditions are shown in Table 1.

[0099] Table 1

[0100]

[0101] In step three, the method for constructing the detection data assimilation scheme is as follows:

[0102] Statistical historical prediction data, and bind different weather conditions with the observation data with the highest accuracy of historical upper air wind field hourly prediction results; for example: according to the actual observation data from March to April 2024, through repeated experiments by assimilating different observation data, the experimental results are shown in Figure 5 and Figure 6 By statistically analyzing all prediction cases from March to April 2024, the observation data with the highest accuracy of the prediction results obtained is the final binding scheme under this weather condition, and the constructed detection data assimilation scheme includes:

[0103] When the weather condition is sunny and few clouds, the Beidou sounding data in the observation data is bound;

[0104] When the weather condition is cloudy, the Beidou sounding data combined with the wind profile radar data in the observation data is bound;

[0105] When the weather condition is overcast and light rain, the Beidou sounding data combined with the microwave radiometer data in the observation data is bound;

[0106] The weather condition is thunderstorm and gale, and the Beidou sounding data in the observation data is combined with the microwave radiometer data.

[0107] According to the actual local ground observation data, March 3, 2024 is a cloudy and rainy day, and according to the Figure 1 procedure, the Beidou sounding data and microwave radiometer data are taken as the assimilation scheme, the data processing is carried out at the corresponding time of 08 on March 3, 2024, and then the data assimilation is carried out to obtain the observation field.

[0108] Step four, design of the prediction process: using the three-dimensional variational assimilation method 3DVAR, the observation field and the background field are assimilated to generate the analysis field, the analysis field is used to start the numerical prediction of the WRF model, and the high-altitude wind field hourly prediction result is obtained;

[0109] 3DVAR is mainly to solve the minimum value of the objective function, mainly by solving the analysis field (the solution of the minimum value), so that the analysis field and the background field and the observation field are best fitted, and at this time the minimum solution of the objective function is also the analysis field; The objective function of the three-dimensional variational assimilation method 3DVAR is:

[0110]

[0111] Wherein, J(x) is the solution of the cost function, x represents the analysis variable on the WRF model grid, x background is the background field, B represents the background field error covariance, y is the observation data, y o represents the observation operator, R represents the observation error covariance, and T represents the transpose of the matrix.

[0112] Preferably, the background field and the observation field are first input into the system. The background field can be the result of global or regional model prediction, and the background field used in the present example is obtained from the WRF model, and the observation field is the assimilation data put into the assimilation system. It should be noted that the WRFDA system only recognizes the BUFR (Binary Universal Form for the Representation of meteorological data) format, and data format conversion is required before the data is put into WRFDA; in addition, statistical control (quality control) is also required, and the background error covariance and the observation error covariance need to be set, and the setting of both is indispensable, and both will have a great influence on the simulation effect of the system. In the WRFDA assimilation process, parameterization setting is also required, such as the setting of the time window for assimilation, and different parameterization settings will result in different assimilation results. In the process of minimizing the cost function, there are mainly two outer loops, and the number of iterations can be set in the WRFDA related module based on the data type and the total amount. After each cycle ends, the prediction result is updated until the optimal solution, that is, the analysis field, is obtained.

[0113] Step five, result correction: as shown in Figure 2 , the high-altitude wind field hourly prediction result is compared with the observation data to obtain the accuracy rate of the high-altitude wind field hourly prediction result; if the accuracy rate does not meet the preset standard of the aircraft route, a double-factor rolling MOS correction method is used to correct and evaluate the high-altitude wind field hourly prediction result, adjust the combination mode of the multi-source heterogeneous data, update the detection data assimilation scheme, and return to step three for cyclic assimilation correction; until the accuracy rate of the high-altitude wind field hourly prediction result that meets the preset standard of the aircraft route is obtained, at this time, the corresponding detection data assimilation scheme is the optimal detection data assimilation scheme; if the accuracy rate of the high-altitude wind field hourly prediction result that meets the preset standard of the aircraft route cannot be obtained, the detection data assimilation scheme corresponding to the highest accuracy rate of the high-altitude wind field hourly prediction result is selected as the optimal detection data assimilation scheme.

[0114] In step five, the method for comparing the high-altitude wind field hourly prediction result with the observation data to obtain the accuracy rate of the high-altitude wind field hourly prediction result is as follows:

[0115] Among them,

[0116]

[0117] In the formula, f w represents the accuracy rate of the high-altitude wind field hourly prediction result obtained by the multi-source heterogeneous data bound to the weather condition w; wherein w is the weather condition, which is divided into four types of weather: sunny, cloudy, overcast, and light rain, thunderstorm and gale; a is the east-west wind weight coefficient, and b is the south-north wind weight coefficient; fmu represents the accuracy rate of the east-west wind in the high-altitude wind field hourly prediction result under the detection data assimilation scheme m, m represents the detection data assimilation scheme, f mv represents the accuracy rate of the north-south wind in the high-altitude wind field hourly prediction result under the detection data assimilation scheme m; N represents whether the i-th layer prediction result in the high-altitude wind field hourly prediction result under m meets the judgment result of the corresponding wind of the aircraft route preset standard; i represents the i-th layer height data; n represents the total number of the predicted wind products; u mi represents the east-west wind of the i-th layer height in the high-altitude wind field hourly prediction result under m, u oi represents the i-th layer high-altitude east-west wind of the observation data; v mi represents the north-south wind of the i-th layer height in the high-altitude wind field hourly prediction result under m, v oi represents the i-th layer high-altitude north-south wind of the observation data.

[0118] The combination mode of the multi-source heterogeneous data to be bound in the detection data assimilation scheme m under different weather conditions includes six combination modes of no observation data, Beidou sounding data, Beidou sounding data combined with wind profile radar data, Beidou sounding data combined with microwave radiometer data, wind profile radar data combined with microwave radiometer data, and Beidou sounding data combined with wind profile radar data and microwave radiometer data.

[0119] In step five, if the high-altitude wind field hourly prediction result accuracy rate that meets the aircraft route preset standard cannot be obtained, the method for selecting the detection data assimilation scheme corresponding to the highest high-altitude wind field hourly prediction result accuracy rate as the optimal detection data assimilation scheme is:

[0120]

[0121] If u mi -u oi ≤u s , then f is 1; if u mi -u oi >u s , then f is 0;

[0122] If v mi -v oi ≤v s , then f is 1; if v mi -v oi >v s , then f is 0;

[0123] In the formula, f wrepresents the maximum value of the high-altitude wind field hourly prediction accuracy under weather condition w; f mu represents the average accuracy of the high-altitude east-west wind under m; f mv represents the average accuracy of the high-altitude north-south wind under m; N represents whether the i-th layer prediction result in the high-altitude wind field hourly prediction result of the k-th sample under m meets the judgment standard of the corresponding wind of the aircraft route preset standard; u s is the east-west wind judgment standard in the aircraft route preset standard; v s is the north-south wind judgment standard in the aircraft route preset standard; f s is the percentage standard in the aircraft route preset standard; K 总 represents the total number of samples under weather condition w; k represents the k-th sample data under m, u s and v s The default value is 5 m / s; f s The default value is 95%.

[0124] For example, the weather condition is thunderstorm and gale, a detection data assimilation scheme m is used for a time-by-time prediction under the thunderstorm and gale weather condition, and the process of comparing the high-altitude wind field hourly prediction result with the observation data is a sample.

[0125] The formula realizes the effect confirmation of the updated detection data assimilation scheme of the multi-source heterogeneous data under different weather conditions, and realizes the selection of the optimal detection data assimilation scheme under different weather conditions in the system running process.

[0126] In step five, the aircraft route preset standard is:

[0127] Under different weather conditions, the overall assimilated data predicts the high-altitude wind hourly prediction accuracy of 20 km below 250 m layer by layer, and the accuracy rate Accuracy is greater than or equal to the percentage standard when the east-west wind component and the north-south wind component wind speed are less than or equal to the judgment standard.

[0128] The preferred value is that the east-west wind component and the north-south wind component wind speed are less than or equal to 5 m / s, and the accuracy rate Accuracy is greater than or equal to 95%.

[0129] In the embodiment of the application, the aircraft route preset standard is divided into east-west wind and north-south wind, and the high-altitude wind field hourly prediction accuracy is judged respectively, so that the problem of insufficient vertical layering and insufficient resolution in the high-altitude wind prediction result of the WRF model is solved, the accuracy of the high-altitude wind prediction is improved, and the safety of the aircraft route is ensured.

[0130] In step five, the method for correcting and evaluating the high-altitude wind field hourly prediction result by using a double-factor rolling MOS correction method is:

[0131] y' t+h =b1yt+h +b2(O t -Y t )+b0

[0132] wherein y t+h is a corrected value at time t+h, y t+h is a predicted value at time t+h, O t is a grid real value at time t, Y t is a predicted value at time t; b0, b1 and b2 are first model coefficients, second model coefficients and third model coefficients automatically established by using a least square method, t is an observation data time, h is a difference value between a correction time and the observation data time, h = 1, 2,..., 24.

[0133] In order to make the skilled in the art further understand the technical solutions of the present application, two examples are provided for illustration.

[0134] Example 1: Taking the east and west wind components and the north and south wind components with a wind speed of ≤5m / s (95%) as the standard, the prediction results are verified; as shown in the following figure, it is the wind field prediction accuracy rate from March to April in 2024, wherein each point is the accuracy rate of the predicted value compared with the observation value on that day, the red point is the east wind accuracy rate, the blue point is the north wind accuracy rate, and the average east wind and north wind accuracy rates of 60 days are 95.48% and 95.14%, respectively; as can be seen from the figure, under different weather conditions, the overall assimilated data prediction accuracy of the 250m high altitude wind below 20km reaches 95.48% for the U wind component and 95.14% for the V wind component, meeting the preset standard of the aircraft route. Figure 3 Figure 3

[0135] Example 2: The background field is obtained from the WRF model, and the observation field is the assimilated data result of the observation data at 8:00 on March 3, 2024. After the WRF model prediction, the 24-hour prediction result is compared with the actual observation data result at 8:00 on March 4, 2024, and the accuracy rate is 94.05% for the U wind component and 96.43% for the V wind component. The prediction result and the actual observation data are shown in the following figure: Figure 4 Figure 4 which is the wind field prediction accuracy rate on March 4, 2024; wherein the red line is the WRF model prediction result, the green line is the observation value, the U wind accuracy rate is 94.05%, the V wind accuracy rate is 96.43%, the 250m layered high altitude wind below 20km is predicted, and the preset standard of the aircraft route is met.

[0136] The beneficial effects of the embodiment of the present application are:

[0137] ​​​The technical scheme disclosed by the embodiment of the application solves the problem of insufficient vertical layering and resolution in the upper air wind prediction result of the WRF model over a specific region at a specific time, and the measured Beidou sounding data, wind profile radar data and microwave radiometer data in the multi-source heterogeneous data have the characteristics of stable quality, reliability, high vertical detection accuracy, can provide a complete description of the three-dimensional structure of the atmosphere, are one of the most basic data for numerical prediction, and are the specimen for verifying the reliability of numerical prediction and other observation data. The data are input into the numerical prediction model for assimilation, improve the prediction accuracy of the atmospheric temperature, humidity and wind profile, and improve the accuracy of weather prediction.

[0138] The application also solves the problem that different assimilation methods and different observation data have great differences in assimilation effect under different weather conditions; in view of the prediction and support demand of high-resolution upper air wind in aerospace weather support, a numerical prediction system taking the WRF model as the core is built by using the multi-source heterogeneous data combination mode of the measured Beidou sounding data, wind profile radar data and microwave radiometer data in the observation data, through observation data preprocessing and quality control, background field error estimation and three-dimensional variational assimilation, the local meteorological data are combined, that is, the observation data with the highest accuracy rate of historical upper air wind field hourly prediction result under different weather conditions are bound, the detection data assimilation scheme under different weather conditions is designed, the initial time weather condition of the WRF model is judged by the measured ground observation data in the observation data, the observation data bound at the corresponding time of the initial time weather condition are input into the WRFDA for assimilation, the observation field and the background field are assimilated to generate an analysis field, the analysis field is used to start the numerical prediction of the WRF model, the upper air wind field hourly prediction result is obtained, if the accuracy rate of the prediction result does not meet the preset standard of the aircraft route, the double-factor rolling MOS correction method is used to correct and evaluate the upper air wind field hourly prediction result, the combination mode of the multi-source heterogeneous data in the detection data assimilation scheme is adjusted, the detection data assimilation scheme is updated for cyclic assimilation correction, through the preset standard of the aircraft route for distinguishing the east-west wind and the north-south wind, the upper air wind prediction below 20km is realized every 250m layering over a specific region at a specific time, the accuracy of the whole layer prediction of the upper air wind at a specific time and a specific region in the task is improved, the vertical resolution requirement is high, and the application is suitable for high-resolution upper air wind hourly prediction. Through experiments, the application can meet the standard that the east-west wind component and the north-south wind component wind speed is less than or equal to 5m / s (95%), accurately describe the detailed information of the distribution of the upper air wind with height, the vertical layering meets the requirements of the aircraft route support, and the safety of the aircraft route is ensured.

[0139] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for hourly forecasting of high-altitude wind fields by cyclic assimilation and correction to ensure the safety of aircraft routes, characterized by: The method includes: Step 1: Collect observation data for the required time period in the mission area. The observation data are measured multi-source heterogeneous data and ground observation data. After quality control of the historical extreme values ​​and background field deviations of the observation data, convert them into a format that can be processed by the obsproc module in the WRF model. Simultaneously, download the numerical forecast model product data for the same time period to obtain the background field. Step 2: Use the default pre-generated background field error information of the WRF model and the global background error covariance matrix CV3 provided by NCEP in the WRF model to build a background field error covariance estimation model; Step 3: Bind different weather conditions to the multi-source heterogeneous data with the highest accuracy of historical hourly high-altitude wind field forecast results to build a sounding data assimilation scheme; estimate the observation error covariance and set the observation error covariance parameters through the obsproc module; estimate the background field error covariance and set the background field error covariance parameters through the background field error covariance estimation model; determine the initial weather conditions of the WRF model based on the measured ground observation data corresponding to the mission area, and based on the sounding data assimilation scheme, input the multi-source heterogeneous data bound to the initial weather conditions into WRFDA and use the three-dimensional variational assimilation method 3DVAR for assimilation to obtain the observation field required by the WRF model; Step 4: Using the three-dimensional variational assimilation method 3DVAR, the observation field and the background field are assimilated to generate an analysis field, and the WRF model numerical forecast is started using the analysis field to obtain the hourly forecast results of the upper-air wind field; Step 5: Compare the hourly forecast results of the high-altitude wind field with the observation data to obtain the accuracy of the hourly forecast results of the high-altitude wind field; if the accuracy does not meet the preset standard of the aircraft route, the double-factor rolling MOS correction method is used to correct and evaluate the hourly forecast results of the high-altitude wind field, adjust the combination method of multi-source heterogeneous data, update the detection data assimilation plan, and return to step 3 for cyclic assimilation correction; until the accuracy of the hourly forecast results of the high-altitude wind field that meets the preset standard of the aircraft route is obtained, at this time, the corresponding detection data assimilation plan is the optimal detection data assimilation plan; if the accuracy of the hourly forecast results of the high-altitude wind field that meets the preset standard of the aircraft route cannot be obtained, the detection data assimilation plan corresponding to the highest accuracy of the hourly forecast results of the high-altitude wind field is selected as the optimal detection data assimilation plan.

2. The method according to claim 1, wherein In step 2, the background field error covariance estimation model constructed is: B=(x background -x ture )(x background -x true ) T ; B 1 / 2 =U p U v U h ; Where B is the background field error covariance, which is a symmetric semi-positive definite matrix, x background is the background field, x true is the real field, T is the transpose of the matrix; U p U v U h The default pre-generated background field error information of the WRF model is used, and the global background error covariance matrix CV3 provided by NCEP in the WRF model is used to achieve the optimal fusion of the background field and the observation data. p is a physical transformation or equilibrium operator, U v is the vertical transformation, U h For horizontal transformation.

3. The method according to claim 1, wherein In step 3, in the exploration data assimilation scheme constructed, Different weather conditions include: sunny with few clouds, cloudy, overcast with light rain, and thunderstorm with strong wind; The multi-source heterogeneous data include Beidou sounding data, wind profiler radar data and microwave radiometer data. The combination of multi-source heterogeneous data to be bound under different weather conditions includes six combinations: no observation data, Beidou sounding data, Beidou sounding data combined with wind profiler radar data, Beidou sounding data combined with microwave radiometer data, wind profiler radar data combined with microwave radiometer data, and Beidou sounding data combined with wind profiler radar data and microwave radiometer data.

4. The method according to claim 1 or 3, wherein: In step 3, the detection data assimilation scheme constructed is: When the weather condition is sunny with few clouds, it is bound to Beidou sounding data; When the weather condition is cloudy, it is combined with Beidou sounding data and wind profiler radar data; When the weather conditions are cloudy and rainy, the data is combined with the Beidou sounding data and the microwave radiometer data; When the weather conditions are thunderstorms and strong winds, the Beidou sounding data is combined with the microwave radiometer data.

5. The method according to claim 1 or 3, wherein: In step 3, the method for determining the initial weather conditions of the WRF model from the measured ground observation data corresponding to the mission area is as follows: The cloud cover in the measured ground observation data corresponding to the mission area is ≤30%, which is considered to be sunny with few clouds; 30%<cloud cover<80%, judged as cloudy; If the cloud cover is ≥ 80%, or the rainfall in one hour is ≤ 2.5 mm, it is considered cloudy with light rain; If the rainfall in one hour is greater than 2.5 mm, or the average wind speed is greater than 8 m / s, it is judged as a thunderstorm with strong winds.

6. The method according to claim 1, wherein The objective function of the three-dimensional variational assimilation method 3DVAR is: Where J(x) is the solution of the cost function, x represents the analysis variable on the WRF model grid, and x background is the background field, B represents the background field error covariance, y is the observation data, y o represents the observation operator, R represents the observation error covariance, and T represents the transpose of the matrix.

7. The method according to claim 1, wherein In step 5, the hourly forecast results of the upper-altitude wind field are compared with the observation data to obtain the accuracy of the hourly forecast results of the upper-altitude wind field: in, Where, f w represents the accuracy of hourly high-altitude wind forecast results obtained from multi-source heterogeneous data bound to weather condition w; where w is the weather condition; a is the weight coefficient of east-west wind, b is the weight coefficient of north-south wind; f mu represents the accuracy of the east-west wind in the hourly forecast results of the upper-air wind field under m, where m represents the detection data assimilation scheme; f mv represents the accuracy of the north-south wind in the hourly forecast results of the high-altitude wind field under m; N represents the judgment result of whether the forecast result meets the wind corresponding to the preset standard of the aircraft route; i represents the height data of the i-th layer; n represents the total number of layers of the forecast wind product; u mi represents the east-west wind at the height of layer i in the hourly forecast results of the upper-altitude wind field under m, u oi represents the east-west wind at the upper level of the i-th layer of observation data; v mi represents the north-south wind at the height of layer i in the hourly forecast results of the high-altitude wind field under m; v oi Represents the north-south wind at the i-th layer of observation data.

8. The method according to claim 1, wherein In step 5, the double-factor rolling MOS correction method is used to revise and evaluate the hourly forecast results of the upper-altitude wind field: y' t+h =b1y t+h +b2(O t -Y t )+b0 Where y' t+h is the corrected value at time t+h, y t+h is the forecast value at time t+h, O t is the grid point situation at time t, Y t is the forecast value at time t; b0, b1 and b2 are the first model coefficient, second model coefficient and third model coefficient automatically established using the least squares method, t is the observation data time, h is the difference between the correction time and the observation data time, h = 1, 2, ..., 24.

9. The method according to claim 7, wherein In step 5, if the accuracy of the hourly forecast results of the upper-altitude wind field that meets the preset standards of the aircraft route cannot be obtained, the method of selecting the detection data assimilation scheme corresponding to the highest accuracy of the hourly forecast results of the upper-altitude wind field as the optimal detection data assimilation scheme is as follows: If u mi -u oi ≤u s ,but is 1; if u mi -u oi >u s ,but is 0; If v mi -v oi ≤v s ,but is 1; if v mi -v oi >v s ,but is 0; Where f' w Indicates the maximum accuracy of hourly forecast results of high-altitude wind field under weather condition w; f' mu represents the average accuracy of the upper-altitude east-west wind under the detection data assimilation scheme m; f' mv represents the average accuracy of the upper north-south wind under the detection data assimilation scheme m; u s The east-west wind judgment standard in the aircraft route preset standard; s The north-south wind judgment standard in the preset standard of the aircraft route; f s The percentage standard of the preset standard for the aircraft route; K 总 represents the total number of samples under weather condition w; k represents the sample number under the detection data assimilation scheme m.

10. The method according to claim 1 or 8, wherein In step 5, the default standard for the aircraft route is: Under different weather conditions, the hourly forecast accuracy of upper-altitude wind at 250m below 20km layered by the overall assimilated data is: the wind speed of the east-west wind component and the north-south wind component ≤ the judgment standard, and the accuracy ≥ the percentage standard.

Images