Millimeter wave radar data assimilation icing prediction method and system

Abstract

The present application relates to the technical field of power grid icing, and discloses a millimeter wave radar data assimilation icing prediction method and system to improve the icing early warning accuracy. The method comprises: prediction calculation region selection; numerical calculation grid construction; for the prediction region, selecting a WRF regional model as a numerical prediction mode, selecting different radiation, convection and boundary layer physical parameterization schemes in the model, and carrying out simulation prediction tests for at least two icing events; and establishing a comprehensive evaluation Taylor diagram to screen out a combination that can minimize the prediction deviation of the physical elements composed of temperature, wind speed, radiation and water vapor; inversing water vapor from radar reflectivity factor; carrying out assimilation calculation according to the radar inversion water vapor data, establishing a minimization of the target function to realize the best fitting of the observation data and the background information; taking the assimilation calculation result as the initial field of the WRF mode prediction to carry out prediction calculation and obtain the icing prediction result.

Description

Technical Field

[0001] This invention relates to the field of power grid icing technology, and in particular to a millimeter-wave radar data assimilation method and system for icing prediction. Background Technology

[0002] In recent years, icing disasters have become more severe and frequent, especially extreme icing events. Icing grows rapidly and forms heavy layers, and conventional prediction methods, lacking the integration of monitoring information, tend to underestimate the severity of predictions.

[0003] Currently, statistical methods for icing prediction offer good retrospective accuracy for historical icing events, but are unsuitable for extreme re-icing environments. Directly using meteorological numerical models, on the other hand, suffers from inherent biases in the initial field, which accumulate over time, significantly reducing the reliability of the results. Therefore, accurately obtaining the initial field is a fundamental prerequisite for accurate prediction.

[0004] Therefore, it is necessary to disclose a millimeter-wave radar data assimilation icing prediction method and system, which can improve the data quality at the initial moment and improve the accuracy of icing early warning by incorporating radar data into the prediction calculation. Summary of the Invention

[0005] The purpose of this invention is to disclose a millimeter-wave radar data assimilation and icing prediction method and system. The method selects prediction areas based on lines prone to icing, constructs a numerical calculation grid and optimizes the parameterization scheme, establishes a radar reflectivity inversion calculation formula, establishes an assimilation calculation formula, and assimilates the results of the inversion calculation to drive numerical model prediction. This improves the data quality at the initial moment and thus enhances the accuracy of icing early warning.

[0006] To achieve the above objectives, the specific steps of the method of the present invention are as follows:

[0007] (1) Selection of prediction calculation area

[0008] Collect long-term historical icing records to identify high-risk icing areas. Areas prone to icing require focused attention and analysis. Collect historical data on wind speed, temperature, radiation, and water vapor in the study area and surrounding regions, and compile and summarize the meteorological data for the icing period.

[0009] (2) Numerical Computation Grid Construction

[0010] A national land use and water distribution dataset, based on ASTER topographic data with a resolution of 30 meters and the latest Landsat 8 imagery, was created. The dataset uses a 3-kilometer grid to evenly divide key areas of interest into one or more grids. During grid division, special micro-topographical areas such as water bodies and hills were grouped into the same grid whenever possible.

[0011] (3) Optimization of the parameterization scheme

[0012] For the prediction area, the WRF regional model was selected as the numerical weather prediction model. Different combinations of physical parameterization schemes for radiation, convection, and boundary layer were chosen within the model to conduct simulation prediction experiments for multiple icing events. A comprehensive evaluation Taylor diagram was established to select the combination that minimizes the prediction bias of physical elements such as temperature, wind speed, radiation, and water vapor. The Taylor diagram comprehensively considers the relationship between the variance, correlation coefficient, and root mean square error of observed and predicted values; its formula is:

[0013] RMSE = S x 2 +S y 2 -2S x S y r

[0014] In the above formula, S x S is the variance of the observed values. y Let be the variance of the predicted values, r be the correlation coefficient between the observed and predicted values, and RMSE be the root mean square error between the observed and predicted values. Their calculation formulas are as follows:

[0015]

[0016]

[0017]

[0018] Where, x n For the observed value, y n X is the predicted value, X is the average of the observed values, and Y is the average of the predicted values.

[0019] (4) Radar inversion calculation

[0020] Water vapor can be derived from radar reflectivity factor. The formula for retrieving water vapor from reflectivity factor is:

[0021] Z = c1 + c2·log 10 (ρq r )

[0022] Where Z is the reflectivity factor in dBZ, C1 and C2 are constants of 43.1 and 17.5 respectively, ρ is the air density, and q r It's water vapor.

[0023] (5) Variational assimilation of three-dimensional meteorological element information

[0024] Based on the water vapor data retrieved from the radar inversion in step (4), assimilation calculations are performed to establish the minimization of the objective function to achieve the best fit between the observation data and the background information. The objective function is defined as follows:

[0025]

[0026] In the formula, J b The degree of fit between the analysis field and the background field is defined, J. o The degree of fit between the analysis field and the observation field is defined, where x is the analysis field to be determined. b As the background field, y o Let H be the observation field, H be the observation operator, B be the background error covariance matrix, and R be the observation error covariance matrix.

[0027] (6) Assimilation and fusion prediction

[0028] The result of assimilation in step (5) is used as the initial field for prediction by the WRF model to obtain the icing prediction result.

[0029] To achieve the above objectives, the present invention also discloses a millimeter-wave radar data assimilation icing prediction system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the following method steps:

[0030] Step S1: Select the prediction calculation area;

[0031] Step S2: Numerical computation grid construction;

[0032] Step S3: Calculate the optimal selection of the parameterization scheme; specifically including:

[0033] For the prediction area, the WRF regional model was selected as the numerical prediction model. Different radiation, convection and boundary layer physical parameterization schemes were selected and combined in the model to carry out simulation prediction experiments for at least two icing events. A comprehensive evaluation Taylor diagram was established to select the combination that minimizes the prediction deviation of the physical elements composed of temperature, wind speed, radiation and water vapor.

[0034] Step S4: Determine water vapor based on radar reflectivity factor;

[0035] Step S5: Variational assimilation of three-dimensional meteorological element information; specifically including:

[0036] Assimilation calculations are performed based on water vapor data retrieved from radar, and the minimization of the objective function is established to achieve the best fit between the observation data and background information.

[0037] Step S6, Assimilation and Fusion Prediction; specifically includes:

[0038] The results of the assimilation calculation are used as the initial field for WRF model prediction to obtain the icing prediction result.

[0039] The specific calculation process for each step can be referred to the above formula, and will not be elaborated further.

[0040] The present invention has the following beneficial effects:

[0041] 1. This invention cleverly converts radar reflectivity information into meteorological element values ​​by inverting water particles for prediction.

[0042] 2. This invention has good versatility and can be used for short-term icing warning calculations in different regions.

[0043] 3. By employing the technology of this invention, it is possible to conduct real-time radar assimilation icing early warning, integrate monitoring information into prediction, and iteratively improve the accuracy of icing prediction.

[0044] The present invention will now be described in further detail with reference to the accompanying drawings. Attached Figure Description

[0045] The accompanying drawings, which form part of this application, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0046] Figure 1 This is a schematic diagram of the millimeter-wave radar data assimilation and icing prediction method disclosed in an embodiment of the present invention.

[0047] Figure 2 This is a Taylor evaluation diagram disclosed in an embodiment of the present invention. Detailed Implementation

[0048] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0049] Example 1

[0050] This embodiment uses the prediction of a 110kV line in Hunan as an example, referring to... Figure 1 The millimeter-wave radar data assimilation icing prediction method in this embodiment includes the following steps:

[0051] (1) Selection of prediction calculation area

[0052] Collect long-term historical icing records to identify high-risk icing areas. Areas prone to icing require focused attention and analysis. Collect historical data on wind speed, temperature, radiation, and water vapor in the study area and surrounding regions, and compile and summarize the meteorological data for the icing period.

[0053] (2) Numerical Computation Grid Construction

[0054] A national land use and water distribution dataset, based on ASTER topographic data with a resolution of 30 meters and the latest Landsat 8 imagery, was created. The dataset uses a 3-kilometer grid to evenly divide key areas of interest into one or more grids. During grid division, special micro-topographical areas such as water bodies and hills were grouped into the same grid whenever possible.

[0055] (3) Optimization of the parameterization scheme

[0056] For the prediction area, the WRF regional model was selected as the numerical weather prediction model. Different combinations of physical parameterization schemes for radiation, convection, and boundary layer were chosen within the model to conduct simulation prediction experiments for multiple icing events. A comprehensive evaluation Taylor diagram was established to select the combination that minimizes the prediction bias of physical elements such as temperature, wind speed, radiation, and water vapor. (Refer to...) Figure 2 The Taylor diagram shown can comprehensively consider the relationship between the variance, correlation coefficient, and root mean square error of observed and predicted values. Its formula is:

[0057] RMSE = S x 2 +S y 2 -2S x S y r

[0058] In the above formula, S x S is the variance of the observed values. y Let be the variance of the predicted values, r be the correlation coefficient between the observed and predicted values, and RMSE be the root mean square error between the observed and predicted values. Their calculation formulas are as follows:

[0059]

[0060]

[0061]

[0062] Where, x n For the observed value, y n X is the predicted value, X is the average of the observed values, and Y is the average of the predicted values.

[0063] (4) Radar inversion calculation

[0064] Water vapor can be derived from radar reflectivity factor. The formula for retrieving water vapor from reflectivity factor is:

[0065] Z = c1 + c2·log 10 (ρq r)

[0066] Here, Z is the reflectivity factor in dBZ, C1 and C2 are constants of 43.1 and 17.5 respectively, ρ is the air density, and q... r It's water vapor.

[0067] (5) Variational assimilation of three-dimensional meteorological element information

[0068] Based on the water vapor data retrieved from the radar inversion in step (4), assimilation calculations are performed to establish the minimization of the objective function to achieve the best fit between the observation data and the background information. The objective function is defined as follows:

[0069]

[0070] In the formula J b The degree of fit between the analysis field and the background field is defined, J. o The degree of fit between the analysis field and the observation field is defined, where x is the analysis field to be determined. b As the background field, y o Let H be the observation field, H be the observation operator, B be the background error covariance matrix, and R be the observation error covariance matrix.

[0071] (6) Assimilation and fusion prediction

[0072] The result of assimilation in step (5) is used as the initial field for prediction by the WRF model to obtain the icing prediction result.

[0073] Example 2

[0074] This embodiment discloses a millimeter-wave radar data assimilation icing prediction system, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the following method steps:

[0075] Step S1: Select the prediction calculation area.

[0076] Step S2: Numerical computation grid construction.

[0077] Step S3: Calculate and optimize the selection of the parameterization scheme. This specifically includes:

[0078] For the prediction area, the WRF regional model was selected as the numerical prediction model. Different radiation, convection and boundary layer physical parameterization schemes were selected and combined in the model to carry out simulation prediction experiments for at least two icing events. A comprehensive evaluation Taylor diagram was established to select the combination that minimizes the prediction deviation of the physical elements composed of temperature, wind speed, radiation and water vapor.

[0079] Step S4: Determine the water vapor based on the radar reflectivity factor.

[0080] Step S5: Variational assimilation of three-dimensional meteorological element information. Specifically, this includes:

[0081] Assimilation calculations are performed based on water vapor data retrieved from radar, and the minimization of the objective function is established to achieve the best fit between the observation data and background information.

[0082] Step S6, Assimilation and Fusion Prediction. Specifically, this includes:

[0083] The results of the assimilation calculation are used as the initial field for WRF model prediction to obtain the icing prediction result.

[0084] Optionally, the Taylor diagram comprehensively considers the relationship between the variance, correlation coefficient, and root mean square error of the observed and predicted values, and its formula is:

[0085] RMSE = S x 2 +S y 2 -2S x S y r

[0086] In the formula, S x S is the variance of the observed values. y Let be the variance of the predicted values, r be the correlation coefficient between the observed and predicted values, and RMSE be the root mean square error between the observed and predicted values; and:

[0087]

[0088]

[0089]

[0090] Where, x n For the observed value, y n X is the predicted value, X is the average of the observed values, and Y is the average of the predicted values.

[0091] Optionally, the specific formula for the reflectivity factor to invert water vapor is as follows:

[0092] Z = c1 + c2·log 10 (ρq r )

[0093] Where Z is the reflectivity factor, C1 and C2 are constants of 43.1 and 17.5 respectively, ρ is the air density, and q r It's water vapor.

[0094] Optionally, the objective function is defined as:

[0095]

[0096] In the formula, J b The degree of fit between the analysis field and the background field is defined, J. o The degree of fit between the analysis field and the observation field is defined, where x is the analysis field to be determined. b As the background field, y o Let H be the observation field, H be the observation operator, B be the background error covariance matrix, and R be the observation error covariance matrix.

[0097] In summary, the beneficial effects of the methods and systems disclosed in the embodiments of the present invention include at least the following:

[0098] 1. This invention cleverly converts radar reflectivity information into meteorological element values ​​by inverting water particles for prediction.

[0099] 2. This invention has good versatility and can be used for short-term icing warning calculations in different regions.

[0100] 3. By employing the technology of this invention, it is possible to conduct real-time radar assimilation icing early warning, integrate monitoring information into prediction, and iteratively improve the accuracy of icing prediction.

[0101] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting icing using millimeter-wave radar data assimilation, characterized in that, include: Step S1: Select the prediction calculation area; Step S2: Numerical computation grid construction; Step S3: Calculate the optimal selection of the parameterization scheme; Specifically, it includes: For the prediction area, the WRF regional model was selected as the numerical prediction model. Different radiation, convection and boundary layer physical parameterization schemes were selected and combined in the model to carry out simulation prediction experiments for at least two icing events. A comprehensive evaluation Taylor diagram was established to select the combination that minimizes the prediction deviation of the physical elements composed of temperature, wind speed, radiation and water vapor. Step S4: Determine water vapor based on radar reflectivity factor; Step S5: Variational assimilation of three-dimensional meteorological element information; specifically including: Assimilation calculations are performed based on water vapor data retrieved from radar, and the minimization of the objective function is established to achieve the best fit between the observation data and background information. Step S6, Assimilation and Fusion Prediction; specifically includes: The results of the assimilation calculation are used as the initial field for WRF model prediction to obtain the icing prediction results.

2. The method according to claim 1, characterized in that, The Taylor diagram comprehensively considers the relationship between the variance, correlation coefficient, and root mean square error of the observed and predicted values. Its formula is: RMSE=S x 2 +S y 2 -2S x S y r In the formula, S x S is the variance of the observed values. y Let be the variance of the predicted values, r be the correlation coefficient between the observed and predicted values, and RMSE be the root mean square error between the observed and predicted values; and: Where, x n For the observed value, y n X is the predicted value, X is the average of the observed values, and Y is the average of the predicted values.

3. The method according to claim 1 or 2, characterized in that, The specific formula for the relationship between reflectivity factor and water vapor is as follows: Z=c1+c2·log 10 (ρq r ) Where Z is the reflectivity factor, C1 and C2 are constants of 43.1 and 17.5 respectively, ρ is the air density, and q r It's water vapor.

4. The method according to claim 3, characterized in that, The objective function is defined as follows: In the formula, J b The degree of fit between the analysis field and the background field is defined, J. o The degree of fit between the analysis field and the observation field is defined, where x is the analysis field to be determined. b As the background field, y o Let H be the observation field, H be the observation operator, B be the background error covariance matrix, and R be the observation error covariance matrix.

5. A millimeter-wave radar data assimilation and icing prediction system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method described in any one of claims 1 to 4.

Images