A method for generating virtual wind lidar based on error analysis

Abstract

The invention belongs to the technical field of radar simulation, and provides a method for generating a virtual wind measuring laser radar based on error analysis, comprising: constructing a virtual wind measuring laser radar based on different data processing methods according to a constructed ideal atmospheric wind field data set, generating a virtual wind measuring laser radar detection data set and calculating the generation time, determining the wind speed error distribution under different ideal wind speed wavelengths and different radar distance resolutions and the ratio of different ideal wind speed wavelengths and different radar distance resolutions through error analysis; calculating the actual deviation value of the error fitting function based on different data processing methods and the error fitting function of a benchmark data processing method; determining the optimal data processing method of the virtual wind measuring laser radar according to the generation time and the actual deviation value to generate a final virtual wind measuring laser radar. The invention overcomes the problem in the prior art that the simulation result generated by the virtual wind measuring laser radar has a large deviation from the actual radar detection situation.

Description

Technical Field

[0001] The invention relates to a method for generating a virtual wind-measuring laser radar based on error analysis, and belongs to the technical field of radar simulation. Background Art

[0002] As an important means of atmospheric wind field detection, coherent Doppler wind lidar has been widely used in atmospheric science research, weather forecasting, and wind energy assessment due to its high precision and high sensitivity. This technology transmits a laser beam and receives the Doppler frequency shift signal scattered back by the atmosphere to invert the wind field information. In recent years, with the continuous advancement of lidar technology, the performance of coherent Doppler wind lidar has also been significantly improved, including improving measurement accuracy, expanding measurement range, and enhancing environmental adaptability. However, despite the rapid development of technology, due to the limitations of radar hardware conditions, especially the lack of range resolution, it still faces great challenges in detecting the fine structure of small-scale wind fields. When the range resolution of the radar is lower than the wind field scale, it will lead to an underestimation of wind shear, thus affecting the accuracy and reliability of the detection results.

[0003] Although coherent Doppler wind lidar has achieved remarkable results in wind field detection, the existing technology still has some obvious defects. First, the wind field measurement error assessment of lidar mainly depends on the radar signal-to-noise ratio and inversion algorithm, and it is difficult to fully consider the additional error introduced by the insufficient radar range resolution. The limitation of this evaluation method is that it fails to fully reflect the impact of radar hardware conditions on the measurement results, resulting in inaccurate error assessment. Secondly, although virtual wind lidar technology provides strong support for radar design and algorithm verification, the existing virtual wind lidar generation methods, such as direct averaging, frequency domain analysis, and time domain analysis, all have limitations to varying degrees. Although the direct averaging method is simple to implement and has fast calculation speed, it fails to fully consider the radar hardware parameters, the random characteristics of the atmospheric phase, and the radar signal acquisition and processing process, resulting in deviations between the virtual detection results and the real lidar detection results. Although the frequency domain analysis method and the time domain analysis method have high accuracy and reliability, the surge in their calculation volume greatly reduces the simulation speed, making it difficult to meet application scenarios with high real-time requirements. In addition, virtual wind lidar often ignores the impact of radar range resolution on wind field scale capture capability during simulation detection, which further increases the deviation between simulation results and actual conditions. Therefore, how to overcome the defects of existing technologies and improve the authenticity and reliability of virtual wind lidar technology has become an urgent problem to be solved. Summary of the invention

[0004] The technical problem to be solved by the present invention is: how to overcome the problem that the simulation results generated by the virtual wind measurement laser radar in the prior art have a large deviation from the actual situation.

[0005] In order to solve the above technical problems, the present invention is implemented by adopting the following technical solutions:

[0006] The present invention provides a method for generating a virtual wind laser radar based on error analysis, comprising:

[0007] An ideal atmospheric wind field data set is constructed based on the radial wind speed data of the atmospheric wind field;

[0008] According to the ideal atmospheric wind field data set, a virtual wind measurement laser radar is constructed based on different data processing methods, and a virtual wind measurement laser radar detection data set is generated by using the virtual wind measurement laser radar and the time for generating the virtual wind measurement laser radar detection data set is calculated;

[0009] Performing error analysis on the virtual wind laser radar detection data set according to the error evaluation index to determine the wind speed error distribution under different ideal wind speed wavelengths and different radar distance resolutions and the ratio of different ideal wind speed wavelengths and different radar distance resolutions;

[0010] Based on the ratios of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distributions under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and an error fitting function of a benchmark data processing method;

[0011] The error fitting function is used to describe the relationship between the error evaluation index and the ratio of the different ideal wind speed wavelengths and different radar distance resolutions;

[0012] Calculating actual deviations between the error fitting functions based on different data processing methods and the error fitting function of the reference data processing method;

[0013] Determine the computational efficiency and data reliability of the virtual wind laser radar according to the time of generating the virtual wind laser radar detection data set and the actual deviation value, determine the optimal data processing method for generating the virtual wind laser radar according to the computational efficiency and data reliability, and determine the final virtual wind laser radar according to the optimal data processing method of the virtual wind laser radar;

[0014] The shorter the time for generating the virtual wind laser radar detection data set is, the better the calculation efficiency of the virtual wind laser radar is; the smaller the actual deviation value is, the better the data reliability of the virtual wind laser radar is.

[0015] Furthermore, the method for acquiring radial wind speed data of the atmospheric wind field comprises:

[0016] According to the statistical law of atmospheric wind field and Fourier analysis method, the wind is simulated as a waveform composed of the superposition of sinusoidal wind speeds with different amplitudes and wavelengths;

[0017] According to the waveform, radial wind speed data of the atmospheric wind field is obtained based on an ideal sinusoidal wind field model.

[0018] Further, according to the ideal atmospheric wind field data set, a virtual wind laser radar is constructed based on different data processing methods, a virtual wind laser radar detection data set is generated, and the time for generating the virtual wind laser radar detection data set is calculated, including:

[0019] According to the ideal atmospheric wind field data set, setting the radar range resolution of the virtual wind measurement laser radar;

[0020] Dividing the maximum detection distance of the virtual wind laser radar into equally spaced distance gates according to the radar distance resolution, and obtaining the radial position of each distance gate of the virtual wind laser radar;

[0021] Different data processing methods are used to process the wind speed value detected by the virtual wind laser radar at the radial position of each range gate, generate a virtual wind laser radar detection data set and calculate the time for generating the virtual wind laser radar detection data set.

[0022] Furthermore, the data processing method includes a direct averaging method, wherein the method of using the direct averaging method to obtain a virtual wind laser radar detection data set generated based on the direct averaging method includes:

[0023] Based on the ideal atmospheric wind field dataset, the arithmetic mean of the ideal wind speed value in each range gate is calculated respectively, and a virtual wind measurement lidar detection dataset generated based on the direct averaging method is obtained.

[0024] Furthermore, the method for obtaining a virtual wind laser radar detection data set generated based on the direct averaging method by using the direct averaging method also includes:

[0025] Based on the ideal atmospheric wind field dataset, the wind speed value at each node position in the ideal atmospheric wind field is interpolated to each detection point position of the virtual wind measurement lidar to obtain the virtual wind measurement lidar detection dataset generated based on the direct averaging method.

[0026] Furthermore, the data processing method also includes an energy weighting method, wherein the method of obtaining a virtual wind laser radar detection data set generated based on the energy weighting method using the energy weighting method includes:

[0027] According to the relationship between the radar range resolution and the full width at half maximum of the transmitted laser pulse in the time domain, a laser radar pulse signal power simulation model at each radial position of the range gate is established to obtain a laser radar pulse signal power simulation result at each radial position of the range gate;

[0028] According to the laser radar pulse signal power simulation results at each range gate radial position, the wind speed value detected by the virtual wind measurement laser radar at each range gate radial position is calculated, and the virtual wind measurement laser radar detection data set generated based on the energy weighted method is obtained.

[0029] Furthermore, the data processing method also includes a time domain analysis method, wherein the method of using the time domain analysis method to obtain a virtual wind laser radar detection data set generated based on the time domain analysis method includes:

[0030] The maximum detection distance of the virtual wind laser radar is layered according to the minimum layer distance of the atmosphere to obtain the echo signal of each layer;

[0031] Each layer of echo signals is superimposed and spliced ​​in time sequence to obtain the time domain model of the echo signal within the radar detection range;

[0032] Fast Fourier transform (FFT) is performed on the heterodyne current within each range gate in the time domain model to obtain a power spectrum corresponding to each range gate.

[0033] According to the power spectrum corresponding to each range gate, the wind speed value detected by the virtual wind laser radar at the radial position of each range gate is calculated using the Doppler frequency shift formula, and a virtual wind laser radar detection data set generated based on the time domain analysis method is obtained;

[0034] Furthermore, before generating the virtual wind laser radar detection data set, it also includes interpolating the detected wind speed values ​​at each detection point of the virtual wind laser radar to each node position in the ideal atmospheric wind field for encryption.

[0035] Further, based on the ratios of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distributions under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and an error fitting function of a benchmark data processing method, including:

[0036] Fitting function between the RMS error generated by direct averaging and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as:

[0037] (1);

[0038] Where RMSE represents the wind speed error distribution under different ideal wind speed wavelengths and different radar range resolutions determined based on the root mean square error, γ represents the ratio of the ideal wind speed wavelength to the radar range resolution, Denotes the error dominant term For the fitting function The contribution coefficient of The number of error-dominant terms The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item The growth rate of , represents the ideal wind speed wavelength, Indicates radar range resolution;

[0039] Fitting function between the RMS error generated based on the energy-weighted method and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as:

[0040] (2);

[0041] In the formula, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item Growth rate;

[0042] Fitting function between the RMS error generated by the time domain analysis method and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as:

[0043] (3);

[0044] In the formula, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item growth rate.

[0045] Further, the calculation speed and data reliability of the virtual wind laser radar are determined according to the time of generating the virtual wind laser radar detection data set and the actual deviation value, the optimal data processing method for generating the virtual wind laser radar is determined according to the calculation speed and data reliability, and the final virtual wind laser radar is determined according to the optimal data processing method of the virtual wind laser radar, including:

[0046] The maximum deviation between the preset error fitting function and the error fitting function based on the time domain analysis method is , the ratio of the ideal wind speed wavelength to the radar range resolution is expressed as ,Will All elements contained are represented as a set ;

[0047] Using time domain analysis as the baseline data processing method:

[0048] According to the maximum deviation value , we obtain the set A of the ratio of the ideal wind speed wavelength to the radar range resolution within the error deviation range of the direct averaging method and the set B of the ratio of the ideal wind speed wavelength to the radar range resolution within the error deviation range of the energy weighted method;

[0049] Among them, set A and set B are expressed as:

[0050] (4);

[0051] (5);

[0052] In the formula, represents the actual deviation value between the error fitting function based on the direct averaging method and the error fitting function of the benchmark data processing method, represents an actual deviation value between the error fitting function based on the energy weighting method and the error fitting function of the reference data processing method;

[0053] when When , a virtual wind laser radar with a ratio of the ideal wind speed wavelength and the radar range resolution is generated using a direct averaging method;

[0054] when When the energy weighting method is used, a virtual wind measuring laser radar is generated at a ratio of the ideal wind speed wavelength and the radar distance resolution;

[0055] when When the time domain analysis method is used, a virtual wind measuring lidar is generated at a ratio of the ideal wind speed wavelength and the radar distance resolution.

[0056] Beneficial Effects

[0057] The method provided by the present invention obtains radial wind speed data of the atmospheric wind field and constructs an ideal atmospheric wind field data set, and then constructs a virtual wind measurement lidar based on different data processing methods to generate a virtual wind measurement lidar detection data set. The ideal atmospheric wind field data set is used in combination with different data processing methods to simulate the virtual wind measurement lidar detection wind field under different conditions, thereby generating a rich virtual wind measurement lidar detection data set, making subsequent error analysis and optimization more comprehensive.

[0058] The method provided by the present invention links the error evaluation index with the radar range resolution and the ideal wind speed wavelength, and obtains the error fitting function through data fitting, which can accurately describe the relationship between the error evaluation index and the ratio of the ideal wind speed wavelength to the radar range resolution.

[0059] The method provided by the present invention can preset the maximum deviation value of the error fitting function based on different data processing methods and the error fitting function of the benchmark data processing method, and based on the error fitting function and the calculation convenience and data reliability of the virtual wind measuring laser radar, generate a virtual wind measuring laser radar under different ideal wind speed wavelengths and different radar distance resolutions, thereby effectively improving the simulation accuracy and practicality of the virtual wind measuring laser radar, and providing more accurate and reliable virtual wind measuring laser radar detection data support for the scientific research of atmospheric wind fields, weather forecasting, wind power generation and other fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flowchart of a method for generating a virtual wind laser radar based on error analysis provided in Example 1 of the present invention;

[0061] Figure 2 It is a schematic diagram comparing the effects of a virtual wind laser radar detection wind field generated based on the direct averaging method, the energy weighting method, and the time domain analysis method with an ideal wind speed wavelength of 120 m and a radar distance resolution of 60 m provided in Example 2 of the present invention and an ideal atmospheric wind field;

[0062] Figure 3 It is a schematic diagram comparing the effects of a virtual wind laser radar detection wind field generated based on the direct averaging method, the energy weighting method, and the time domain analysis method with an ideal wind speed wavelength of 480m and a radar distance resolution of 30m provided in Example 2 of the present invention and an ideal atmospheric wind field;

[0063] Figure 4It is a schematic distribution diagram of the root mean square error RMSE, mean relative error MRE and mean absolute error MAE of the virtual wind measurement laser radar generated based on the energy weighting method provided in Example 2 of the present invention under different radar distance resolutions and ideal wind speed models of different wavelengths;

[0064] Figure 5 It is a schematic diagram of the fitting function curve between the root mean square error RMSE, the mean relative error MRE and the mean absolute error MAE generated based on the direct averaging method, the energy weighting method and the time domain analysis method provided in Example 2 of the present invention and the ratio of the ideal wind speed wavelength to the radar distance resolution. DETAILED DESCRIPTION

[0065] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0066] The term "and / or" is only a description of the association relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " generally indicates that the related objects are in an "or" relationship.

[0067] Example 1

[0068] like Figure 1 As shown, this embodiment introduces a method for generating a virtual wind laser radar based on error analysis, including:

[0069] Step 1: Construct an ideal atmospheric wind field dataset based on the acquired radial wind speed data of the atmospheric wind field.

[0070] By acquiring real atmospheric wind field data, the present invention can construct an ideal atmospheric wind field data set close to the actual situation, making the construction of the virtual wind measurement laser radar more accurate.

[0071] Step 2: According to the ideal atmospheric wind field dataset, a virtual wind measurement lidar is constructed based on different data processing methods, and the virtual wind measurement lidar is used to generate a virtual wind measurement lidar detection dataset and calculate the time for generating the virtual wind measurement lidar detection dataset.

[0072] The present invention utilizes an ideal atmospheric wind field dataset and combines different data processing methods such as direct averaging, energy weighting and time domain analysis to simulate virtual wind lidar detection wind fields under different conditions, generating a rich virtual wind lidar detection dataset, making subsequent error analysis and optimization more comprehensive.

[0073] The present invention also calculates the time for generating a virtual wind measurement laser radar detection data set. The shorter the calculation time for the virtual wind measurement laser radar to generate the virtual wind measurement laser radar detection data set, the better the calculation speed of the virtual wind measurement laser radar.

[0074] Step 3: Perform error analysis on the virtual wind lidar detection data set according to the error evaluation index to determine the wind speed error distribution under different ideal wind speed wavelengths and different radar distance resolutions and the ratio of different ideal wind speed wavelengths and different radar distance resolutions.

[0075] The present invention performs error analysis on a virtual wind measurement lidar detection data set, evaluates the ideal wind speed error under different conditions, clarifies the error distribution under different ideal wind speed wavelengths and different radar distance resolutions, and the ratio of the ideal wind speed wavelength to the radar distance resolution, and based on the wind speed error distribution under different ideal wind speed wavelengths and different radar distance resolutions, obtains the error fitting function through data fitting.

[0076] Step 4: Based on the ratio of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distribution under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and the error fitting function of the benchmark data processing method.

[0077] The error fitting function is used to describe the relationship between the error evaluation index and the ratio of the ideal wind speed wavelength to the radar range resolution.

[0078] The error fitting function can describe the relationship between the error evaluation index and the ratio of the ideal wind speed wavelength to the radar range resolution. Through the error fitting function, the error conditions under different radar parameters and different wind field scales can be predicted.

[0079] Step 5: Calculate the actual deviation value between the error fitting function based on different data processing methods and the error fitting function of the reference data processing method.

[0080] The present invention compares the actual deviation value of the error fitting function based on different data processing methods and the error fitting function of the benchmark data processing method with the preset maximum deviation value of the error fitting function based on different data processing methods and the error fitting function of the benchmark data processing method.

[0081] Step six: Determine the calculation speed and data reliability of the virtual wind measurement lidar based on the time of generating the virtual wind measurement lidar detection data set and the actual deviation value, determine the optimal data processing method for generating the virtual wind measurement lidar based on the calculation speed and data reliability, and determine the final virtual wind measurement lidar based on the optimal data processing method of the virtual wind measurement lidar.

[0082] Among them, the shorter the time for generating the virtual wind measurement laser radar detection data set, the better the calculation speed of the virtual wind measurement laser radar; the smaller the actual deviation value, the better the data reliability of the virtual wind measurement laser radar.

[0083] The present invention utilizes the time for generating a virtual wind measuring laser radar detection data set and the actual deviation value between the error fitting function based on different data processing methods and the error fitting function of the benchmark data processing method, combined with the calculation quickness and data reliability requirements of the virtual wind measuring laser radar, to generate a virtual wind measuring laser radar that meets different conditions. The generated virtual wind measuring laser radar not only has high precision and reliability, but can also adapt to different radar parameters and wind speed conditions, providing more flexible and diverse options for practical applications, and providing strong technical support for the fields of scientific research on atmospheric wind fields, weather forecasting, wind power generation, etc.

[0084] Example 2

[0085] Based on the same inventive concept as Example 1, this example introduces a method for generating a virtual wind laser radar based on error analysis, including:

[0086] Step 1: Obtain radial wind speed data of the atmospheric wind field and construct an ideal atmospheric wind field dataset.

[0087] The method for acquiring radial wind speed data of the atmospheric wind field comprises:

[0088] According to the statistical law of atmospheric wind field and Fourier analysis method, the wind is simulated as a waveform composed of the superposition of sinusoidal wind speeds with different amplitudes and wavelengths;

[0089] According to the waveform, radial wind speed data of the atmospheric wind field is obtained based on an ideal sinusoidal wind field model.

[0090] The ideal sinusoidal wind field model expression is:

[0091] (1);

[0092] in, Represents the atmospheric wind field node The radial position The ideal wind speed value at Indicates the distance step between two atmospheric wind field nodes, representing the minimum atmospheric stratification distance. Indicates the maximum detection distance, represents the wind speed amplitude, Indicates the ideal wind speed wavelength, reflecting the size of the wind field. Indicates the maximum detection distance The distance step between two nodes is The total number of nodes, sin represents the sine function.

[0093] Step 2: According to the ideal atmospheric wind field dataset, a virtual wind laser radar is constructed based on different data processing methods, a virtual wind laser radar detection dataset is generated, and the time for generating the virtual wind laser radar detection dataset is calculated, including:

[0094] According to the ideal atmospheric wind field data set, the radar range resolution of the virtual wind laser radar is set , according to the radar range resolution The maximum detection distance , divided into equally spaced gates, and the The radial position of the virtual wind lidar range gate , its expression is:

[0095] (2);

[0096] In the formula, represents the total number of equally spaced gates, Represents the rounding function.

[0097] Dividing the maximum detection distance of the virtual wind laser radar into equally spaced distance gates according to the radar distance resolution, and obtaining the radial position of each distance gate of the virtual wind laser radar;

[0098] Different data processing methods are used to process the wind speed value detected by the virtual wind laser radar at the radial position of each range gate, generate a virtual wind laser radar detection data set and calculate the time for generating the virtual wind laser radar detection data set.

[0099] The data processing method includes a direct averaging method, wherein the method of using the direct averaging method to obtain a virtual wind laser radar detection data set generated based on the direct averaging method includes:

[0100] Based on the ideal atmospheric wind field dataset, the arithmetic mean of the ideal wind speed value within each range gate is calculated to obtain a virtual wind lidar detection dataset generated based on the direct averaging method. , where the virtual wind lidar detection dataset generated based on the direct averaging method A virtual wind lidar detection data It is expressed as:

[0101] (3);

[0102] in, Indicates The starting point of the range gate is used to process the nodes of the atmospheric wind field. Indicates The endpoint of each range gate processes the node of the atmospheric wind field.

[0103] The method for obtaining a virtual wind laser radar detection data set generated based on the direct averaging method by using the direct averaging method also includes:

[0104] Based on the ideal atmospheric wind field dataset, the wind speed value at each node position in the ideal atmospheric wind field is interpolated to each detection point position of the virtual wind measurement lidar to obtain the virtual wind measurement lidar detection dataset generated based on the direct averaging method.

[0105] The data processing method further includes an energy weighting method, wherein the method of using the energy weighting method to obtain a virtual wind laser radar detection data set generated based on the energy weighting method includes:

[0106] According to the radar range resolution and the full width at half maximum of the emitted laser pulse The relationship between the laser radar pulse signal power at each range gate radial position is established. , and obtain the simulation results of the laser radar pulse signal power at each range gate radial position, where the laser pulse waveform is Gaussian.

[0107] Among them, the radar range resolution and the full width at half maximum of the emitted laser pulse The relationship is expressed as:

[0108] (4);

[0109] In the formula, Represents the speed of light.

[0110] Among them, the laser radar pulse signal power simulation model at each range gate radial position , expressed as:

[0111] (5);

[0112] In the formula, represents the logarithmic function with base 2, (·) indicates An exponential function with base .

[0113] According to the laser radar pulse signal power simulation results at each range gate radial position, the wind speed value detected by the virtual wind laser radar at each range gate radial position is calculated, and the virtual wind laser radar detection data set generated based on the energy weighted method is obtained. , where the virtual wind lidar detection dataset generated based on the energy weighted method A virtual wind lidar detection data It is expressed as:

[0114] (6);

[0115] In the formula, express The ideal atmospheric wind field node at express The ideal atmospheric wind field node at .

[0116] The data processing method further includes a time domain analysis method, wherein the method of using the time domain analysis method to obtain a virtual wind laser radar detection data set generated based on the time domain analysis method includes:

[0117] The maximum detection distance of the virtual wind laser radar is layered according to the minimum layer distance of the atmosphere to obtain the echo signal of each layer;

[0118] Each layer of echo signals is superimposed and spliced ​​in time sequence to obtain the time domain model of the echo signal within the radar detection range;

[0119] Performing fast Fourier transform (FFT) on the heterodyne current within each range gate in the time domain model to obtain a power spectrum corresponding to each range gate;

[0120] Wherein, the calculation expression of the heterodyne current is:

[0121] (7);

[0122] In the formula, Indicates the current time within the range of the gate The heterodyne current, represents the real part of the complex form of the heterodyne current, represents the AOM frequency shift frequency, M represents the acousto-optic modulator, Represents the atmospheric wind field node The collection time, Indicates the maximum detection distance The distance step between two atmospheric wind field nodes is The total number of nodes, Represents the atmospheric wind field node The random factor corresponding to the speckle effect of the echo signal, represents the local oscillator frequency, LO represents the local oscillator frequency, represents the laser pulse power function, where Represents the atmospheric wind field node The radial position of represents the speed of light, Represents the atmospheric wind field node The radial position The square of the transmittance at For the system The total optical efficiency, Represents the atmospheric wind field node The radial position The backscattering coefficient at Represents the atmospheric wind field node The radial position The transmittance at Indicates telescope The radiation area, represents the exponential function, Indicates the wave number of the laser emitted by the virtual wind laser radar, Represents the atmospheric wind field node The radial position The ideal wind speed value at the location.

[0123] According to the power spectrum corresponding to each range gate, the wind speed value detected by the virtual wind lidar at the radial position of each range gate is calculated using the Doppler frequency shift formula, and the virtual wind lidar detection data set generated based on the time domain analysis method is obtained. ; Among them, the wind speed value detected by the virtual wind laser radar at the radial position is expressed as:

[0124] (8);

[0125] In the formula, Indicates the wind speed value detected by the virtual wind laser radar at the radial position, Indicates The Doppler shift at the radial position of the range gate is Indicates the wavelength of the signal emitted by the virtual wind lidar.

[0126] Before generating the virtual wind laser radar detection data set, the method also includes interpolating the detected wind speed values ​​at each detection point of the virtual wind laser radar to each node position in the ideal atmospheric wind field for encryption.

[0127] Step 3: Perform error analysis on the virtual wind lidar detection data set according to the error evaluation index to determine the wind speed error distribution under different ideal wind speed wavelengths and different radar distance resolutions and the ratio of different ideal wind speed wavelengths and different radar distance resolutions.

[0128] This embodiment uses error evaluation indicators including root mean square error RMSE, mean relative error MRE and mean absolute error MAE to perform error analysis on the virtual wind laser radar detection data set. Figure 4 This is a schematic distribution diagram of the root mean square error RMSE, mean relative error MRE and mean absolute error MAE of the virtual wind measurement lidar generated based on the energy weighted method provided in this embodiment under different radar distance resolutions and ideal wind speed models at different wavelengths.

[0129] Step 4: Based on the ratios of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distributions under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and an error fitting function of a benchmark data processing method.

[0130] In this embodiment, the ideal wind speed wavelength The value range is ,in, Indicates the total number of wavelengths with a wavelength step length of 60m and a range from 60m to 480m. Indicates wavelength serial number; radar range resolution The value range is ,in, Indicates the total number of radar range resolutions ranging from 3m to 60m with a step size of 3m. Indicates the radar range resolution number.

[0131] Figure 2 It is a schematic diagram comparing the effects of a virtual wind measurement lidar detection wind field generated based on the direct averaging method, energy weighting method and time domain analysis method with an ideal wind speed wavelength of 120m and a radar distance resolution of 60m provided by an embodiment of the present invention and an ideal atmospheric wind field.

[0132] Figure 3 It is a schematic diagram comparing the effects of a virtual wind measurement lidar wind field detected by the ideal wind speed wavelength of 480m and a radar distance resolution of 30m provided by an embodiment of the present invention and an ideal atmospheric wind field generated based on the direct averaging method, the energy weighting method and the time domain analysis method.

[0133] In this embodiment, the root mean square error (RMSE) is used as the error evaluation index, the ratio of the ideal wind speed wavelength and the radar range resolution is defined, and the error fitting function is obtained through data fitting based on the wind speed error distribution under different ideal wind speed wavelengths and different radar range resolutions.

[0134] In this embodiment, a data processing method is selected from different data processing methods as a reference data processing method.

[0135] Based on the ratios of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distributions under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and an error fitting function of a benchmark data processing method, including:

[0136] Fitting function between the RMS error generated by direct averaging and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as:

[0137] (9);

[0138] Where RMSE represents the wind speed error distribution under different ideal wind speed wavelengths and different radar range resolutions determined based on the root mean square error, γ represents the ratio of the ideal wind speed wavelength to the radar range resolution, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item The growth rate of , represents the ideal wind speed wavelength, Indicates radar range resolution;

[0139] Fitting function between the RMS error generated based on the energy-weighted method and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as:

[0140] (10);

[0141] In the formula, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item Growth rate;

[0142] Fitting function between the RMS error generated by the time domain analysis method and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as:

[0143] (11);

[0144] In the formula, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item growth rate.

[0145] A schematic diagram of a fitting function curve between the root mean square error RMSE, the mean relative error MRE and the mean absolute error MAE generated based on the direct averaging method, the energy weighting method and the time domain analysis method provided in an embodiment of the present invention and the ratio of the ideal wind speed wavelength to the radar range resolution is shown in FIG. Figure 5 shown.

[0146] Step 5: determining the computational efficiency and data reliability of the virtual wind laser radar according to the time of generating the virtual wind laser radar detection data set and the actual deviation value, determining the optimal data processing method for generating the virtual wind laser radar according to the computational efficiency and data reliability, and determining the final virtual wind laser radar according to the optimal data processing method of the virtual wind laser radar, including:

[0147] The actual deviation value between the error fitting function based on the different data processing methods and the error fitting function of the reference data processing method is calculated.

[0148] The maximum deviation between the preset error fitting function and the error fitting function based on the time domain analysis method is , the ratio of the ideal wind speed wavelength to the radar range resolution is expressed as ,Will All elements contained are represented as a set .

[0149] Using time domain analysis as the baseline data processing method:

[0150] According to the maximum deviation value , we obtain the set A of the ratio of the ideal wind speed wavelength to the radar range resolution within the error deviation range of the direct averaging method and the set B of the ratio of the ideal wind speed wavelength to the radar range resolution within the error deviation range of the energy weighted method;

[0151] Among them, set A and set B are expressed as:

[0152] (4);

[0153] (5);

[0154] In the formula, represents the actual deviation value between the error fitting function based on the direct averaging method and the error fitting function of the benchmark data processing method, represents an actual deviation value between the error fitting function based on the energy weighting method and the error fitting function of the reference data processing method;

[0155] when When , a virtual wind laser radar with a ratio of the ideal wind speed wavelength and the radar range resolution is generated using a direct averaging method;

[0156] when When the energy weighting method is used, a virtual wind measuring laser radar is generated at a ratio of the ideal wind speed wavelength and the radar distance resolution;

[0157] when When the time domain analysis method is used, a virtual wind measuring lidar is generated at a ratio of the ideal wind speed wavelength and the radar distance resolution.

[0158] The specific functions and effects of the above steps are referred to the relevant contents of the method in Example 1 and will not be elaborated here.

[0159] In summary, the method provided by the present invention obtains the radial wind speed data of the atmospheric wind field and constructs an ideal atmospheric wind field data set, and then constructs a virtual wind measurement lidar based on different data processing methods to generate a virtual wind measurement lidar detection data set. By using the ideal atmospheric wind field data set and combining different data processing methods, the virtual wind measurement lidar detection wind field under different conditions is simulated, and a rich virtual wind measurement lidar detection data set is generated, so that the subsequent error analysis and optimization are more comprehensive.

[0160] The method provided by the present invention links the error evaluation index with the radar range resolution and the ideal wind speed wavelength by defining a proportional relationship, and obtains an error fitting function through data fitting, which can accurately describe the relationship between the error evaluation index and the ratio of the ideal wind speed wavelength to the radar range resolution.

[0161] The method provided by the present invention can preset the maximum deviation value of the error fitting function based on different data processing methods and the error fitting function of the benchmark data processing method, and based on the error fitting function and the calculation convenience and data reliability of the virtual wind measuring laser radar, generate a virtual wind measuring laser radar under different ideal wind speed wavelengths and different radar distance resolutions, thereby effectively improving the simulation accuracy and practicality of the virtual wind measuring laser radar, and providing more accurate and reliable virtual wind measuring laser radar detection data support for the scientific research of atmospheric wind fields, weather forecasting, wind power generation and other fields.

[0162] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0163] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0164] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0165] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0166] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the enlightenment of the present invention, ordinary technicians in this field can also make many forms without departing from the scope of protection of the purpose of the present invention and the claims, which all fall within the protection of the present invention.

Claims

1. A method for generating a virtual wind laser radar based on error analysis, characterized in that: include: An ideal atmospheric wind field data set is constructed based on the radial wind speed data of the atmospheric wind field; According to the ideal atmospheric wind field data set, a virtual wind measurement laser radar is constructed based on different data processing methods, and a virtual wind measurement laser radar detection data set is generated by using the virtual wind measurement laser radar and the time for generating the virtual wind measurement laser radar detection data set is calculated; Performing error analysis on the virtual wind laser radar detection data set according to the error evaluation index to determine the wind speed error distribution under different ideal wind speed wavelengths and different radar distance resolutions and the ratio of different ideal wind speed wavelengths and different radar distance resolutions; Based on the ratios of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distributions under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and an error fitting function of a benchmark data processing method; The error fitting function is used to describe the relationship between the error evaluation index and the ratio of the different ideal wind speed wavelengths and different radar distance resolutions; Calculating actual deviations between the error fitting functions based on different data processing methods and the error fitting function of the reference data processing method; The calculation speed and data reliability of the virtual wind measuring lidar are determined according to the time of generating the virtual wind measuring lidar detection data set and the actual deviation value, the optimal data processing method for generating the virtual wind measuring lidar is determined according to the calculation speed and data reliability, and the final virtual wind measuring lidar is determined according to the optimal data processing method of the virtual wind measuring lidar.

2. The method for generating a virtual wind laser radar based on error analysis according to claim 1, characterized in that: The method for acquiring radial wind speed data of the atmospheric wind field comprises: According to the statistical law of atmospheric wind field and Fourier analysis method, the wind is simulated as a waveform composed of the superposition of sinusoidal wind speeds with different amplitudes and wavelengths; According to the waveform, radial wind speed data of the atmospheric wind field is obtained based on an ideal sinusoidal wind field model.

3. The method for generating a virtual wind laser radar based on error analysis according to claim 1, characterized in that: According to the ideal atmospheric wind field data set, a virtual wind laser radar is constructed based on different data processing methods, a virtual wind laser radar detection data set is generated, and the time for generating the virtual wind laser radar detection data set is calculated, including: According to the ideal atmospheric wind field data set, setting the radar range resolution of the virtual wind measurement laser radar; Dividing the maximum detection distance of the virtual wind laser radar into equally spaced distance gates according to the radar distance resolution, and obtaining the radial position of each distance gate of the virtual wind laser radar; Different data processing methods are used to process the wind speed value detected by the virtual wind laser radar at the radial position of each range gate, generate a virtual wind laser radar detection data set and calculate the time for generating the virtual wind laser radar detection data set.

4. The method for generating a virtual wind laser radar based on error analysis according to claim 3, characterized in that: The data processing method includes a direct averaging method, wherein the method of using the direct averaging method to obtain a virtual wind laser radar detection data set generated based on the direct averaging method includes: Based on the ideal atmospheric wind field dataset, the arithmetic mean of the ideal wind speed value in each range gate is calculated respectively, and a virtual wind measurement lidar detection dataset generated based on the direct averaging method is obtained.

5. The method for generating a virtual wind laser radar based on error analysis according to claim 4, characterized in that: The method for obtaining a virtual wind laser radar detection data set generated based on the direct averaging method by using the direct averaging method also includes: Based on the ideal atmospheric wind field dataset, the wind speed value at each node position in the ideal atmospheric wind field is interpolated to each detection point position of the virtual wind measurement lidar to obtain the virtual wind measurement lidar detection dataset generated based on the direct averaging method.

6. The method for generating a virtual wind laser radar based on error analysis according to claim 4 or 5, characterized in that: The data processing method further includes an energy weighting method, wherein the method of using the energy weighting method to obtain a virtual wind laser radar detection data set generated based on the energy weighting method includes: According to the relationship between the radar range resolution and the full width at half maximum of the transmitted laser pulse in the time domain, a laser radar pulse signal power simulation model at each radial position of the range gate is established to obtain a laser radar pulse signal power simulation result at each radial position of the range gate; According to the laser radar pulse signal power simulation results at each range gate radial position, the wind speed value detected by the virtual wind measurement laser radar at each range gate radial position is calculated, and a virtual wind measurement laser radar detection data set generated based on the energy weighted method is obtained.

7. The method for generating a virtual wind laser radar based on error analysis according to claim 6, characterized in that: The data processing method further includes a time domain analysis method, wherein the method of using the time domain analysis method to obtain a virtual wind laser radar detection data set generated based on the time domain analysis method includes: The maximum detection distance of the virtual wind laser radar is layered according to the minimum layer distance of the atmosphere to obtain the echo signal of each layer; Each layer of echo signals is superimposed and spliced ​​in time sequence to obtain the time domain model of the echo signal within the radar detection range; Performing fast Fourier transform (FFT) on the heterodyne current within each range gate in the time domain model to obtain a power spectrum corresponding to each range gate; According to the power spectrum corresponding to each range gate, the wind speed value detected by the virtual wind lidar at the radial position of each range gate is calculated using the Doppler frequency shift formula, and a virtual wind lidar detection data set generated based on the time domain analysis method is obtained.

8. The method for generating a virtual wind laser radar based on error analysis according to claim 1, characterized in that: Before generating the virtual wind laser radar detection data set, the method also includes interpolating the detected wind speed values ​​at each detection point of the virtual wind laser radar to each node position in the ideal atmospheric wind field for encryption.

9. The method for generating a virtual wind laser radar based on error analysis according to claim 1, characterized in that: Based on the ratios of the different ideal wind speed wavelengths and the different radar distance resolutions and the wind speed error distributions under the different ideal wind speed wavelengths and the different radar distance resolutions, data fitting is performed to obtain error fitting functions based on different data processing methods and an error fitting function of a benchmark data processing method, including: Fitting function between the RMS error generated by direct averaging and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as: （1）； Where RMSE represents the wind speed error distribution under different ideal wind speed wavelengths and different radar range resolutions determined based on the root mean square error, γ represents the ratio of the ideal wind speed wavelength to the radar range resolution, Denotes the error dominant term For the fitting function The contribution coefficient of The number of error-dominant terms The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item The growth rate of , represents the ideal wind speed wavelength, Indicates radar range resolution; Fitting function between the RMS error generated based on the energy-weighted method and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as: （2）； In the formula, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item Growth rate; Fitting function between the RMS error generated by the time domain analysis method and the ratio of the ideal wind speed wavelength to the radar range resolution , expressed as: （3）； In the formula, Denotes the error dominant term For the fitting function The contribution coefficient of Denotes the error dominant term The growth rate, Indicates error adjustment item For the fitting function The contribution coefficient of Indicates error adjustment item growth rate.

10. The method for generating a virtual wind laser radar based on error analysis according to claim 9, characterized in that: Determining the computational efficiency and data reliability of the virtual wind laser radar according to the time of generating the virtual wind laser radar detection data set and the actual deviation value, determining the optimal data processing method for generating the virtual wind laser radar according to the computational efficiency and data reliability, and determining the final virtual wind laser radar according to the optimal data processing method of the virtual wind laser radar, including: The maximum deviation between the preset error fitting function and the error fitting function based on the time domain analysis method is , the ratio of the ideal wind speed wavelength to the radar range resolution is expressed as ,Will All elements contained are represented as a set ; Using time domain analysis as the baseline data processing method: According to the maximum deviation value , we obtain the set A of the ratio of the ideal wind speed wavelength to the radar range resolution within the error deviation range of the direct averaging method and the set B of the ratio of the ideal wind speed wavelength to the radar range resolution within the error deviation range of the energy weighted method; Among them, set A and set B are expressed as: （4）； （5）； In the formula, represents the actual deviation value between the error fitting function based on the direct averaging method and the error fitting function of the benchmark data processing method, represents the actual deviation value between the error fitting function based on the energy weighting method and the error fitting function of the reference data processing method; when When , a virtual wind laser radar with a ratio of the ideal wind speed wavelength and the radar range resolution is generated using a direct averaging method; when When the energy weighting method is used, a virtual wind measuring laser radar is generated at a ratio of the ideal wind speed wavelength and the radar distance resolution; when When the time domain analysis method is used, a virtual wind measuring lidar is generated at a ratio of the ideal wind speed wavelength and the radar distance resolution.

Images