Wind profile radar signal and data processing method and system

Abstract

The invention relates to a wind profile radar signal and data processing method and system, and the method comprises the steps: constructing a working parameter combination of a wind profile radar, and designing a radar detection mode of the wind profile radar based on the working parameter combination; the wind profile radar collects external wind field data based on the radar detection mode to obtain a wind profile radar signal; performing signal processing based on the obtained wind profile radar signal; performing pulse compression, moving target detection and Doppler spectrum analysis on the wind profile radar signal; extracting wind profile radar data from the processed wind profile radar signal, and processing the wind profile radar data; performing wind field inversion and quality control on the wind profile radar data; and constructing a three-dimensional wind field based on the processed wind profile radar data. According to the scheme, complete acquisition of weak meteorological echo signals under a clear sky condition is effectively ensured, and the accuracy of wind field inversion is improved.

Description

Technical Field

[0001] This invention relates to the field of meteorology, and more particularly to a method and system for processing wind profiler radar signals and data. Background Technology

[0002] In today's era, with the sustained and steady development of my country's national economy, various sectors are increasingly reliant on meteorological support, and the requirements for its accuracy, timeliness, and comprehensiveness are becoming increasingly stringent. Meteorological information is as fundamental as a cornerstone for many industries. Whether it is the hydrological and water conservancy industry, which is closely related to people's livelihoods, or atmospheric environmental monitoring and weather forecasting, which are related to people's daily lives, all rely on accurate meteorological data as support.

[0003] Wind profiler radar, with its unique functional characteristics, stands out among numerous meteorological detection devices, becoming a powerful assistant in meteorological support work. It can operate 24 hours a day without interruption, continuously and stably providing relevant departments and research institutions with a series of crucial meteorological elements, such as dynamically changing horizontal wind fields, vertical airflow, and atmospheric refractive index structure constants. This feature makes it irreplaceable in the field of meteorological detection.

[0004] In the aerospace field, the meteorological data provided by wind profiler radar offers crucial reference information for aircraft takeoffs and landings, flight path planning, and spacecraft launches and orbital operations. By monitoring atmospheric wind fields and other information in real time, aerospace departments can anticipate potential risks from weather changes and take corresponding countermeasures to ensure flight safety and the smooth execution of missions.

[0005] In the field of hydrology and water conservancy, accurate meteorological data is of great significance for river level prediction, reservoir operation, and flood disaster early warning. Meteorological data provided by wind profiler radar helps relevant departments to prepare in advance, rationally allocate water resources, ensure the safe operation of water conservancy facilities, and minimize losses caused by floods and other disasters.

[0006] In the field of atmospheric monitoring, as people pay increasing attention to environmental protection and air quality, data such as atmospheric refractive index structure constant monitored by wind profiler radar can help researchers gain a deeper understanding of the physical properties and changing patterns of the atmosphere, providing strong data support for research on atmospheric pollution diffusion, climate change, and other related issues.

[0007] For weather forecasting, wind profiler radar data is indispensable. The high-precision, real-time meteorological data it provides greatly enriches the data sources for weather forecasts, enabling forecast models to more accurately simulate atmospheric motion. This improves the accuracy and timeliness of weather forecasts, providing more reliable meteorological guidance for people's daily travel and production activities.

[0008] However, existing wind profiler radars have significant problems under clear-sky conditions. Clear-sky echo signals are extremely weak, greatly increasing the difficulty of detection. Under clear-sky conditions, there are relatively few scatterers in the atmosphere capable of generating radar echoes, resulting in low echo power received by the radar. These weak clear-sky echo signals are highly susceptible to interference from various types of noise during transmission, such as thermal noise generated by electronic equipment and electromagnetic noise from the surrounding environment. This noise can mask the characteristics of the echo signal, making it difficult to accurately extract effective meteorological information during signal processing.

[0009] Furthermore, traditional signal processing algorithms perform poorly when dealing with such weak clear-sky echoes. Because the algorithm design does not fully consider the special properties of clear-sky echoes, misjudgments or omissions are prone to occur during detection and analysis. For example, when the intensity of a clear-sky echo signal is close to that of noise, the algorithm may misjudge the echo signal as noise and ignore it, leading to the loss of important meteorological information. Moreover, existing hardware also has limitations in receiving and processing weak signals. The hardware's sensitivity and dynamic range are limited, making it unable to effectively amplify and accurately process extremely weak clear-sky echo signals, further restricting the detection capability of wind profiler radar in clear-sky conditions. These problems severely affect the ability of wind profiler radar to acquire accurate meteorological data under clear-sky conditions, thus adversely impacting related fields that rely on this data. Summary of the Invention

[0010] The technical problem to be solved by the present invention is to provide a method and system for processing wind profiler radar signals and data.

[0011] To achieve the above-mentioned objectives, the present invention provides a method for processing wind profiler radar signals and data, comprising: S1. Construct the working parameter combination of the wind profiler radar, and design the radar detection mode of the wind profiler radar based on the working parameter combination; S2. The wind profiler radar collects external wind field data based on the radar detection mode to obtain a wind profiler radar signal; S3. Perform signal processing based on the obtained wind profiler radar signal; wherein, pulse compression, moving target detection and Doppler spectrum analysis are performed on the wind profiler radar signal; S4. Extract wind profiler radar data from the processed wind profiler radar signal and process the wind profiler radar data: including wind field inversion and quality control of the wind profiler radar data. S5. Construct a three-dimensional wind field based on the processed wind profile radar data.

[0012] According to one aspect of the present invention, in step S1, the step of constructing the working parameter combination of the wind profiler radar includes: height resolution, maximum unambiguous Doppler velocity, velocity resolution, and time resolution. The height resolution can be set in multiple ways based on the detection height; The maximum unambiguous Doppler velocity can be set in multiple ways based on the detection height; The temporal resolution is obtained based on the type of height resolution and the number of probe beams.

[0013] According to one aspect of the invention, the height resolution is provided with three levels: low, medium, and high, which are 75m, 150m, and 300m, respectively. There are two settings for the maximum unambiguous Doppler velocity, namely: At detection altitudes below 5 km, the maximum unambiguous Doppler velocity is set to 15 m / s, and at detection altitudes above 5 km, the maximum unambiguous Doppler velocity is set to 20 m / s. The velocity resolution is less than 0.2 m / s; The time resolution is expressed as:

[0014] in, Indicates the beam number. The category number indicating the height resolution. For time-domain accumulation, For the number of points in the Fast Fourier Transform (FFT), The spectral mean The number of detection modes used. The number of detection beams used.

[0015] According to one aspect of the present invention, in step S1, the step of designing the radar detection mode of the wind profiler radar based on the combination of operating parameters involves determining the detection mode parameters in the radar detection mode based on the combination of operating parameters to complete the design of the radar detection mode. The detection mode parameters include: pulse width, pulse repetition period, pulse coherence averaging count, Fast Fourier Transform (FFT) points, and spectral average. The steps then include: The pulse width is determined based on the height resolution requirements, and the pulse repetition period is determined based on the detection height that matches the height resolution. The pulse coherence average number of times is determined based on the formula for the maximum unambiguous Doppler velocity and the determined maximum unambiguous Doppler velocity and pulse repetition period; The number of points for the Fast Fourier Transform (FFT) is determined based on the maximum unambiguous Doppler velocity and velocity resolution. With atmospheric stabilization time as a constraint, the dwell time range of a single beam is set. Under the premise of satisfying the dwell time range of a single beam, the average number of spectra is set to the maximum integer value that meets the signal-to-noise ratio improvement requirements.

[0016] According to one aspect of the present invention, in step S2, the wind profiler radar acquires external wind field data based on the radar detection mode to obtain a wind profiler radar signal, wherein the wind profiler radar acquires external wind field data using a Doppler transceiver system to obtain a wind profiler radar signal. The wind profiler radar uses 3 or 5 beams for detection and data acquisition.

[0017] According to one aspect of the present invention, in step S3, signal processing is performed based on the obtained wind profiler radar signal; wherein the steps of pulse compression, moving target detection, and Doppler spectrum analysis of the wind profiler radar signal include: Acquire orthogonal I and Q video signals output by the wind profiler radar; A / D sampling and saturation detection are performed on the quadrature I and Q video signals. For the intermediate frequency direct sampling signal, saturation detection is performed first, and the receiving channel is adjusted through the AGC control loop to ensure that the input data is not saturated before sampling. Time-domain filtering involves designing a linear phase filter to suppress secondary range folding echo interference and rain echo interference, thereby improving the signal-to-noise ratio and reducing the computational load of FFT. Time-domain coherent averaging, adding echoes in phase according to the phase relationship of adjacent periods, improves echo signal-to-noise ratio and target detection capability; Spectral analysis involves windowing the coherently averaged signal in the time domain, combining them into a complex signal, and then performing FFT analysis. Frequency domain filtering employs filters with good rectangular coefficients and small notch widths to suppress ground clutter while preserving meteorological echoes. Spectral averaging involves averaging the values ​​of P power spectral density functions at corresponding frequencies to further improve the signal-to-noise ratio.

[0018] According to one aspect of the present invention, in step S4, wind profiler radar data is extracted from the processed wind profiler radar signal, and the wind profiler radar data is processed: wherein the step of performing wind field inversion and quality control on the wind profiler radar data includes: S41. Target Detection and Spectral Moment Calculation: Identify the echo signals of meteorological targets using a power spectrum peak detection algorithm, and calculate the noise level, Doppler velocity, spectral width, and signal-to-noise ratio based on the power spectrum peak; S42. Consistent averaging: Time-domain averaging of multiple Doppler velocity measurements within the same beam direction and the same distance gate is performed to suppress random measurement errors; S43. Wind vector inversion: Based on Doppler velocity measurements with different beam orientations, the horizontal wind vector components are calculated using the least squares method; S44. Data quality control: The reliability of labeled data is verified through wind shear checks, continuity checks, and two-dimensional median checks; S45. Anti-interference processing: The power spectrum is processed using half-plane cancellation, data smoothing, pattern recognition, multi-peak extraction, median filtering, and cluster analysis.

[0019] According to one aspect of the present invention, step S44, the data quality control step, includes: S441. Spatial consistency test based on wind field shear threshold; S442. Outlier removal based on time continuity constraints; S443. A two-dimensional median filter is used to smooth the wind field data.

[0020] According to one aspect of the present invention, step S45, the anti-interference processing step, includes: S451. Suppress ground clutter using a half-plane cancellation algorithm; S452. Spectral smoothing is performed using a Savitzky-Golay filter; S453. Identifying multi-peak spectral features based on K-means clustering to separate overlapping meteorological targets; S454. Combine morphological median filtering to eliminate impulse interference noise.

[0021] To achieve the above-mentioned objectives, the present invention provides a wind profiler radar signal and data processing system, comprising: The radar detection mode setting module is used to construct the working parameter combination of the wind profiler radar and design the radar detection mode of the wind profiler radar based on the working parameter combination. The signal receiving module is used to receive the wind profiler radar signal output by the wind profiler radar during the process of collecting external wind field data based on the radar detection mode. The signal processing module is used to process the wind profiler radar signal; specifically, it performs pulse compression, moving target detection, and Doppler spectrum analysis on the wind profiler radar signal. The data processing module is used to extract wind profiler radar data from the processed wind profiler radar signal and process the wind profiler radar data, including wind field inversion and quality control of the wind profiler radar data. The display module is used to construct and display the three-dimensional wind field from the processed wind profile radar data.

[0022] According to one aspect of the present invention, this solution employs multiple mode selection settings for height resolution, effectively resolving the conflict between detection height and range resolution, and ensuring optimal overall radar performance. In particular, by combining short and long pulses and employing pulse compression technology, the height resolution is balanced with the detection height, ensuring overlap in detection height between modes. This effectively achieves comprehensive coverage of the detection range under clear sky conditions, significantly enhancing the detection effectiveness of this solution.

[0023] According to one aspect of the present invention, this approach significantly enhances the detection capability under clear sky conditions and weak wind fields (such as boundary layer light wind circulation) by setting a specific velocity resolution, and effectively reduces wind speed errors.

[0024] According to one aspect of the present invention, the maximum unambiguous Doppler velocity is set at different detection heights, ensuring that velocity ambiguity does not occur in most cases. Even in rare cases of velocity ambiguity, it is only a single instance and can be identified during data processing. Based on this, high resolution in the near-field (e.g., 0-5 km) is guaranteed, while sufficient detection sensitivity in the far-field (e.g., 5-20 km) is achieved. Furthermore, the high overlap design between modes eliminates the vertical detection blind zone caused by traditional single-pulse modes, thus improving the continuity of the wind field's vertical profile.

[0025] According to one aspect of the present invention, by directly digitizing the wind profiler radar signal, the noise accumulation and distortion problems in the traditional analog signal processing link are avoided, ensuring the complete acquisition of weak meteorological echo signals under clear sky conditions, thereby improving the accuracy of wind field inversion.

[0026] According to one aspect of the present invention, by employing algorithms such as digital filtering and adaptive clutter suppression, the pollution of the power spectrum by noise such as ground clutter and radio frequency interference is effectively eliminated, ensuring data reliability under complex meteorological conditions.

[0027] According to one aspect of the present invention, this approach can flexibly adjust the pulse width and sampling strategy, taking into account both long-range detection and high-resolution requirements, thus overcoming the limitations of traditional radar in terms of detection range and accuracy.

[0028] According to one aspect of the present invention, this approach effectively avoids spectral aliasing caused by traditional FFT analysis and improves the continuity of wind field data detection. Attached Figure Description

[0029] Figure 1 This diagram illustrates the steps of a wind profiler radar signal and data processing method according to one embodiment of the present invention. Detailed Implementation

[0030] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The embodiments cannot be described in detail here, but the embodiments of the present invention are not limited to the following embodiments.

[0031] like Figure 1 As shown, according to one embodiment of the present invention, a wind profiler radar signal and data processing method of the present invention includes: S1. Construct the working parameter combination of the wind profiler radar, and design the radar detection mode of the wind profiler radar based on the working parameter combination; S2. The wind profiler radar acquires external wind field data based on radar detection modes to obtain wind profiler radar signals; S3. Perform signal processing based on the obtained wind profiler radar signal; including pulse compression, moving target detection and Doppler spectrum analysis of the wind profiler radar signal; S4. Extract wind profiler radar data from the processed wind profiler radar signal and process the wind profiler radar data: including wind field inversion and quality control of the wind profiler radar data. S5. Construct a three-dimensional wind field based on the processed wind profile radar data.

[0032] According to one embodiment of the present invention, the essence of wind profiler radar's echo detection is the scattering of electromagnetic waves by the non-uniform structure of the atmospheric refractive index caused by atmospheric turbulence. Therefore, the echo principle of wind profiler radar is based on the relationship between the atmospheric refractive index structure constant and the radar reflectivity. Tatarski derived this result using Kolmogrove's concept of local isotropic turbulence, and expressed it as: (1) in, For radar reflectivity, For the radar wavelength, Represents the atmospheric refractive index structure constant; where, The value varies greatly depending on the season and regional weather, and comprehensive measurement data is still lacking. Doviak derived the value based on data measured in Colorado during the winter months in the United States. The change in value with height can be expressed as: (2) in, The height is expressed in meters (m).

[0033] When the turbulent atmospheric medium fills the radar illumination volume, the expression for the radar echo power of the wind profile can be obtained according to the meteorological radar equation: (3) in, For transmission power, The effective area of ​​the antenna. For distance, For distance resolution, For feeder transmission efficiency.

[0034] Substituting equation (1) into equation (3), we get: (4).

[0035] Due to clear skies and atmospheric conditions Generally, the size is relatively small. In order to receive weak clear-sky echoes, according to equation (4), wind profiler radar should generally have a larger [specification / capacity]. Due to the use of longer wavelengths, the antenna area is generally larger to meet antenna gain requirements. Even so, wind profiler radar struggles to receive weak echo signals. Therefore, wind profiler radar requires pulse accumulation in signal processing to improve the signal-to-noise ratio.

[0036] Pulse accumulation is divided into coherent accumulation and non-coherent accumulation. Coherent accumulation, also called coherent averaging or pre-detector accumulation, is completed before the echo signal envelope is detected, at which point the echo pulses have a strict phase relationship. Pulses from the same range library... When consecutive echo pulses are accumulated, the output signal-to-noise ratio will be improved. The number of coherent averaging iterations cannot be infinitely large in practice; it must be limited by the maximum unambiguous Doppler velocity. The pulse repetition period is... Time domain echo after After multiple averaging, the maximum unambiguous Doppler velocity for: (5).

[0037] then (6).

[0038] Obtained sequentially from the same distance library Averaging the power spectral density values ​​at each corresponding frequency of the power spectrum can also improve the signal-to-noise ratio (SNR). This process is called spectral averaging. Spectral averaging is also known as incoherent accumulation or post-detection accumulation. While incoherent averaging can also improve the SNR, the improvement is less significant due to the SNR loss caused by the nonlinearity of the envelope detector. and The frequency of the spectrum should not be too high, otherwise it will reduce the temporal resolution of radar detection. It is generally selected between several and a dozen times.

[0039] Therefore, in step S1, the step of constructing the working parameter combination of the wind profiler radar includes: height resolution, maximum unambiguous Doppler velocity, velocity resolution, and time resolution; wherein, the height resolution is set in multiple ways based on the detection height; the maximum unambiguous Doppler velocity is set in multiple ways based on the detection height; and the time resolution is obtained based on the type of height resolution and the number of detection beams.

[0040] In this embodiment, as shown in formula (4), the height resolution is related to the maximum detection height. For wind profiler radar, it usually adopts multiple observation modes. One mode uses short pulse transmission to obtain good height resolution and minimum detection height, but the maximum detection height is relatively limited, which is called low mode. Another mode uses long pulse transmission (or pulse compression technology) to obtain a higher maximum detection height, but the height resolution is slightly worse, and the first detection height is higher, which is called high mode. As long as the detection height ranges of different modes overlap, the wind profiler can be linked. Taking tropospheric wind profiler radar as an example, in order to obtain better low-altitude performance and height resolution, the minimum detection height is required to be 150m and the height resolution is 75m. In order to ensure these two indicators, the low-altitude mode must use a pulse width of less than 0.5us. Due to the small pulse width, its detection height is greatly affected. According to actual detection tests, under the condition of a detection pulse width of 0.5us, the maximum detection height is around 3km. At altitudes above 6km, due to As altitude decreases exponentially, detection capability becomes a significant issue, necessitating the use of larger detection pulse widths. Therefore, three altitude resolution options were selected: low, medium, and high, at 75m, 150m, and 300m respectively. This effectively addresses the conflict between detection altitude and range resolution, ensuring optimal overall radar performance. In particular, the altitude resolution was balanced with the detection altitude through pulse compression technology, combining short and long pulses and ensuring overlapping detection altitudes between modes. This effectively achieves comprehensive detection coverage under clear-sky conditions, significantly enhancing the detection effectiveness of this approach.

[0041] In this embodiment, according to equation (5), the maximum unambiguous Doppler velocity depends on the pulse repetition period and the number of coherent averages. To reduce the probability of velocity ambiguity, the maximum unambiguous Doppler velocity should not be too small. For wind profiler radar, the zenith angle of its slant beam is generally 15°, so the projection of the maximum wind speed of 80 m / s onto the slant beam is 20 m / s, and the falling velocity of raindrops is generally less than 10 m / s (which can be estimated using empirical formulas). According to equation (7), the Doppler velocity measured by the vertical beam of the wind profiler radar is generally less than 10 m / s, while the Doppler velocity measured by the slant beam is generally less than 30 m / s. In fact, this is the extreme case, because the falling velocity of raindrops generally increases gradually with altitude from 8km, only reaching a relatively stable terminal velocity below 3km. Meanwhile, the horizontal wind speed at low altitudes is generally less than 30m / s. Based on this, the maximum unambiguous Doppler velocity is set in two ways: 15m / s at detection altitudes below 5km and 20m / s at detection altitudes above 5km. This maximum unambiguous Doppler velocity setting effectively ensures that velocity ambiguity will not occur in most cases. Even if ambiguity does occur, it is only a single instance and can be identified during data processing. Therefore, a transmission strategy combining short and long pulses, along with pulse compression technology, is adopted. This ensures high resolution in the near-field (e.g., 0-5km) and sufficient detection sensitivity in the far-field (e.g., 5-20km). Furthermore, the high overlap design between modes eliminates the vertical detection blind zone caused by traditional single-pulse modes, improving the continuity of the wind field's vertical profile.

[0042] In this embodiment, velocity resolution directly affects velocity measurement accuracy. Its value must be lower than the wind speed measurement accuracy value to ensure the required wind speed measurement accuracy. The wind speed measurement accuracy requirement for wind profiler radar is generally 1 m / s. Therefore, the projection value on the oblique beam of a wind profiler radar with a zenith angle of 15° is 1 m / s × sin15° = 0.25 m / s. This is the basic requirement for velocity resolution. Considering that a smaller spectral line interval is beneficial for target detection, the velocity resolution value is set to be less than 0.2 m / s. Furthermore, at this velocity resolution, the detection capability under clear sky conditions and weak wind fields (such as boundary layer light wind circulation) is significantly enhanced, and the wind speed error is reduced.

[0043] In this embodiment, the time resolution is expressed as:

[0044] in, Indicates the beam number. The category number indicating the height resolution. For time-domain accumulation, For the number of points in the Fast Fourier Transform (FFT), The spectral mean The number of detection modes used. The number of detection beams used.

[0045] By setting the time resolution as described above, and comprehensively considering factors such as the number of transmitted beams, altitude resolution, number of modes, number of time-domain accumulations, number of Fast Fourier Transform (FFT) points, and number of spectral averages, parallel processing of data and transmission is achieved. This is also more beneficial for ensuring sufficient scanning capability at a refresh rate of seconds. Through the deep coupling of physical constraints and algorithm requirements, the dilemma of "high resolution, large coverage, high precision, and real-time performance" in weather radar is solved.

[0046] According to one embodiment of the present invention, in step S1, the step of designing the radar detection mode of the wind profiler radar based on the combination of operating parameters involves determining the detection mode parameters in the radar detection mode based on the combination of operating parameters to complete the design of the radar detection mode. The detection mode parameters include: pulse width, pulse repetition period, pulse coherence averaging count, Fast Fourier Transform (FFT) points, and spectral average. The steps then include: The pulse width is determined based on the height resolution requirements, and the pulse repetition period is determined based on the detection height matching the height resolution; in this embodiment, the pulse repetition period... The corresponding detection range should ideally be at least 1.5 times the required detection altitude. This avoids the near-field blind zone or insufficient far-field sensitivity problems caused by traditional fixed pulse parameters, and enables continuous and stable detection of meteorological targets at all altitudes.

[0047] The pulse coherence average number of times is determined based on the maximum unambiguous Doppler velocity formula and the determined maximum unambiguous Doppler velocity and pulse repetition period; in this embodiment, due to the pulse repetition period... Once determined, the average number of pulse coherences can be determined using equation (5). Furthermore, the determined pulse coherence average number of times... It not only suppresses the contamination of the spectrum by random noise, but also prevents the loss of information on rapid changes in the wind field due to excessive averaging, and significantly reduces the ambiguity probability of velocity inversion.

[0048] The Fast Fourier Transform (FFT) points are determined based on the maximum unambiguous Doppler velocity and velocity resolution. In this embodiment, the velocity resolution of the wind profiler radar is the ratio of the maximum unambiguous Doppler velocity to half the FFT points, thus determining the specific value of the FFT points. Specifically, the FFT points are not less than 2 × (maximum unambiguous Doppler velocity value / velocity resolution). By scientifically matching the FFT points with the velocity resolution requirements, spectral leakage or wasted computational resources caused by traditional empirical values ​​are avoided, ensuring the effective extraction of weak wind field signals while maintaining reasonable real-time requirements.

[0049] With atmospheric stabilization time as a constraint, a single-beam dwell time range is set. Under the premise of satisfying the single-beam dwell time range, the spectral average is set to the maximum integer value that meets the signal-to-noise ratio improvement requirement. In this embodiment, the set single-beam dwell time range is the dwell time range of the Moy beam pointing upwards towards the berberine genus, generally 20s~30s. Therefore, the spectral average is determined based on the following formula, expressed as:

[0050] in, Indicates the dwell time of a single beam. This indicates the average number of pulse coherences.

[0051] Based on this, while meeting the requirements for monitoring the temporal evolution of wind fields, the signal-to-noise ratio is improved through sufficient signal accumulation, resolving the contradiction between data quality and refresh rate in the traditional fixed dwell time mode. By modeling the physical correlation between parameters, the radar system can automatically generate the optimal detection mode under different observation scenarios.

[0052] According to one embodiment of the present invention, in step S2, the wind profiler radar acquires external wind field data based on radar detection mode to obtain a wind profiler radar signal. The wind profiler radar uses a Doppler transceiver system to acquire external wind field data to obtain a wind profiler radar signal. Specifically, the wind profiler radar uses either 3 or 5 beams for detection and acquisition. If 3 beams are used, one beam is vertically pointing (hereinafter referred to as the mid-beam), and the other two are azimuthally orthogonal inclined pointing. The Doppler velocities of each beam are measured, and under the assumption of wind field horizontal uniformity, three wind components are obtained by simultaneously solving these systems. If 5 beams are used for detection and acquisition, one beam is vertically pointing (hereinafter referred to as the mid-beam), and the other four are azimuthally orthogonal inclined pointing. The Doppler velocities of each beam are measured, and under the assumption of wind field horizontal uniformity, three wind components are obtained by simultaneously solving these systems.

[0053] According to one embodiment of the present invention, in step S3, signal processing is performed based on the obtained wind profiler radar signal; wherein, the steps of pulse compression, moving target detection, and Doppler spectrum analysis of the wind profiler radar signal include: Acquire orthogonal I and Q video signals output by the wind profiler radar; A / D sampling and saturation detection are performed on the quadrature I and Q video signals. For the intermediate frequency direct sampling signal, saturation detection is performed first, and the receiving channel is adjusted through the AGC control loop to ensure that the input data is not saturated before sampling. Temporal filtering is employed, using a linear phase filter to suppress interference from quadratic range-folded echoes and rain echoes, thereby improving the signal-to-noise ratio (SNR) and reducing the computational complexity of the Fast Fourier Transform (FFT). In this embodiment, the measured signal echoes include not only the useful signal within the detection range but also quadratic range-folded echoes caused by interference signals from outside that range, such as meteorological echoes, aircraft, wind turbines, and birds. Rain echoes and aircraft echoes are generally much stronger than turbulent echoes. The presence of interference echoes hinders the detection of the useful signal and increases the dynamic range of the signal. Therefore, temporal filtering before processing the useful signal can suppress interference echoes, improve the echo SNR, and simultaneously reduce the computational complexity of the FFT, thus shortening the computation time.

[0054] Time-domain coherent averaging, adding echoes in phase according to the phase relationship of adjacent periods, improves echo signal-to-noise ratio and target detection capability; Spectral analysis involves windowing the signal after coherent averaging in the time domain, and then performing FFT analysis on the orthogonal I and Q video signals after processing and combining them into a complex signal. Frequency domain filtering employs a filter with good rectangular coefficients and a small notch width to suppress ground clutter while preserving meteorological echoes. In this embodiment, the notch width of the filter is set to be less than or equal to 1Hz, thereby achieving sufficient suppression of ground clutter. The rectangular coefficients of the filter ensure that the actual frequency response curve closely approximates the frequency response curve of an ideal rectangular filter, enabling a rapid decrease from passband gain to stopband attenuation near the cutoff frequency, thus reducing interference between frequency bands.

[0055] Spectral averaging involves averaging the values ​​at corresponding frequencies of P power spectral density functions to further improve the signal-to-noise ratio. In this embodiment, the spectral averaging number can be set to 8, 12, 16, etc.

[0056] By real-time detection of signal saturation and dynamic adjustment of the receiving gain, nonlinear distortion caused by strong echoes and quantization errors of weak signals are avoided, thus expanding the system's dynamic range and ensuring synchronous and accurate acquisition of meteorological targets of varying intensities, from heavy precipitation to weak turbulence. A digital filter with strictly linear phase characteristics is employed to eliminate radio frequency interference and ground clutter while preserving the integrity of the meteorological echo phase information. Combined with time-domain coherent accumulation technology based on phase consistency, the signal-to-noise ratio is improved, significantly enhancing the detection probability of weak meteorological targets. Adaptive selection of the transform length to satisfy the Fast Fourier Transform (FFT) points ensures a spectral resolution better than 1 Hz. A frequency-domain filter with a steep transition band introduces only less than 0.2 dB of signal amplitude distortion while suppressing ground clutter. Frequency point-to-point averaging of P independent power spectra is performed, suppressing random noise fluctuations through statistical characteristics, reducing the standard deviation of the final wind speed estimate, and particularly improving the reliability of boundary layer wind field inversion under clear-sky conditions.

[0057] By employing pre-saturation detection and AGC dynamic adjustment, the intermediate frequency signal is ensured to complete digital sampling within the optimal dynamic range, avoiding nonlinear distortion caused by strong echoes and quantization errors of weak signals, thereby improving the effective dynamic range of the system and eliminating spectral leakage problems introduced by signal saturation. A linear phase filter is used to accurately remove interference components such as double-folded echoes, reducing the computational load of invalid Fast Fourier Transform (FFT) points while preserving the phase characteristics of meteorological echoes. Combined with time-domain coherent accumulation technology based on phase consistency, the signal-to-noise ratio is improved, significantly enhancing the detection probability of low-observable targets under clear-sky conditions. Spectral leakage energy is suppressed by adaptive windowing and scientifically setting the number of FFT points (matching effective bandwidth and Doppler resolution requirements). With the help of a frequency domain filter with a steep transition band, the ground clutter suppression ratio and meteorological signal amplitude distortion are simultaneously optimized. Frequency point-to-point statistical averaging is performed on P independent power spectra, and random noise fluctuations are suppressed by the law of large numbers, making the standard deviation of the final wind speed estimate approach the theoretical Cramer-Rao lower limit, especially achieving control of velocity measurement errors in clear-sky turbulence observations.

[0058] According to one embodiment of the present invention, in step S4, wind profiler radar data is extracted from the processed wind profiler radar signal, and the wind profiler radar data is processed: wherein the step of performing wind field inversion and quality control on the wind profiler radar data includes: S41. Target Detection and Spectral Moment Calculation: The echo signal of the meteorological target is identified by the power spectrum peak detection algorithm, and the noise level, Doppler velocity, spectral width and signal-to-noise ratio are calculated based on the power spectrum peak. In this embodiment, the signal spectral peak is identified from the entire echo signal power spectrum by target detection. Based on the noise level, the first moment (i.e., Doppler velocity), the second moment (i.e., Doppler velocity spectral width) and the signal-to-noise ratio can be obtained by calculating the signal spectrum above the noise level.

[0059] S42. Consistent averaging: Time-domain averaging is performed on multiple Doppler velocity measurements within the same beam direction and range gate to suppress random measurement errors. This method can simply, practically, and effectively eliminate isolated interference from aircraft, automobiles, etc. In this embodiment, 10 measurements are averaged within the same range gate.

[0060] S43. Wind vector inversion: Based on Doppler velocity measurements with different beam directions, the horizontal wind vector components (i.e., u, v, and w components) are solved by the least squares method. S44. Data quality control: The reliability of the data is marked by wind shear check, continuity check and two-dimensional median check, and doubtful or invalid data is removed; S45. Anti-interference processing: Half-plane cancellation (utilizing ground clutter symmetry), data smoothing, pattern recognition, multi-peak extraction, median filtering, and cluster analysis are used to process the "contaminated" power spectrum.

[0061] Through the above settings, the power spectrum peak detection technology based on adaptive design effectively distinguishes meteorological echoes from interference signals in complex noise backgrounds, solving the problem of missed detection in weak signal scenarios using traditional fixed threshold methods, and significantly improving the system's ability to identify low-observable targets under clear sky conditions. The temporal coherent averaging algorithm is used to fuse multiple observation data, and a phase consistency verification mechanism is used to suppress random measurement errors, effectively improving the reliability and repeatability of Doppler velocity measurements. Based on a multi-beam scanning strategy and least squares optimization algorithm, the inherent geometric error limitations of single-beam measurements are overcome, achieving accurate calculation of horizontal wind field components and significantly improving the analytical capability of complex wind field structures. Through multi-dimensional wind field feature verification (including spatial continuity, temporal consistency, and physical rationality), an automatic data credibility labeling mechanism is established, providing a reliable basis for subsequent data applications. By comprehensively utilizing intelligent algorithms such as spectral domain cancellation, pattern recognition, and cluster analysis, accurate extraction of meteorological echoes is achieved in scenarios with multiple interferences, effectively solving the misjudgment problem of traditional methods in environments with strong ground clutter and biological targets. By establishing a closed-loop quality control chain from raw signal analysis to final product generation, the system is equipped with the ability to adapt to complex meteorological conditions and interference environments, providing highly reliable wind field observation data for meteorological monitoring.

[0062] According to one embodiment of the present invention, step S44, the data quality control step, includes: S441. Spatial consistency test based on wind field shear threshold; S442. Outlier removal based on time continuity constraints; S443. A two-dimensional median filter is used to smooth the wind field data.

[0063] By setting a wind field shear threshold, the physical rationality of wind speed gradients at adjacent distance gates or altitude layers is verified, effectively identifying and eliminating discontinuous wind field data caused by beam obstruction or velocity ambiguity, ensuring that the inversion results conform to the fluid continuity law of atmospheric motion. Dynamic constraints are established based on the time-scale characteristics of atmospheric evolution, automatically filtering out abrupt data points that do not conform to meteorological evolution laws within adjacent scanning cycles, solving the problem of false wind field jumps caused by instantaneous interference or system noise. A two-dimensional median filtering algorithm is used to suppress the impact of isolated noise points on the overall wind field while retaining the small-scale characteristics of the real wind field, especially effectively eliminating local data anomalies caused by radar sidelobe interference, improving the spatial analysis usability of wind field products. By integrating atmospheric dynamic constraints and digital signal processing technology, the final output three-dimensional wind field data simultaneously meets the requirements of physical rationality and measurement reliability.

[0064] According to one embodiment of the present invention, step S45, the anti-interference processing step, includes: S451. Suppress ground clutter using a half-plane cancellation algorithm; S452. Spectral smoothing is performed using a Savitzky-Golay filter; S453. Identifying multi-peak spectral features based on K-means clustering to separate overlapping meteorological targets; S454. Combine morphological median filtering to eliminate impulse interference noise.

[0065] By establishing a dynamic clutter reference spectrum model and performing adaptive cancellation operations, the system achieves directional suppression of ground clutter while preserving meteorological echo phase information, solving the technical problem of weak precipitation signal loss caused by traditional static filtering. A Savitzky-Golay filter is used to jointly perform differential smoothing in the time and frequency domains, effectively suppressing random noise while maintaining the integrity of meteorological spectral peak shape parameters, avoiding the spectral broadening effect caused by conventional moving average methods. Multimodal decomposition of complex power spectra is performed based on the K-means clustering algorithm, and radial velocity decoupling of adjacent meteorological targets (such as stratiforms and convective clouds) is achieved by automatically identifying spectral peak subspaces, overcoming the application limitations of traditional single-peak detection models. Combining morphological opening operations and adaptive median filtering techniques, abnormal spectral lines generated by transient interference such as lightning are detected and removed simultaneously in the time and frequency domains, maintaining the temporal continuity of meteorological echoes. By establishing a technical closed loop of "feature extraction-interference modeling-intelligent filtering," the system can maintain an effective data acquisition rate even under harsh observation environments.

[0066] According to one embodiment of the present invention, a wind profiler radar signal and data processing system is provided, comprising: a radar detection mode setting module, a signal receiving module, a signal processing module, a data processing module, and a display module; wherein, the radar detection mode setting module is used to construct a combination of operating parameters for the wind profiler radar and design the radar detection mode of the wind profiler radar based on the combination of operating parameters; the signal receiving module is used to receive the wind profiler radar signal output during the process of the wind profiler radar acquiring external wind field data based on the radar detection mode; the signal processing module is used to perform signal processing on the wind profiler radar signal, including pulse compression, moving target detection, and Doppler spectrum analysis; the data processing module is used to extract wind profiler radar data from the processed wind profiler radar signal and process the wind profiler radar data, including wind field inversion and quality control of the wind profiler radar data; the display module is used to construct and display a three-dimensional wind field from the processed wind profiler radar data.

[0067] The above description is merely an example of a specific solution of the present invention. For any devices and structures not described in detail herein, it should be understood that they are implemented using common devices and methods already available in the art.

[0068] The above description is merely one embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the invention by those skilled in the art. Any modifications, equivalent substitutions, or improvements 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 processing wind profiler radar signals and data, characterized in that, include: S1. Construct the working parameter combination of the wind profiler radar, and design the radar detection mode of the wind profiler radar based on the working parameter combination; S2. The wind profiler radar collects external wind field data based on the radar detection mode to obtain a wind profiler radar signal; S3. Perform signal processing based on the obtained wind profiler radar signal; wherein, pulse compression, moving target detection and Doppler spectrum analysis are performed on the wind profiler radar signal; S4. Extract wind profiler radar data from the processed wind profiler radar signal and process the wind profiler radar data: including wind field inversion and quality control of the wind profiler radar data. S5. Construct a three-dimensional wind field based on the processed wind profile radar data.

2. The wind profiler radar signal and data processing method according to claim 1, characterized in that, In step S1, the step of constructing the working parameter combination of the wind profiler radar includes: height resolution, maximum unambiguous Doppler velocity, velocity resolution, and time resolution. The height resolution can be set in multiple ways based on the detection height; The maximum unambiguous Doppler velocity can be set in multiple ways based on the detection height; The temporal resolution is obtained based on the type of height resolution and the number of probe beams.

3. The wind profiler radar signal and data processing method according to claim 2, characterized in that, The height resolution is set to three levels: low, medium, and high, which are 75m, 150m, and 300m respectively. There are two settings for the maximum unambiguous Doppler velocity, namely: At detection altitudes below 5 km, the maximum unambiguous Doppler velocity is set to 15 m / s, and at detection altitudes above 5 km, the maximum unambiguous Doppler velocity is set to 20 m / s. The velocity resolution is less than 0.2 m / s; The time resolution is expressed as: in, Indicates the beam number. The category number indicating the height resolution. For time-domain accumulation, For the number of points in the Fast Fourier Transform (FFT), The spectral mean The number of detection modes used. The number of detection beams used.

4. The wind profiler radar signal and data processing method according to claim 3, characterized in that, In step S1, the step of designing the radar detection mode of the wind profiler radar based on the combination of operating parameters involves determining the detection mode parameters in the radar detection mode based on the combination of operating parameters to complete the design of the radar detection mode. The detection mode parameters include: pulse width, pulse repetition period, pulse coherence averaging count, Fast Fourier Transform (FFT) points, and spectral average. The steps include: The pulse width is determined based on the height resolution requirements, and the pulse repetition period is determined based on the detection height that matches the height resolution. The pulse coherence average number of times is determined based on the formula for the maximum unambiguous Doppler velocity and the determined maximum unambiguous Doppler velocity and pulse repetition period; The number of points for the Fast Fourier Transform (FFT) is determined based on the maximum unambiguous Doppler velocity and velocity resolution. With atmospheric stabilization time as a constraint, the dwell time range of a single beam is set. Under the premise of satisfying the dwell time range of a single beam, the average number of spectra is set to the maximum integer value that meets the signal-to-noise ratio improvement requirements.

5. The wind profiler radar signal and data processing method according to claim 4, characterized in that, In step S2, the wind profiler radar acquires external wind field data based on the radar detection mode to obtain a wind profiler radar signal. In this step, the wind profiler radar uses a Doppler transceiver system to acquire external wind field data to obtain a wind profiler radar signal. The wind profiler radar uses 3 or 5 beams for detection and data acquisition.

6. The wind profiler radar signal and data processing method according to claim 5, characterized in that, In step S3, signal processing is performed based on the obtained wind profiler radar signal; wherein the steps of pulse compression, moving target detection, and Doppler spectrum analysis of the wind profiler radar signal include: Acquire orthogonal I and Q video signals output by the wind profiler radar; A / D sampling and saturation detection are performed on the quadrature I and Q video signals. For the intermediate frequency direct sampling signal, saturation detection is performed first, and the receiving channel is adjusted through the AGC control loop to ensure that the input data is not saturated before sampling. Time-domain filtering involves designing a linear phase filter to suppress secondary range folding echo interference and rain echo interference, thereby improving the signal-to-noise ratio and reducing the computational load of FFT. Time-domain coherent averaging, adding echoes in phase according to the phase relationship of adjacent periods, improves echo signal-to-noise ratio and target detection capability; Spectral analysis involves windowing the coherently averaged signal in the time domain, combining them into a complex signal, and then performing FFT analysis. Frequency domain filtering employs filters with good rectangular coefficients and small notch widths to suppress ground clutter while preserving meteorological echoes. Spectral averaging involves averaging the values ​​of P power spectral density functions at corresponding frequencies to further improve the signal-to-noise ratio.

7. The wind profiler radar signal and data processing method according to claim 6, characterized in that, In step S4, wind profiler radar data is extracted from the processed wind profiler radar signal, and the wind profiler radar data is processed. The steps of wind field inversion and quality control of the wind profiler radar data include: S41. Target Detection and Spectral Moment Calculation: Identify the echo signals of meteorological targets using a power spectrum peak detection algorithm, and calculate the noise level, Doppler velocity, spectral width, and signal-to-noise ratio based on the power spectrum peak; S42. Consistent averaging: Time-domain averaging of multiple Doppler velocity measurements within the same beam direction and the same distance gate is performed to suppress random measurement errors; S43. Wind vector inversion: Based on Doppler velocity measurements with different beam orientations, the horizontal wind vector components are calculated using the least squares method; S44. Data quality control: The reliability of labeled data is verified through wind shear checks, continuity checks, and two-dimensional median checks; S45. Anti-interference processing: The power spectrum is processed using half-plane cancellation, data smoothing, pattern recognition, multi-peak extraction, median filtering, and cluster analysis.

8. The wind profiler radar signal and data processing method according to claim 7, characterized in that, Step S44, the data quality control step, includes: S441. Spatial consistency test based on wind field shear threshold; S442. Outlier removal based on time continuity constraints; S443. A two-dimensional median filter is used to smooth the wind field data.

9. The wind profiler radar signal and data processing method according to claim 8, characterized in that, Step S45, the anti-interference processing step, includes: S451. Suppress ground clutter using a half-plane cancellation algorithm; S452. Spectral smoothing is performed using a Savitzky-Golay filter; S453. Identifying multi-peak spectral features based on K-means clustering to separate overlapping meteorological targets; S454. Combine morphological median filtering to eliminate impulse interference noise.

10. A wind profiler radar signal and data processing system, characterized in that, include: The radar detection mode setting module is used to construct the working parameter combination of the wind profiler radar and design the radar detection mode of the wind profiler radar based on the working parameter combination. The signal receiving module is used to receive the wind profiler radar signal output by the wind profiler radar during the process of collecting external wind field data based on the radar detection mode. The signal processing module is used to process the wind profiler radar signal; specifically, it performs pulse compression, moving target detection, and Doppler spectrum analysis on the wind profiler radar signal. The data processing module is used to extract wind profiler radar data from the processed wind profiler radar signal and process the wind profiler radar data, including wind field inversion and quality control of the wind profiler radar data. The display module is used to construct and display the three-dimensional wind field from the processed wind profile radar data.

Images