CN115932759A - Differential propagation phase shift rate measurement method of dual-polarization Doppler weather radar - Google Patents

Abstract

The invention relates to a differential propagation phase shift rate measuring method of a dual-polarization Doppler weather radar, which comprises the following steps: basic data processing steps: sequentially carrying out ground clutter identification and clutter filtering processing on the original data; phi _DP And (3) data quality control: sequentially processing data processed by basic data processing step _DP Initial radial filling,. Phi _DP Data quality classification process, phi _DP Unfolding treatment,. Phi. _DP Radial interpolation process and phi _DP Filtering; k _DP Calculating and processing steps: will pass through phi _DP Phi obtained in the data quality control step _DP The data sequentially passes through K _DP Least squares method of calculating sum K _DP The measured value is judged and processed to obtain K _DP And (4) measuring the value. The invention can accurately estimate K _DP The parameters may be precipitation estimation, attenuation correction and precipitation of dual polarization weather radarThe water particle phase state identification provides effective data support, and further provides important auxiliary data for early warning and forecasting of dangerous weather.

Description

Differential propagation phase shift rate measuring method of dual-polarization Doppler weather radar

Technical Field

The invention relates to the technical field of radar detection, in particular to a differential propagation phase shift rate measuring method of a dual-polarization Doppler weather radar.

Background

The dual-polarization Doppler weather radar has great potential in the aspects of precipitation estimation, precipitation particle phase state identification and the like, and differential propagation phase shift rate (K) _DP ) Is an important parameter obtained by dual-polarization weather radar measurement due to K _DP Is hardly influenced by rainfall attenuation, radar calibration error and radar beam filling coefficient, is almost in linear change relation with rainfall intensity R, is not sensitive to rainfall spectrum change but has excellent characteristics of unique sensitivity and the like on rainfall particles, particularly on the improvement of aragonite and snowflake phase states, and therefore K is _DP The data has wide application prospect in radar attenuation correction, radar precipitation estimation, dangerous weather early warning and forecast and other aspects.

Therefore, how to measure the differential propagation phase shift (Φ) from a dual polarization radar _DP ) To obtain accurate and reliable K _DP Still is a difficult problem to be solved; a number of related studies in the prior art have been performed from Φ _DP Pretreatment of (A) or (K) _DP The algorithm design of the self-body is started to obtain reliable K _DP And measuring the parameter. However, currently, there is no reliable K in the business yet _DP Application of algorithms, K measured by different algorithms _DP The result difference is obvious, and the radial inversion result also has the problems of edge abnormal values, radial burrs, obvious jitter and the like. In addition, part on K _DP The algorithmic study of parameter estimation is limited by the sample or processing time under simulation, and does not realize engineering application in actual probing.

It is noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present disclosure and therefore may include information that does not constitute prior art that is already known to a person of ordinary skill in the art.

Disclosure of Invention

The invention aims to overcome the defects of the prior art and provides a pairThe differential propagation phase shift rate measuring method of the polarization Doppler weather radar solves the problem of K in the prior art _DP The measurement is not adequate.

The purpose of the invention is realized by the following technical scheme: a method of differential propagation phase shift rate measurement of a dual polarization doppler weather radar, the method of measurement comprising:

basic data processing steps: sequentially carrying out ground clutter recognition and clutter filtering processing on the original data;

Φ _DP and a data quality control step: sequentially processing the data after the basic data processing step _DP Initial radial filling,. Phi _DP Data quality classification process, phi _DP Unfolding treatment of phi _DP Radial interpolation process and phi _DP Filtering;

K _DP calculating and processing steps: will pass through phi _DP Phi obtained in the data quality control step _DP The data sequentially passes through K _DP Least squares method of calculating sum K _DP The measured value is judged and processed to obtain K _DP And (4) measuring the value.

The ground clutter identification processing comprises: by analyzing different characteristics and distribution of ground clutter and weather echoes, determining horizontal texture TDBZ of reflectivity, radial change SPIN of reflectivity and clutter phase array calibration value CPA on a distance unit in a certain range as an identification fuzzy basis, and selecting a reasonable threshold value by using a fuzzy logic method for identification;

the clutter filtering process includes: and marking the distance unit containing the ground clutter, and performing ground clutter suppression processing on the marked distance unit by setting clutter filters with different suppression degrees of a time domain and a frequency domain.

The phi _DP The radial initial fill includes: for phi _DP Filling the initial effective value with the data in the radial direction by filling the ineffective cells with the adjacent effective distance bins, and performing the filling of phi _DP Data quality classification processing or _DP Unfolding treatment or phi _DP Radial interpolation process or phi _DP After the filtering processing, invalid data or data which does not meet the requirements are required to be processedAnd filling the effective value to update the processed data.

The phi _DP The data quality classification processing comprises the following steps:

sequentially calculating phi of each distance unit along the radial direction _DP Standard deviation and ZDr standard deviation, and setting corresponding threshold values respectively;

meanwhile, the data quality is judged by combining the zero lag correlation coefficient and the signal-to-noise ratio of the current distance library and distributing a method meeting a threshold judgment condition, and when continuous M distance libraries meet the threshold judgment, the phi marked as good data quality _DP Data, when not meeting the threshold judgment, marked as difference data phi _DP The left start unit of the data interval.

Passing through the phi _DP After the data quality classification processing, it is necessary to determine whether the data marked as the difference data interval is large precipitation particles, which specifically includes the following contents:

calculating the average value of the reflectivity factors of the meteorological echo information in the region, and when the average value is more than or equal to a set value, comparing the average value of the zero lag correlation coefficient in the region with the standard deviation phi _DP Calculating the average value and standard deviation, judging whether the identification result of the interval is determined as a difference data interval or not by checking whether the identification result meets a threshold value or not, finally filling the effective value of the interval determined as the difference data, and updating the radial phi _DP Data is obtained.

The phi _DP The unfolding process includes:

according to phi _DP Determining a range of starting distances and phi of effective precipitation echoes in the radial direction _DP A starting value;

sequentially calculating phi of M distance libraries _DP Standard deviation and mean, the continuous distance library phi _DP Taking an initial distance library with the standard deviation smaller than a set fluctuation threshold and the zero lag correlation coefficient larger than a set threshold as an effective initial distance library of the current radial unfolding processing;

selecting a plurality of successive range bins for phi _DP Calculation of the standard deviation and mean value,. Phi _DP Phi of distance library interval with standard deviation smaller than threshold _DP Mean value ofAs the radial direction phi _DP An initial differential phase reference value of the unfolding process;

sequentially carrying out phi on all the remaining distance bins along the radial direction _DP Calculating standard deviation and average value, if phi is recognized _DP If there is abnormal fluctuation, the obtained phi _DP The average value is stored, and the estimated phi of the planar soul and the corresponding distance unit is used _DP Comparing the parameters to obtain a difference;

judging whether phi occurs or not by checking continuity between radial distance libraries _DP Folding, and performing differential phase unfolding processing on the units identified as being folded to recover the true phi _DP And (6) measuring the values.

Phi is _DP The radial interpolation process includes: phi marking two segments that are not adjacent as good data quality _DP Interpolation processing is carried out on the parameter estimation value area, and after interpolation filling is finished on the whole radial data, phi generated again in the radial direction is stored _DP And obtaining the difference result of the parameters.

Phi is _DP The filtering process includes: FIR low-pass filter pair phi adopting loop iteration _DP Filtering the measured value, specifically including the following:

filtering each distance library and unfolding processed phi before filtering _DP And comparing, when the absolute value of the difference value of the two values is greater than a set threshold value, performing iterative filtering of the next cycle by using the filtered data, otherwise performing iteration by using the data before filtering.

Said K _DP The least square method calculation comprises: selecting phi of M successive distance bins _DP Measured value by K _DP And phi _DP Calculating K by the estimation formula _DP For calculating K _DP The number N of the distance bins of the parameter is selected according to the measured value of the reflectivity factor of the corresponding distance bin.

Said K is _DP The measured value judgment and processing comprises the following steps:

in the radial direction to K _DP The measured values are checked, and when the measured values exceed a plurality of continuous distance bins and are a certain fixed value, K is determined _DP The measured value being an invalid valueAnd K of all distance bins in the radial direction _DP In the estimation, K marked as a bad data quality interval _DP Setting the value as an invalid value;

to K _DP The measured value is subjected to radial inspection, the echo property is judged through the characteristic parameters, and when the echo property is judged to be a meteorological echo, K is output _DP And if not, setting the measured value as an invalid value.

The invention has the following advantages: a method for measuring differential propagation phase shift rate of dual-polarization Doppler weather radar is implemented by aiming at various possible pairs K _DP Factor of influence of data quality and current K _DP Analyzing the problems of abnormal result, radial jitter fluctuation and the like in algorithm estimation, and respectively designing corresponding algorithms to phi used for estimation _DP The data is subjected to data quality control. For phi _DP The data is subjected to a relatively complete and comprehensive data quality control processing process, and the obtained K is processed _DP And (4) reasonably processing the data. The algorithm is applied to engineering and real-time data processing in dual-polarization weather radar detection. Analysis of estimated K _DP The spatial distribution of (1) is in accordance with the real precipitation condition and is equal to phi _DP And the distribution of the reflectivity factors is more consistent, the distribution characteristics of precipitation particles can be more reasonably reflected, and the designed K _DP The estimation algorithm has reasonable and radially continuous data processing effect.

Drawings

FIG. 1 is a schematic flow diagram of the present invention;

FIG. 2 shows the present invention _DP A flow diagram of data unfolding processing;

FIG. 3 is a diagram of the present invention _DP A flow diagram of data filtering processing;

FIG. 4 shows the invention K _DP Least squares schematic representation of the use.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the detailed description of the embodiments of the present application provided below in connection with the appended drawings is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present application without making any creative effort, shall fall within the protection scope of the present application. The invention is further described below with reference to the accompanying drawings.

The invention particularly relates to a differential propagation phase shift rate measuring method of a dual-polarization Doppler weather radar, which is used for realizing reasonable and feasible dual-polarization weather radar K _DP Estimation algorithm of parameters to obtain correct and accurate K _DP And (5) parameter estimation results. The double bias parameter K _DP The estimation algorithm is applied to a large amount of radar data for testing after the design is completed, the obtained estimation result has better consistency with the reflectivity factor on the whole, and particularly, the good consistency is kept on the development and change trends of the echo region and the radial direction. In addition, the estimation algorithm is applied to radar complete machine detection, and K meeting the weather echo characteristic is obtained _DP Measurement result, and can satisfy the pair K _DP Real-time estimation of the demand of the parameter.

In the process of K _DP Before parameter estimation, the double offset parameter phi needs to be aligned _DP The data quality control is carried out, and the estimation of K is guaranteed while the meteorological echo information is kept to the maximum extent _DP Phi of _DP Reliability and validity of data. For phi _DP The data quality control processing of the data comprises the following steps: ground feature identification, ground clutter suppression, phi, of raw data _DP Data quality analysis and classification, phi _DP Unfolding treatment,. Phi. _DP Radial interpolation processing and phi _DP The filtering process of (2). At a pair of phi _DP After the data quality control is finished, K is calculated by adopting a least square normal linear regression method _DP Parametric measurement values and checking for radial run-out of order dataAnd (5) removing.

In actual detection, the differential phase information directly estimated from radar detection is the total differential propagation phase Ψ _DP 。Ψ _DP Involving a forward propagation phase shift Φ _DP And a backscatter phase shift δ, as in the following equation:

Ψ _DP (r)＝Φ _DP (r)+δ(r)

the value of δ is close to zero under the condition that Rayleigh scattering is satisfied, at which time the detected total differential propagation phase can be directly approximated as a differential propagation phase shift Φ _DP For K _DP And (4) estimating parameters, wherein r is the length of the radar range bin.

Polarization parameter K _DP Is a phase shift (phi) by differential propagation _DP ) The slope of the phase change is calculated as phi _DP The quality of data will determine K _DP Accuracy and precision of the parameter estimation. Thus, in design K _DP Before the parameter estimation algorithm, first, the parameter is measured for phi _DP The parameters are subjected to data quality control to obtain a calculation K _DP The parameters provide reliable underlying data. Wherein the differential propagation phase shift rate K _DP And phi _DP The relationship of (a) is shown as follows:

in the formula, phi _DP (0) For the radar system initial differential phase, ε is the measurement error in the actual probe and is determined by other error sources, such as: errors caused by clutter of ground objects.

In actual detection, phi is caused by interference of non-meteorological targets such as ground clutter _DP The parameters have abnormal fluctuation phenomena, and the data quality is influenced. In order to reduce the data quality influence caused by ground clutter, ground clutter suppression processing is performed on the original data. In order to reduce the loss of the ground clutter suppression on the weather echo, the ground clutter is firstly identified.

Thus, as shown in fig. 1, the present invention specifically includes the following:

step one, basic data processing step: sequentially carrying out ground clutter identification and clutter filtering processing on the original data;

step two, phi _DP And (3) data quality control: sequentially processing the data after the basic data processing step _DP Initial radial filling,. Phi _DP Data quality classification process, phi _DP Unfolding treatment of phi _DP Radial interpolation process and phi _DP Filtering;

step three, K _DP Calculating and processing steps: will pass through phi _DP Phi obtained in the data quality control step _DP The data sequentially passes through K _DP Least squares method of calculating sum K _DP The measured value is judged and processed to obtain K _DP And (6) measuring the values.

Further, the ground clutter recognition processing comprises: by analyzing different characteristics and distribution of ground clutter and weather echoes, selecting horizontal texture TDBZ of reflectivity, radial variation SPIN of reflectivity and clutter phase array calibration value CPA on a distance unit with a proper threshold as an identification fuzzy basis, and selecting a reasonable threshold by using a fuzzy logic method for identification;

the expressions of TDBZ, SPIN and CPA are respectively as follows:

in the above equation, the TDBZ and SPIN parameters of the current range bin i are based on a concatenation of radar reflectivity factors (dBZ) estimated from successive M range binsAnd calculating a continued relation, namely an adjacent distance library. X _i Is the time IQ sequence value of the ith distance library. thresh is the decision threshold for the sum of the selected neighboring distance bin dBZ differences.

Wherein, clutter filtering process includes: and marking the distance unit containing the ground clutter, and performing ground clutter suppression processing on the marked distance unit by setting clutter filters with different suppression degrees of a time domain and a frequency domain.

Dual-polarization Doppler weather radar with dual-transmitting and dual-receiving system _DP The unambiguous measurement range of the parameters is generally from 0 to 360 °. When phi is _DP After the parameter exceeds the unambiguous range in estimation, the radar will give it a non-true value in the measurement range and return a folded measurement result, resulting in phi _DP The estimation result is no longer reliable. In the process of precipitation, phi is very easy to occur in the process of transmission of electromagnetic waves _DP Especially when the radar wavelength is short. Therefore, the radar needs to minimize the occurrence of folding in detection and design a reasonable algorithm for phi _DP Parametric unfolding.

φ _DP (0) Is a constant and does not directly affect K _DP Estimation of the parameters, but too high phi _DP (0) Is likely to cause phi _DP And (4) folding. The initial differential phase of the radar is determined by the system phases of the two channels H and V of the radar, and can be corrected by software. To reduce phi _DP The invention sets reasonable initial differential phase for radar system by software compensation, makes the initial phase of system close to the lower boundary of non-fuzzy measurement range, and makes full use of the whole phi _DP The measurement range is not obscured.

Due to phi _DP (0) Drift may occur and easily exceed phi when the propagation path is long, especially during precipitation or when detection is performed using high-band radar _DP Does not obscure the measurement range, therefore, it is also necessary to measure for phi _DP And performing phase unfolding processing. In order to reduce the differential propagation phase shift phi in practical detection _DP The existing pulsatility and various burrs eliminate the data quality influence caused by abnormal fluctuationTo make phi _DP The data has better continuity and the distribution of differential phase shift is more concentrated. Before unfolding processing, phi is first processed _DP And performing L-point median filtering processing on the data. The choice of L is related to the wavelength of the radar and the range bin length in practical applications.

In addition, the phi obtained due to the influence of interference of non-meteorological echo, disturbance of atmospheric environment field, fluctuation of particles, random error, system error and the like introduced during measurement _DP The measured value of (2) often has the phenomena of burr, large amplitude fluctuation oscillation and the like in the radial direction. For a better quality of phi _DP Data, glitches and ripples can be removed by filtering while phi is _DP In the case of poor data quality or low signal-to-noise ratio, the burrs or fluctuations existing in some radial distance bins will affect the overall data quality in the radial direction, resulting in estimated K _DP The parameter change is scattered, even the specific distribution characteristics can not be identified, so that K _DP The measured values of the parameters have a large error. Therefore, it is necessary to first _DP Data are specifically classified as better and worse _DP The data are processed separately.

Due to K _DP The estimation of the measured value is derived from distance accumulation least square regression, and when the continuity of the echo area in the radial direction is poor, especially in the process of small-range and local weather, the K at the edge of the echo is easily caused _DP Unreliable measurements, even resulting in outliers. To avoid the above problem and the subsequent processing algorithm consists in determining the effective distance unit, Φ _DP The influence caused by the discontinuity of echo edge in the data unfolding, smoothing and filtering needs to be firstly performed _DP And (4) initial filling in the radial direction.

Wherein phi _DP The radial initial fill includes: for phi _DP Filling the initial effective value with the data in the radial direction by filling the ineffective cells with the adjacent effective distance bins, and performing the filling of phi _DP Data quality classification process or _DP Unfolding treatment or phi _DP Radial interpolation process or phi _DP The invalid data or the number which does not meet the requirement after the filtering processingAnd carrying out effective value filling to update the processed data.

Further, Φ _DP The data quality classification processing comprises the following steps:

at the moment of carrying out phi _DP When data is classified, phi of each distance unit is calculated in turn along the radial direction _DP And the standard deviation and the ZDr standard deviation are combined with the zero lag correlation coefficient and the signal-to-noise ratio of the current distance library to respectively meet the threshold judgment condition. Marking as phi of better data quality when consecutive M distance bins satisfy threshold determination _DP Data, when not satisfied, is marked as poor data quality phi _DP The left start unit of the data interval. Based on the parametric characterization of the meteorological echoes and statistical analysis of the data, wherein the zero lag correlation coefficient is set to 0.7, Φ _DP The abnormal fluctuation threshold value was set to 15 °, and the abnormal fluctuation threshold value of Zdr was set to 2.0.

To avoid the influence of special large-grained precipitation particles, such as hail, it is necessary to determine whether the data marked as the poor data interval is large-grained precipitation particles. When the judgment is carried out, the average value of the reflectivity factors of the meteorological echo information in the interval is calculated, and when the average value is more than or equal to 30.0dbz, the average value of the zero lag correlation coefficient in the interval is compared with the standard deviation phi _DP The average value and the standard deviation are calculated, and whether the identification result of the interval is determined to be a poor data interval is judged through checking whether the threshold is met. Filling the regions marked as poor data with effective values, and updating phi in radial direction _DP Data is obtained.

Further, as shown in FIG. 2, Φ _DP The unfolding process includes:

at a pair of phi _DP After the data are classified, begin to process phi _DP And performing unfolding treatment. The processing method should first depend on phi _DP Determining a starting range and a range of effective precipitation echoes in the radial direction _DP A starting value. This starting range bin is set as a cell where the radar near zone is near the edge of the precipitation echo and no phase folding occurs. In addition, data within 3 and 4km of radar near zone are susceptible to ground objects and noiseThe radial data fluctuation is large. Therefore, in the embodiment, the difference phase standard deviation and the average value of the continuous M distance banks are calculated in sequence from 5km of the distance radar, and the continuous distance bank phi is calculated _DP Without abnormal fluctuation (i.e.. Phi.) _DP Standard deviation less than a set fluctuation threshold) and zero lag correlation coefficient>The starting distance bin of 0.75 serves as the valid starting distance bin for the current radial unfolding process. According to a large quantity phi _DP Statistical analysis of data, selecting 10 continuous distance bins for phi analysis _DP And calculating standard deviation and average value. Phi _DP The abnormal fluctuation threshold is set to 15.0, and the average value of the differential phase shifts in the range bin interval in which the standard deviation of the differential phase shifts is smaller than the threshold is taken as the radial phi _DP The initial differential phase reference value of the unfolding process.

After the initial phase in the radial direction is determined, calculation of the standard deviation and the average value of differential propagation phase shift is carried out on all the remaining distance banks in turn in the radial direction. If phi is recognized _DP And if the abnormal fluctuation exists, storing the obtained difference phase average value. And the average value is compared with the estimated phi of the corresponding distance unit _DP Comparing the difference between the parameters, and determining if phi occurs by checking the continuity between the radial distance bins _DP Folding, and performing differential phase unfolding processing on the units identified as being folded to recover the true phi _DP And (6) measuring the values.

In actual detection, due to phi _DP The parametric measurements are susceptible to clutter, atmospheric interference, and noise and error, phi in the radial direction _DP The measurement results often have fluctuations and undulations. The median filtering in the above steps has a certain effect on the burr of the smooth part, but cannot ensure the phi after quality control _DP The parameter radially holds the property of cumulative increments. K thus obtained _DP The parameters generally have negative values.

Therefore, it is necessary to proceed with _DP Radial interpolation process to ensure phi _DP Distance cumulative incremental feature of (a) and (K) _DP A non-negative characteristic of (d); it includes marking two segments that are not adjacent as phi of good data quality _DP Performing interpolation processing on the parameter estimation value region when the whole radial number is processedAfter interpolation filling is completed, the radially regenerated phi is saved _DP The difference of the parameters results.

According to the formula Ψ _DP (r)＝Φ _DP (r) + δ (r), the total differential phase shift in the radial direction comprising the forward propagating phase shift Φ _DP And a backscatter differential phase δ. The effect of δ will not be considered when the meteorological echo sounding satisfies the Rayleigh scattering condition. However, when the precipitation particles no longer satisfy the Rayleigh scattering condition, δ is no longer approximately zero, and even larger values may occur, and the influence of δ is no longer negligible. Studies have shown that for stronger rain areas, the presence of large precipitation particles will produce a significant delta effect, and the value of delta increases with increasing precipitation particles. In this case, Ψ is directly utilized _DP As phi _DP Measured value K _DP A large error occurs in the estimation. Therefore, using a dual polarization parameter Φ _DP Estimating K _DP In the parameter setting, the delta component in the differential phase measurement value must be removed first.

When the above phi is completed _DP After the quality control of the parameters, to eliminate the delta value and further eliminate the fluctuation caused by the interference, the phi obtained by the above method is needed _DP And filtering the parameters. Phi _DP Is a distance cumulant which changes with the increase of the distance, and phi can be obtained by selecting lower cut-off frequency by adopting traditional filters such as FIR, IIR and the like _DP The average trend of the data, however, when the value of δ in a plurality of consecutive distance bins is non-zero, the variation caused by δ cannot be effectively suppressed. When the filtering method of loop iteration is adopted, the problem of fluctuation of the data curve is solved, and delta can be automatically detected. Therefore, the invention adopts the FIR low-pass filter pair phi of the loop iteration _DP The measured values are filtered.

As shown in FIG. 3, phi _DP The filtering process includes: the phi after the filtering and before the filtering on each distance library is subjected to unfolding processing _DP And comparing, when the absolute value of the difference value between the two is greater than a set threshold value, performing iterative filtering of the next cycle by using the filtered data, and otherwise, performing iteration by using the data before filtering. The data for the iteration is selected asOn the premise of eliminating delta and abnormal fluctuation, the authenticity of the measured data is guaranteed to the greatest extent, the fluctuation trend of the original data is kept, and the data is prevented from being excessively processed.

Wherein the filter is designed as a FIR low-pass iterative filter based on a window function. The window function may control the response according to the terminal. The unit impulse response of the FIR low-pass filter is:

h(n)＝h _d (n)w(n)

wherein h is _d (n) is the ideal impulse response, w (n) is the selected window function;

the output sequence y (N) of the FIR filter of order N is then:

wherein, N is the order of the designed filter, N is the serial number of the discrete sequence, and the determination of the order of the filter is related to the radar distance library length and the filtering width. And if the length of the radar distance library is set as R and the filter width is determined as R, the order of the FIR low-pass iterative filter is as follows:

as the order of the filter increases within a certain range, phi in the radial direction _DP The data becomes smoother, the curve change is more stable, and phi can be effectively eliminated _DP Pulsatility of measured value of phi _DP The continuity of the measured value in the radial direction is better. But as the order continues to increase, the radial direction Φ _DP Will gradually decrease and even destroy phi during the smoothing process _DP Signal fluctuations in the measurement caused by the meteorological echo signals result in a smooth transition. Therefore, the filter in the invention can be designed with a proper order according to actual needs.

As shown in fig. 4, the above-mentioned phi is completed _DP After the data quality control step, the obtained phi is used _DP Data calculation K _DP . The invention adopts the minimum twoMethod for linear regression by multiplication and calculation of K _DP . In practical application, in order to ensure K _DP Measurement accuracy of, subduct K _DP The pulsatility of the measured values, selecting phi of M continuous distance banks _DP Measured value of K _DP Calculating; k _DP The estimation formula of (c) is:

where M is the number of consecutive distance bins selected, r _i The detection range of the ith range bin.

For calculating K _DP The number N of the distance bins of the parameter is selected according to the measured value of the reflectivity factor of the corresponding distance bin. For strong echo regions, a shorter fitting distance is selected, and for regions with weaker echoes, a longer fitting distance is selected. K of the adaptive variable distance method _DP The design of the algorithm can reduce the sum of phi _DP And the influence on the weak echo region is reduced while errors caused by data abnormal fluctuation, noise and the like are avoided, and the characteristics of the strong echo region are reserved. The invention adaptively adjusts the fitting distance according to the actually detected corresponding reflectivity factor.

Further, K _DP The measured value judgment and processing comprises the following steps:

in the radial direction to K _DP The measured values are checked, and when the measured values exceed 5 continuous distance banks and are a certain fixed value, the measured values are determined as the phi invalid in the previous pair _DP K obtained by calculating data filling step _DP When K is determined _DP The measured value is an invalid value, and K of all distance libraries in the radial direction is completed _DP In the estimation, K marked as a bad data quality interval _DP Setting as invalid value;

to K _DP The measured value is radially checked, the echo property is judged through characteristic parameters, and when the echo property is judged to be a meteorological echo, K is output _DP And if not, setting the measured value as an invalid value.

By pair of phi _DP Data quality control of data and for K _DP The estimation method is designed to process and verify radar data in real time in the whole radar detection after the data of a plurality of stations are checked. Through the K _DP Estimating K obtained by algorithm design _DP The measured value has good corresponding relation with the reflectivity factor under the weather conditions of weak echo and strong echo, shows the trend of increasing along with the increase of the reflectivity factor, and is used for measuring phi _DP Unfolding and filtering of jitter, glitches, etc., K _DP Better radial continuity and reduced negative value, and integrally embodies K _DP A trend toward growth.

The foregoing is illustrative of the preferred embodiments of this invention, and it is to be understood that the invention is not limited to the precise form disclosed herein and that various other combinations, modifications, and environments may be resorted to, falling within the scope of the concept as disclosed herein, either as described above or as apparent to those skilled in the relevant art. And that modifications and variations may be effected by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (10)

1. A differential propagation phase shift rate measurement method of a dual-polarization Doppler weather radar is characterized by comprising the following steps: the measuring method comprises the following steps:

basic data processing steps: sequentially carrying out ground clutter identification and clutter filtering processing on the original data;

Φ _DP and a data quality control step: sequentially processing data processed by basic data processing step _DP Initial radial filling,. Phi _DP Data quality classification process, phi _DP Unfolding treatment,. Phi. _DP Radial interpolation process and phi _DP Filtering;

K _DP calculating and processing steps: will pass through phi _DP Phi obtained in the data quality control step _DP The data sequentially passes through K _DP Least squares method of calculating sum K _DP The measured value is judged and processed to obtain K _DP And (4) measuring the value.

2. The method of claim 1, wherein the method comprises the steps of: the ground clutter recognition processing comprises: determining horizontal texture TDBZ of the reflectivity, radial change SPIN of the reflectivity and a clutter phase array calibration value CPA on a distance unit in a certain range as an identification fuzzy basis by analyzing different characteristics and distribution of ground clutter and weather echo, and selecting a reasonable threshold value by using a fuzzy logic method for identification;

the clutter filtering process includes: and marking the distance unit containing the ground clutter, and performing ground clutter suppression processing on the marked distance unit by setting clutter filters with different suppression degrees of a time domain and a frequency domain.

3. The method of claim 1, wherein the method comprises the steps of: the phi _DP The radial initial fill includes: for phi _DP Filling the initial effective value of the data in the radial direction by filling the ineffective cells with the adjacent effective distance library, and filling phi _DP Data quality classification processing or _DP Unfolding treatment or phi _DP Radial interpolation process or phi _DP After the filtering process, the invalid data or the data which does not meet the requirement are required to be subjected to effective value filling so as to update the processed data.

4. The method of claim 1, wherein the method comprises the steps of: the phi _DP The data quality classification processing comprises the following steps:

sequentially calculating phi of each distance unit along radial direction _DP Standard deviation and ZDr standard deviation, and setting corresponding threshold values respectively;

meanwhile, the data quality is judged by combining the zero lag correlation coefficient of the current distance library and the method that the signal-to-noise ratio distribution meets the threshold judgment condition, and when the continuous M distance libraries meet the threshold judgment, the phi marked as good data quality _DP Data, when not meeting the threshold judgment, marked as difference data phi _DP The left start unit of the data interval.

5. The method of claim 4, wherein the method comprises the steps of: passing through the phi _DP After the data quality classification processing, it is necessary to determine whether the data marked as the difference data interval is large precipitation particles, which specifically includes the following contents:

calculating the average value of the reflectivity factors of the meteorological echo information in the region, and when the average value is more than or equal to a set value, comparing the average value of the zero lag correlation coefficient in the region with the standard deviation phi _DP Calculating the average value and standard deviation, judging whether the identification result of the interval is determined as a difference data interval or not by checking whether the identification result meets a threshold value or not, finally filling the effective value of the interval determined as the difference data, and updating the radial phi _DP Data is obtained.

6. The method of claim 1, wherein the method comprises the steps of: the phi _DP The unfolding process includes:

according to phi _DP Determining a range of starting distances and phi of effective precipitation echoes in the radial direction _DP A starting value;

sequentially calculating phi of M distance libraries _DP Standard deviation and mean value, and a continuous distance library phi _DP The initial distance library with the standard deviation smaller than a set fluctuation threshold and the zero lag correlation coefficient larger than a set threshold is used as an effective initial distance library for the current radial unfolding processing;

selecting a plurality of successive range bins for phi _DP Calculation of the standard deviation and mean value,. Phi _DP Phi of distance library interval with standard deviation less than threshold _DP Average value as the radial direction phi _DP Unfolding the processed initial differential phase reference values;

sequentially carrying out phi on all the remaining distance bins along the radial direction _DP Calculating standard deviation and average value, if phi is recognized _DP If there is abnormal fluctuation, the obtained phi _DP The average value is stored and the planar soul and the estimated phi of the corresponding distance unit _DP Comparing the parameters to obtain a difference;

judging whether phi occurs or not by checking continuity between radial distance bins _DP Folding, and performing differential phase unfolding processing on the units identified as being folded to recover the true phi _DP And (4) measuring the value.

7. The method of claim 1, wherein the method comprises the steps of: the phi _DP The radial interpolation process includes: phi marking two segments that are not adjacent as good data quality _DP Interpolation processing is carried out on the parameter estimation value area, and after interpolation filling is finished on the whole radial data, phi generated again in the radial direction is stored _DP And obtaining the difference result of the parameters.

8. The method of claim 1, wherein the method comprises the steps of: the phi _DP The filtering process includes: FIR low-pass filter pair phi adopting loop iteration _DP Filtering the measured value, specifically including the following:

filtering each distance library and unfolding processed phi before filtering _DP And comparing, when the absolute value of the difference value between the two is greater than a set threshold value, performing iterative filtering of the next cycle by using the filtered data, and otherwise, performing iteration by using the data before filtering.

9. The method of claim 1, wherein the method comprises the steps of: said K _DP The least square method calculation comprises: selecting phi of M continuous distance libraries _DP Measured value by K _DP And phi _DP Calculating K by the estimation formula _DP For calculating K _DP The number N of the distance bins of the parameter is selected according to the measured value of the reflectivity factor of the corresponding distance bin.

10. The method of claim 1, wherein the method comprises the steps of: said K _DP The measurement value judgment and processing comprises the following steps:

in the radial direction to K _DP The measured values are checked, and when the measured values exceed a plurality of continuous distance bins to a certain fixed value, K is determined _DP The measured value is an invalid value, and K of all distance libraries in the radial direction is completed _DP In the estimation, K marked as a bad data quality interval is used _DP Setting the value as an invalid value;

to K _DP The measured value is subjected to radial inspection, the echo property is judged through the characteristic parameters, and when the echo property is judged to be a meteorological echo, K is output _DP And if not, setting the measured value as an invalid value.

Images