X-band dual-polarized phased array rain radar system and detection method
Technical Field
The invention relates to the technical field of meteorological radars, in particular to an X-band dual-polarized phased array rain radar system and a detection method.
Background
The dual-polarized phased array radar is a radar system combining a polarization measurement technology and a phased array technology, can perform functions of a traditional phased array radar, such as rapid and flexible beam pointing and tracking of a plurality of targets, and can provide more information about characteristics of the targets through measurement of polarization information. The dual-polarized phased array radar can acquire the characteristics of the reflected signals of the target in different polarization states, so that the target can be identified and classified more accurately, and the dual-polarized phased array radar is particularly very useful for water vapor particle analysis in meteorological monitoring.
The Chinese patent with the publication number of CN113933844A discloses a phased array multiband integrated receiving and transmitting radar and a radar detection method, which comprises a wave control unit and a phased array antenna connected with the wave control unit, wherein the phased array antenna comprises a first receiving and transmitting assembly, a second receiving and transmitting assembly and a third receiving assembly, the first receiving and transmitting assembly is used for transmitting a plurality of first beams to a plurality of different directions of the sky based on elevation angle azimuth information and receiving first information corresponding to the first beams, the second receiving and transmitting assembly is used for transmitting a plurality of second beams to a plurality of different directions of the sky based on the elevation angle azimuth information and receiving second information corresponding to the second beams, the third receiving assembly is used for receiving third information corresponding to the third beams, and the wave control unit is used for determining elevation angle azimuth information and also used for determining detection information based on the first information, the second information and the third information. The method has the advantages that the transmitting direction of the detection radar is not required to be regulated, the detection data of a plurality of satellites can be obtained simultaneously, the cloud and rain distribution information in a certain range is detected, and the complexity of operation is reduced.
The Chinese patent with publication number of CN111983617A discloses a dual-polarization phased array weather radar, which is formed by alternately configuring a horizontal polarization antenna and a vertical polarization antenna to form a waveguide crack planar antenna array, a DTRU subsystem formed by a DTRU module and a heat dissipation fan, a digital beam forming/signal processing subsystem formed by a signal processing and light transmission and timing synchronization module, a servo subsystem, a frequency source and distribution network subsystem and a monitoring subsystem, and can realize the requirements of multiple working modes of dual-emission, single-emission dual-emission single-emission of a whole system. The system can realize pitching dimension phase scanning, single/multi-beam receiving work requirements, quickly detect weather target information, effectively monitor occurrence and development of disaster weather such as heavy rain, hail, tornado and the like, simultaneously realize double polarization with good performance of quantitatively measuring echo intensity, quantitatively estimate large-scale precipitation, realize fixed point, quantitative and timed precipitation prediction and provide reliable and scientific detection data for weather guarantee.
However, the above prior art still has the following problems:
(1) The high-precision precipitation estimation depends on the accuracy of radar dual-polarized quantity measurement, and requires that equivalent radiation power, beam direction, beam width and side lobe level of dual-polarized beams have higher consistency;
(2) Particle characteristics of different phases are not considered, and influence of the different particle phases on precipitation estimation cannot be effectively eliminated;
(3) The ground area matching is not considered, and the rain monitoring effect on the fine, near-ground and accurate surfaces is poor.
Disclosure of Invention
In order to achieve the purpose of the invention, the invention realizes an X-band dual-polarized phased array rain radar system by the following technical scheme, which comprises hardware equipment, a processing module and an accessory matching;
the hardware equipment comprises an antenna array subsystem and a radar base subsystem;
the processing module comprises a polarization channel consistency calibration sub-module, a particle phase state identification sub-module and a rainfall estimation sub-module;
the auxiliary accessories comprise power supply equipment, a generator, lightning protection facilities, an optical fiber network and a data storage.
The radar base subsystem comprises a beam synthesis control unit, a system monitoring unit, a signal processing unit, a main power supply and a servo control unit.
Further, the dual-polarized array antenna adopts a planar microstrip patch array antenna;
the array surface of the dual-polarized array antenna adopts a rectangular caliber, the pitching and the azimuth are separable, the pitching direction is 64 rows of dual-polarized microstrip patch linear arrays, each linear array is provided with 32 radiating units in the azimuth direction, and the feed interfaces of all subarrays of the antenna are positioned on the back surface of the radiating surface.
Further, the radiating units are rectangular radiating patches, horizontal and vertical coupling gaps and feed transmission lines from the upper layer to the lower layer respectively.
The algorithm of the polarization channel consistency calibration submodule is as follows:
Equivalent conversion of the aperture field and the far field is adopted, after the amplitude and phase measurement of the aperture field, far field conversion is carried out to obtain a far field pattern,
And initializing a particle swarm population by utilizing l ogi st ics mapping based on the objective function, and obtaining an optimized far-field pattern through a particle swarm algorithm.
The classification algorithm of the particle phase state recognition sub-module comprises melting layer recognition, data confidence calculation and membership function calculation, and specifically comprises the following steps:
(1) Carrying out melting layer identification based on a quasi-vertical profile technology, and acquiring melting layer identification information and precipitation type identification information;
(2) Establishing a membership function by combining the scattering characteristics of the condensate with the identification information of the melting layer and the identification information of the precipitation type;
(3) Optimizing a membership function by using a K-means cluster analysis algorithm;
(4) And integrating the information of each polarization parameter by using a fuzzy logic method, and carrying out various particle phase state identification and classification.
The rainfall estimation submodule comprises a falling area correction model and a random forest model, and the random forest model training steps are as follows:
(1) Acquiring automatic station rainfall data, altitude and longitude and latitude of radar grid points, and radar polarization parameters including radar degree reflectivity factors, differential reflectivity factors and differential propagation phase shifts;
(2) Taking the automatic station rainfall corresponding to the radar lattice points as an output value and other data as an input value, and training an initial model;
(3) And constraining the grid point data of the rainfall product according to the probability density distribution of the intensity grid point data to generate a random forest model.
The invention also comprises a detection method of the X-band dual-polarized phased array rain-measuring radar, which comprises the following steps:
S1, radar parameter setting, wherein a user terminal sends parameters to a system monitoring unit through a network, and the system monitoring unit respectively sends the parameters to a signal processing unit and a beam synthesis control unit;
s2, the beam synthesis control unit calculates beam direction data and sends the beam direction data to the digital intermediate frequency processing unit and the radio frequency receiving and transmitting unit;
S3, the dual-polarized array antenna completes the transmission and the reception of signals;
And S4, the processing module completes data calculation through the channel consistency calibration sub-module, the particle phase state identification sub-module and the rainfall estimation sub-module, and transmits the data to a remote user terminal through a network.
The beneficial effects of the invention are as follows:
1. Setting a polarization channel consistency calibration submodule, giving an initial value by establishing a mapping function of far-field pattern distribution difference and channel amplitude and combining Logistic mapping, and obtaining an optimized far-field pattern based on a particle swarm algorithm to more accurately approximate a target pattern;
2. Setting a particle phase state identification submodule, and establishing a recognition algorithm based on fuzzy logic by using the characteristic expression of particles in different phases on different radar polarization amounts so as to realize automatic recognition of different particle phases such as light rain, medium rain, heavy rain, big drop, hail, snow, wet snow, ice crystal, ground clutter and the like, effectively eliminate the influence of different particle phases on precipitation estimation, and improve the accuracy of precipitation estimation;
3. setting a rainfall estimation submodule, correcting errors of the rainfall on the ground through a falling area correction model, replacing a static rainfall relation by using a random forest model, and carrying out constraint processing on lattice point data of a rainfall product, so that the correlation coefficient is improved, the root mean square error is reduced, and the measurement accuracy is obviously improved;
4. the planar microstrip patch array antenna is designed through interval arrangement, and has the characteristics of small area, high phase center consistency and the like, and high consistency, low side lobe and high cross polarization isolation of dual polarized beams are ensured.
Drawings
FIG. 1 is a schematic diagram of an X-band dual-polarized phased array rain radar system of the present invention;
Detailed Description
The present invention will be further described in detail with reference to the following examples, which are only for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
Example 1
According to the embodiment shown in fig. 1, an X-band dual-polarized phased array rain radar system is provided, which comprises hardware equipment, a processing module and an accessory kit;
the hardware equipment comprises an antenna array subsystem and a radar base subsystem;
the processing module comprises a polarization channel consistency calibration sub-module, a particle phase state identification sub-module and a rainfall estimation sub-module;
the auxiliary accessories comprise power supply equipment, a generator, lightning protection facilities, an optical fiber network and a data storage.
The radar base subsystem comprises a beam synthesis control unit, a system monitoring unit, a signal processing unit, a main power supply and a servo control unit.
More specifically, the dual polarized array antenna adopts a planar microstrip patch array antenna;
the array surface of the dual-polarized array antenna adopts a rectangular caliber, the pitching and the azimuth are separable, the pitching direction is 64 rows of dual-polarized microstrip patch linear arrays, each linear array is provided with 32 radiating units in the azimuth direction, and the feed interfaces of all subarrays of the antenna are positioned on the back surface of the radiating surface.
More specifically, the radiating element is a rectangular radiating patch, horizontal and vertical coupling slits, and a feed transmission line from the upper layer to the lower layer, respectively.
More specifically, the radiating element spacing satisfies:
where λ is the wavelength of the antenna operating frequency, θ is the scan angle, and Δθ is half the beamwidth.
More specifically, the algorithm of the polarization channel consistency calibration sub-module is as follows:
And performing far field conversion to obtain a far field pattern after measuring the amplitude and phase of the aperture field by adopting equivalent conversion of the aperture field and the far field, wherein the far field pattern of each polarization is expressed as:
Wherein, u=kdcos(θ),θ=0~180°,The number of radar units is N, the distance is d, and the measured amplitude and phase distribution of each channel are I n and alpha n respectively;
The given directional diagram is f (theta), and the amplitude I n and the phase alpha n of the array elements are changed to form the same directional diagram as f (theta) under the condition that the number N of the array elements, the unit spacing d and the wavelength lambda are kept unchanged;
sampling a given direction diagram, setting the number of sampling points as M, and constructing an objective function:
wherein S max = max (S (I, α, θ));
Based on the objective function, initializing a particle swarm population by using logistics mapping, and obtaining an optimized far-field pattern through a particle swarm algorithm. The convergence efficiency of the algorithm can be effectively increased by combining the vectorized objective function and utilizing the particle swarm algorithm, the optimized directional diagram can be enabled to approach the objective directional diagram more accurately, and the result can reach expectations better by increasing constraint conditions.
More specifically, the classification algorithm of the particle phase state recognition sub-module comprises melting layer recognition, data confidence calculation and membership function calculation, and specifically comprises the following steps:
(1) Carrying out melting layer identification based on a quasi-vertical profile technology, and acquiring melting layer identification information and precipitation type identification information;
(2) Establishing a membership function by combining the scattering characteristics of the condensate with the identification information of the melting layer and the identification information of the precipitation type;
(3) Optimizing a membership function by using a K-means cluster analysis algorithm;
(4) And integrating the information of each polarization parameter by using a fuzzy logic method, and carrying out various particle phase state identification and classification.
Before identification, quality control such as noise removal, attenuation correction, KDP segmentation least square fitting and the like is needed to be carried out on the radar original observed data so as to improve the reliability of the dual-polarized observed data.
The horizontal resolution of the X-band dual-polarized phased array rain measuring radar is 30m, the X-band dual-polarized phased array rain measuring radar has the characteristics of high space-time resolution and easiness in network distribution, the number of elevation angles of the X-band dual-polarized phased array rain measuring radar is unequal in 35-40 layers, denser observation data can be provided in the vertical direction, and the X-band dual-polarized phased array rain measuring radar has a greater advantage in particle phase identification.
More specifically, the rainfall estimation submodule comprises a landing zone correction model and a random forest model, and the random forest model training steps are as follows:
(1) Acquiring automatic station rainfall data, altitude and longitude and latitude of radar grid points, and radar polarization parameters including radar degree reflectivity factors, differential reflectivity factors and differential propagation phase shifts;
(2) Taking the automatic station rainfall corresponding to the radar lattice points as an output value and other data as an input value, and training an initial model;
(3) And constraining the grid point data of the rainfall product according to the probability density distribution of the intensity grid point data to generate a random forest model.
More specifically, the drop zone correction model calculation steps are as follows:
(1) The radar minimum elevation angle is usually within the observation range of the boundary layer, the movement of the raindrops can be regarded as the movement of the landing points in the boundary layer, and the movement equation of the raindrops within the boundary layer considers the gravity, the buoyancy and the drag force (resistance) of the raindrops, which can be expressed as:
f(NRe)=a1NRe b1+a2NRe b2
Wherein, Indicating the weight force to which the raindrops are subjected,And (3) withRepresents the buoyancy and resistance experienced by the raindrops, S represents the area through which the water drops pass,Representing the falling speed of the raindrops,Represents the velocity of ambient air, C d represents the drag coefficient, N Re represents the reynolds number, r represents the water drop radius, ρ a represents the air density, ρ w represents the rain drop density, η represents the air viscosity, and in addition the coefficients a 1=0.17,a2=1.0×10-6, b1=0.632, b2=2.25;
Combining the above formulas to obtain:
Wherein, Representing the falling speed of the raindrops,Representing the velocity of ambient air, N Re representing the reynolds number, r representing the water droplet radius, ρ a representing the air density, ρ w representing the rain droplet density, η representing the air viscosity;
(2) Calculating the raindrop falling time according to the raindrop falling speed and the height of the radar grid point from the ground;
(3) And combining the automatic station interpolation horizontal wind field, calculating to obtain the offset after the raindrops drop, and correcting the error of the radar estimated rainfall and the landing rainfall.
Example two
An X-band dual-polarized phased array rain detection radar detection method comprises the following steps:
S1, radar parameter setting, wherein a user terminal sends parameters to a system monitoring unit through a network, and the system monitoring unit respectively sends the parameters to a signal processing unit and a beam synthesis control unit;
s2, the beam synthesis control unit calculates beam direction data and sends the beam direction data to the digital intermediate frequency processing unit and the radio frequency receiving and transmitting unit;
S3, the dual-polarized array antenna completes the transmission and the reception of signals;
And S4, the processing module completes data calculation through the channel consistency calibration sub-module, the particle phase state identification sub-module and the rainfall estimation sub-module, and transmits the data to a remote user terminal through a network.
The foregoing has shown and described the basic principles, principal features and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, and that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and various changes and modifications may be made without departing from the spirit and scope of the invention, which is defined in the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.