A phased array weather radar beam equalization method based on a preset particle swarm algorithm

Abstract

The application discloses a phased array weather radar beam equalization method based on a preset particle swarm algorithm, and the method comprises the following steps: establishing upper and lower limit constraint curves of an equalization beam optimization target according to index requirements; designing amplitude-phase results of Taylor weighting and scanning phase weighting according to the index requirements, and generating an initial particle swarm according to the amplitude-phase results; establishing an evaluation function according to array characteristics; calculating an array radiation beam through amplitude and phase of each channel of the array, scoring the particle swarm according to a matching degree of the radiation beam and the optimization target, selecting a best particle in the swarm, and setting the preset particle swarm as a best particle in each generation; all particle swarms participate in iteration, the newly generated particle swarm is scored by using the evaluation function constantly, and a best particle in the swarm and the best particle in each generation are selected constantly; the iteration movement process is repeated until a specified iteration number is reached or the best particle in the swarm completely satisfies the optimization target curve.

Description

Technical Field

[0001] This invention relates to antenna and microwave technology, and in particular to a method for beam equalization of phased array weather radar based on a pre-set particle swarm optimization algorithm. Background Technology

[0002] In the field of weather radar, radar systems have gradually transitioned from traditional mechanically scanned radar to phased array radar. Phased array radar can increase the range of weather detection, enhance signal anti-interference capabilities, and provide flexible beam pointing. It can also perform rapid scanning without inertia, controlling the full airspace scan time to within one minute. It has excellent monitoring capabilities for rapidly changing weather processes such as tornadoes, hail, and wind shear. This makes phased array radar widely favored in the meteorological field.

[0003] However, according to the weather radar equation:

[0004]

[0005] In the formula: P r To receive echo power; P t λ is the peak power; G is the antenna gain; λ is the radar wavelength; h is the pulse length; R is the distance between the radar and the target; θ is the horizontal beamwidth; φ is the vertical beamwidth; |K| 2 is the dielectric constant term; Z is the radar reflectivity factor of the meteorological target.

[0006] It is evident that meteorological detection is sensitive to the beamwidth of the radar beam. During scanning, the radar beam must be kept absolutely stable to minimize correction processes and detection errors. However, the beamwidth and beam gain of phased array radars change with the scanning angle, which imposes certain limitations on their application in meteorological detection.

[0007] Patent application number 201711226971.8 discloses a small-aperture conical horn with beam equalization. This design compresses the aperture of the radiating section in the E-plane direction and uses perturbation to excite the TM. 11 The design achieves beam equalization in both the E-plane and H-plane. However, this design only achieves beam equalization in the E-plane and H-plane for a single antenna, which falls short of the requirements for phased array weather radar applications.

[0008] In the field of phased array radar, there is still no good solution to the beam equalization problem during the scanning process, so as to ensure that the radar beamwidth and gain remain consistent during the scanning process. Summary of the Invention

[0009] To address the problems existing in the prior art, this invention provides a solution to the beam equalization problem in phased array radar during the scanning process. It achieves beam equalization by optimizing the radar beam shape during scanning using a pre-set particle swarm optimization algorithm, generating the optimal amplitude and phase distribution at each scanning angle. This ensures that the phased array radar beam is equalized at all scanning angles, meaning that the radar beamwidth and gain remain consistent. This technology solves the problem of phased array radar beam widening and gain decreasing with scanning angle, meeting the application requirements of meteorological radar and enhancing the advantages of phased array radar in meteorological detection. This is a beam equalization method for phased array meteorological radar based on a pre-set particle swarm optimization algorithm.

[0010] The objective of this invention is achieved through the following technical solutions.

[0011] A method for beam equalization of phased array weather radar based on a pre-set particle swarm optimization algorithm includes the following steps:

[0012] Step 1: Establish upper and lower limit constraint curves for equalized beam optimization targets based on the indicator requirements;

[0013] Step 2: Design the amplitude and phase results using Taylor weighting and scanning phase weighting according to the performance requirements, and generate the initial particle swarm based on the amplitude and phase results. in …is a Taylor-weighted amplitude distribution for each channel. It is a scan phase weighted with θ pointing, S n It is the nth particle, and N is the number of particles in the swarm. That is, the pre-set particle swarm corresponds to the non-equilibrium radiation beam, which is closer to the optimization target;

[0014] Step 3: Establish an evaluation function F = lg(∑max(LP,0) + ∑max(PH,0)) based on the array characteristics, where H and L are the upper and lower limit curves of the optimization target, respectively, and P is the array radiation beam result curve calculated for the corresponding particles;

[0015] Step 4: Calculate the array radiation beam P using the amplitude and phase of each channel, and score the particle swarm based on the degree of matching between the radiation beam and the optimized target, selecting the best particle in the swarm. Set the preset particle swarm to the best particles of all time.

[0016] Step 5: Involve the entire particle swarm in the iteration, and the iterative motion vector V includes the best motion vectors from each iteration. Group optimal motion vector and inertial motion vector V R The process consists of three parts: continuously using evaluation functions to score newly generated particle populations, and continuously selecting the best particle in the population and the best particle in each generation.

[0017] Step 6: Repeat the iterative motion process until the specified number of iterations is reached or the optimal swarm of particles fully satisfies the optimization target curve;

[0018] Step 7: Verify whether the radiation beam result corresponding to the optimal particle of the swarm, i.e. the optimal channel amplitude phase, meets the beam equalization requirements and store the optimal channel amplitude phase corresponding to the scanning angle;

[0019] Step 8: Change the scanning angle indicator and repeat the above steps.

[0020] In step one, the upper limit curve of the optimization target defines the upper limit profile of the beam, the lower limit curve of the optimization target defines the lower limit profile of the beam, and the beamwidth is also defined by the upper and lower limit curves of the optimization target. The upper and lower limit constraint curves of the optimization target constrain the beamwidth and also constrain the sidelobes, so that the optimization target achieves high sidelobe suppression while ensuring that the beamwidth remains unchanged.

[0021] In step two, each preset particle is a set of Taylor-weighted amplitude distribution and scan-phase-weighted phase distribution required by the index; the particle population size N is set by the accuracy and speed required by the algorithm.

[0022] In step three, the number of calculation points for the upper and lower limit curves H and L, as well as the radiation beam P, is the same, and they are mapped one-to-one with the spatial angles.

[0023] The iterative motion in step five is characterized by: V R =k3V - Where k1, k2, k3 are iteration factors, R(i) is a uniformly distributed random value in the interval (0,1), and V - It is the motion vector from the previous iteration. The values ​​of the iteration factors k1, k2, and k3 are constants and change according to the required iteration speed.

[0024] Step six, which continues until the specified number of iterations is reached or the optimal particle in the population fully satisfies the optimization target curve, does not yield a unique optimization result. This step can be performed multiple times to obtain multiple different optimization results.

[0025] In step seven, it is necessary to verify whether the radiation beam result generated by the optimal particle of the swarm, i.e. the optimal channel amplitude phase, meets the beam equalization requirements. The iteration result must meet the amplitude phase quantization requirements of each radar channel and be applied to the actual radar array. The beam result after amplitude phase quantization of each channel is verified by electromagnetic simulation software.

[0026] In step eight, the scanning angle parameter is changed: the scanning angle setting must satisfy all scanning spatial domains, and the change rule is θ. i =iθ p i∈(0,…,θ)m / θ p -1), where θ i It is the beam direction, θ p It is the beamwidth, θ m The indicator requires the maximum beam pointing.

[0027] Compared with the existing technology, the advantages of the present invention are: 1. The present technology adopts the pre-set particle swarm method, which provides a better solution before iteration, greatly improving the optimization speed of the algorithm.

[0028] 2. This technology establishes equalized beam optimization target upper and lower limit curves for each scanning angle, ensuring that the beamwidth and gain of the array remain unchanged during the scanning process, thus meeting the application requirements of weather radar.

[0029] 3. During the scanning process, this technology also incorporates a high sidelobe suppression optimization target, which maintains high sidelobe suppression while ensuring that the beamwidth and gain remain unchanged during the array scanning process. Attached Figure Description

[0030] Figure 1 This is a flowchart of the algorithm in this invention.

[0031] Figure 2 The present invention provides the target curve and optimization results for scanning 30 degrees in an embodiment.

[0032] Figure 3 The results of beam scanning in the embodiments of the present invention are only Taylor-weighted.

[0033] Figure 4 The beam scanning results in the embodiments of the present invention are optimized by the preset particle swarm optimization technique.

[0034] Figure 5 This is the amplitude and phase optimization result of each channel in the beam equalization of the embodiment of the present invention when scanning is not performed.

[0035] Figure 6 The results of beam equalization and amplitude / phase optimization for each channel in an embodiment of the present invention are shown.

[0036] Figure 7 The results of beam equalization and amplitude / phase optimization for each channel in an embodiment of the present invention are shown.

[0037] Figure 8 The results of beam equalization and amplitude / phase optimization for each channel in an embodiment of the present invention are shown. Detailed Implementation

[0038] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0039] A method for beam equalization of phased array weather radar based on a pre-set particle swarm optimization algorithm includes the following steps:

[0040] Step 1: Establish upper and lower limit constraint curves for equalized beam optimization targets based on the indicator requirements.

[0041] Step 2: Design the amplitude and phase results using Taylor weighting and scanning phase weighting according to the performance requirements, and generate the initial particle swarm based on the amplitude and phase results. in …is a Taylor-weighted amplitude distribution for each channel. It is a scan phase weighted with θ pointing, S n It is the nth particle, and N is the number of particles in the swarm. That is, the pre-set particle swarm corresponds to the non-equilibrium radiation beam, which is closer to the optimization target.

[0042] Step 3: Establish an evaluation function F = lg(∑max(LP,0) + ∑max(PH,0)) based on the array characteristics, where H and L are the upper and lower limit curves of the optimization target, respectively, and P is the array radiation beam result curve calculated for the corresponding particles.

[0043] Step 4: Calculate the array radiation beam P using the amplitude and phase of each channel, and score the particle swarm based on the degree of matching between the radiation beam and the optimized target, selecting the best particle in the swarm. Set the preset particle swarm to the best particles of all time.

[0044] Step 5: Involve the entire particle swarm in the iteration, and the iterative motion vector V is derived from the best motion vector in each iteration. Group optimal motion vector and inertial motion vector V R It consists of three parts. The evaluation function is continuously used to score the newly generated particle swarm, and the best particle in the swarm and the best particle in each generation are continuously selected.

[0045] Step 6: Repeat the iterative motion process until the specified number of iterations is reached or the optimal particle population fully satisfies the optimization target curve.

[0046] Step 7: Verify whether the radiation beam result corresponding to the optimal particle of the swarm, i.e. the optimal channel amplitude phase, meets the beam equalization requirements and store the optimal channel amplitude phase corresponding to the scanning angle.

[0047] Step 8: Change the scanning angle indicator and repeat the above steps.

[0048] In step one, the upper and lower limit constraint curves of the equalization beam optimization target are established: the upper limit curve of the optimization target defines the upper limit profile of the beam, the lower limit curve of the optimization target defines the lower limit profile of the beam, and the beamwidth is also defined by the upper and lower limit curves of the optimization target.

[0049] In step one, the upper and lower limit constraint curves for the equalization beam optimization target are established. These curves can constrain not only the beamwidth but also the sidelobes, enabling the optimization target to achieve high sidelobe suppression while maintaining a constant beamwidth.

[0050] Step two generates a pre-defined particle swarm: each pre-defined particle is a set of Taylor-weighted amplitude distributions and scan-phase-weighted phase distributions required by the specifications. Using a pre-defined particle swarm can speed up the algorithm optimization process and allow the optimal result to converge as quickly as possible.

[0051] In step two, a pre-defined particle swarm is generated: the number of particles N is set by the required precision and speed of the algorithm.

[0052] In step three, an evaluation function is established: the number of calculation points for the upper and lower limit curves H and L, as well as the radiation beam P, are the same, and they are mapped one-to-one with the spatial angles.

[0053] Iterative motion in step five: V R =k3V - Where k1, k2, k3 are iteration factors, R(i) is a uniformly distributed random value in the interval (0,1), and V - It is the motion vector from the previous iteration.

[0054] The iteration factors k1, k2, and k3 are constants and can be changed according to the required iteration speed.

[0055] Step six, until the specified number of iterations is reached or the optimal particle in the population fully satisfies the optimization objective curve: the optimization result is not unique, this step can be performed multiple times to obtain multiple different optimization results.

[0056] Step seven verifies whether the radiation beam generated by the optimal particle swarm, i.e., the optimal channel amplitude phase, meets the beam equalization requirements: the iteration results must meet the amplitude phase quantization requirements of each radar channel and be applicable to actual radar arrays. The beam results after amplitude phase quantization of each channel can be verified using electromagnetic simulation software. Therefore, step six can be performed multiple times to verify and select the best optimization result from the optimization results in step six.

[0057] Step eight involves changing the scan angle parameters: the scan angle setting must satisfy all scan spatial domains, and the rule for changing it is θ. i =iθ p i∈(0,…,θ) m / θ p -1), where θ i It is the beam direction, θ p It is the beamwidth, θ mThe indicator requires the maximum beam pointing.

[0058] All these steps can be applied not only to 1D linear arrays but also to 2D area arrays. In application, it is only necessary to perform all the steps described in claim 1 in each of the two dimensions.

[0059] Example

[0060] like Figure 1 As shown, a beam equalization method for phased array weather radar based on a pre-set particle swarm optimization algorithm includes the following steps. This embodiment takes a linear array composed of 160 elements as an example, with the following specifications: scanning angle of 30 degrees, sidelobe suppression of 35dB, and beamwidth maintained at 1 degree.

[0061] Step 1: Establish the upper and lower limit constraint curves of the optimization target corresponding to the scanning angle;

[0062] Step 2: Based on the sidelobe suppression requirements and the scanning angle to be optimized, establish a preset particle swarm optimization algorithm with 35dB Taylor amplitude weighting and corresponding scanning angle phase weighting. in …is a Taylor-weighted amplitude distribution for each channel. It is a scan phase weighted with θ pointing, S n It is the nth particle, and N is the number of particles in the swarm.

[0063] Step 3: Establish an evaluation function F = lg(∑max(LP,0) + vmax(PH,0)) for 160 units, where H and L are the upper and lower limit curves of the optimization target, respectively, and P is the curve of the array radiation beam result obtained by the particle calculation.

[0064] Step 4: Calculate the array radiation beam P using the amplitude and phase of each channel, and score the particle swarm based on the degree of matching between the radiation beam and the optimized target, selecting the best particle in the swarm. Set the preset particle swarm to the best particles of all time.

[0065] Step 5: Involve the entire particle swarm in the iteration, and the iterative motion vector V is derived from the best motion vector in each iteration. Group optimal motion vector and inertial motion vector V R It consists of three parts. The evaluation function is continuously used to score the newly generated particle swarm, and the best particle in the swarm and the best particle in each generation are continuously selected.

[0066] Step Six: Repeat the iterative process until the specified number of iterations is reached or the optimal swarm of particles fully satisfies the optimization target curve. The optimization results and optimization target curve at a scanning angle of 30 degrees are shown below. Figure 2 As shown.

[0067] Step 7: Verify whether the radiation beam result corresponding to the optimal particle of the swarm, i.e. the optimal channel amplitude phase, meets the beam equalization requirements and store the optimal channel amplitude phase corresponding to the scanning angle.

[0068] Step 8: Change the scanning angle and repeat steps 1 through 7.

[0069] In this embodiment, V R =k3V - Where k1, k2, and k3 are iteration factors, and k1 = 2, k2 = 2, and k3 = 0.4, R(i) is a uniformly distributed random value in the interval (0,1), and V - It is the motion vector from the previous iteration.

[0070] Figure 3 The image shows the beamwidth under Taylor weighting only. At a scan angle of 0 degrees, the beamwidth is 0.84 degrees and the gain is 24.2 dBi; at 10 degrees, the beamwidth is 0.85 degrees and the gain is 24.1 dBi; at 20 degrees, the beamwidth is 0.89 degrees and the gain is 23.7 dBi; and at 30 degrees, the beamwidth is 0.97 degrees and the gain is 23.2 dBi. It can be seen that as the scan angle increases, the beamwidth widens and the gain decreases. Beamwidth variations during scanning cannot meet the application requirements of weather radar.

[0071] Figure 4 The scanning beam results of a phased array weather radar using beam equalization technology based on a pre-set particle swarm optimization algorithm are shown. At a scanning angle of 0 degrees, the beamwidth is 1 degree and the gain is 23.3 dBi. At this point, the amplitude and phase distribution of each channel changes from... Figure 5 As shown; when the scanning angle is 10 degrees, the beamwidth is 1 degree, and the gain is 23.3 dBi. At this time, the amplitude and phase distribution of each channel changes from... Figure 6 As shown; when the scanning angle is 20 degrees, the beamwidth is 1 degree, and the gain is 23.1 dBi. At this time, the amplitude and phase distribution of each channel changes from... Figure 7 As shown; when the scanning angle is 30 degrees, the beamwidth is 1 degree, and the gain is 23.0 dBi. At this time, the amplitude and phase distribution of each channel changes from... Figure 8 As shown, the scanning beam optimized by this invention maintains a beamwidth of 1 degree during the scanning process, with a gain change of only 0.3dB. Furthermore, the sidelobe suppression remains at 35dB during the scanning process, ensuring the radar's high stability, high effectiveness, and high anti-interference capability, fully meeting the application requirements of meteorological radar.

Claims

1. A phased array weather radar beam equalization method based on a preset particle swarm algorithm, characterized in that, Includes the following steps: Step 1: Establish upper and lower limit constraint curves for equalized beam optimization targets based on the indicator requirements; Step 2: Design the amplitude and phase results using Taylor weighting and scanning phase weighting according to the performance requirements, and generate the initial particle swarm based on the amplitude and phase results. in It is a Taylor-weighted amplitude distribution for each channel. It is a scan phase weighted with θ pointing, S n It is the nth particle, and N is the number of particles in the swarm. That is, the pre-set particle swarm corresponds to the non-equilibrium radiation beam, which is closer to the optimization target; Step 3: Establish an evaluation function F = lg(∑max(LP,0) + ∑max(PH,0)) based on the array characteristics, where H and L are the upper and lower limit curves of the optimization target, respectively, and P is the array radiation beam result curve calculated for the corresponding particles; Step four: calculate the array radiation beam P through the amplitude and phase of each channel of the array, score the particle swarm according to the matching degree of the radiation beam and the optimization target, and select the best particle in the swarm Set the preset particle swarm as the best particle of each generation Step 5: Involve the entire particle swarm in the iteration, and the iterative motion vector V includes the best motion vectors from each iteration. Group optimal motion vector and inertial motion vector V R The process consists of three parts: continuously using evaluation functions to score newly generated particle populations, and continuously selecting the best particle in the population and the best particle in each generation. Step 6: Repeat the iterative motion process until the specified number of iterations is reached or the optimal swarm of particles fully satisfies the optimization target curve; Step 7: Verify whether the radiation beam result corresponding to the optimal particle of the swarm, i.e. the optimal channel amplitude phase, meets the beam equalization requirements and store the optimal channel amplitude phase corresponding to the scanning angle; Step 8: Change the scanning angle indicator and repeat the above steps.

2. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that In step one, the upper limit curve of the optimization target defines the upper limit profile of the beam, the lower limit curve of the optimization target defines the lower limit profile of the beam, and the beamwidth is also defined by the upper and lower limit curves of the optimization target. The upper and lower limit constraint curves of the optimization target constrain the beamwidth and also constrain the sidelobes, so that the optimization target achieves high sidelobe suppression while ensuring that the beamwidth remains unchanged.

3. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that In step two, each preset particle is a set of Taylor-weighted amplitude distribution and scan-phase-weighted phase distribution required by the index; the number of particles N is set by the accuracy and speed required by the algorithm.

4. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that In step three, the number of calculation points for the upper and lower limit curves H and L, as well as the radiation beam P, is the same, and they are mapped one-to-one with the spatial angles.

5. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that The iterative motion in step five is characterized by: V R =k3V - Where k1, k2, k3 are iteration factors, R(i) is a uniformly distributed random value in the interval (0,1), and V - It is the motion vector from the previous iteration. The values ​​of the iteration factors k1, k2, and k3 are constants and change according to the required iteration speed.

6. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that Step six, which continues until the specified number of iterations is reached or the optimal particle in the population fully satisfies the optimization target curve, does not yield a unique optimization result. This step can be performed multiple times to obtain multiple different optimization results.

7. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that In step seven, it is necessary to verify whether the radiation beam result generated by the optimal particle of the swarm, i.e. the optimal channel amplitude phase, meets the beam equalization requirements. The iteration result must meet the amplitude phase quantization requirements of each radar channel and be applied to the actual radar array. The beam result after amplitude phase quantization of each channel is verified by electromagnetic simulation software.

8. The phased array weather radar beam equalization method based on the preset particle swarm algorithm according to claim 1, characterized in that In step eight, the scanning angle parameter is changed: the scanning angle setting must satisfy all scanning spatial domains, and the change rule is θ. i =iθ p i∈(0,…,θ) m / θ p -1), where θ i It is the beam direction, θ p It is the beamwidth, θ m The indicator requires the maximum beam pointing.

Images