Sea wave simulation method based on model matching in marine radar

Through the spectrum analysis and model matching technology of sea radar image sequences, the problem of difficulty in obtaining and simulation of wave spectrum is solved, and high accuracy and low-cost measurement of wave simulation are achieved.

CN119047194APending Publication Date: 2024-11-29XIDIAN UNIV +1

Patent Information

Application Number
CN202411211445.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

In the existing wave simulation technology, the wave spectrum is difficult to obtain and the simulation does not conform to the actual sea conditions, resulting in large errors in the simulation results.

Method used

The sea radar image sequence is used for spectral analysis, the wave spectrum model is selected through model matching, and the wave spectrum model is simulated by linear superposition method. The mean difference and mean square deviation of the on-site wave spectrum and the wave spectrum model are used as selection criteria to ensure the simulated wave spectrum model. Match with the actual environment.

Benefits of technology

Accurate measurement and simulation of wave spectrum is achieved, the accuracy of wave simulation is improved, the cost and manpower and material investment of traditional measurement tools are reduced, and the errors caused by improper model selection are avoided.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119047194A_ABST
    Figure CN119047194A_ABST
Patent Text Reader

Abstract

The invention discloses a sea wave simulation method based on model matching in a marine radar. The sea wave simulation method comprises the following implementation steps: acquiring an image spectrum from a radar image sequence by adopting a spectrum analysis method; obtaining an energy spectrum from the image spectrum, and solving the ocean current velocity by using the energy spectrum; performing band-pass filtering on the energy spectrum to filter out non-sea wave information; a wave number spectrum is obtained through energy spectrum integration, the wave number spectrum is corrected, and a field sea wave spectrum is obtained through calculation; constructing a sea wave spectrum model by using the actual environmental parameters, and matching the sea wave spectrum model with a field sea wave spectrum; and generating a simulation direction spectrum through the matched optimal sea wave spectrum model and the direction distribution function, and obtaining simulated sea waves according to the simulation direction spectrum and a linear superposition method. The problem that the sea wave spectrum is difficult to obtain is solved, errors caused by the fact that the sea wave spectrum model does not conform to the actual sea condition are avoided, and the accuracy of sea wave simulation is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of radar technology, and further relates to a wave simulation method based on model matching in a marine radar in the field of wave simulation technology. The present invention can provide accurate data that conforms to actual sea conditions for wave information measurement. Background Art

[0002] Marine radar has high temporal and spatial resolution and can monitor waves around the clock. Compared with traditional measurement tools such as buoys, it has a wider measurement range and higher real-time performance. In actual engineering, in order to reduce the adverse effects caused by data shortages, simulated wave height and other values ​​can provide accurate data for applications such as wave information inversion. A widely used wave simulation technology is based on wave spectrum simulation. This method mainly combines the wave spectrum model with the directional distribution function, and then generates waves through linear superposition. When using traditional tools such as buoys to measure the wave spectrum, the measurement is difficult, and the measurement results can only reflect the wave spectrum in a small sea area. The measurement is easily affected by sea conditions, so in many cases the wave spectrum is difficult to obtain. In addition, the subjectively selected wave spectrum model will not conform to the actual sea conditions. The difference between the simulated environmental parameters and the actual environmental parameters will seriously affect the simulation effect.

[0003] Dalian Maritime University disclosed a real-time wave height simulation method for various sea conditions in its patent application "A real-time wave simulation method for various sea conditions" (patent application number CN 202210267649.4, application publication number CN 114818262 A). The implementation steps of this method are: (1) Select the wave spectrum form as needed; (2) Obtain the wave spectrum parameters through the wave parameter information table or real wave data; (3) Determine the amplitude and initial phase of the corresponding harmonic based on the wave spectrum and the discretized angular frequency, so as to obtain the harmonic component; (4) Select the diffusion function and combine the harmonic components at each frequency with the diffusion function; (5) Superimpose all models after combining each harmonic component with the corresponding diffusion function, and finally obtain a three-dimensional wave simulation model. This method has the advantages of strong real-time performance and good authenticity, and has broad application prospects in the fields of wave simulation and ship motion control. However, this method still has the following two shortcomings: First, this method proposes to query the wave spectrum parameters through the wave parameter information table, and the sea condition classification in the wave parameter information table is only ten levels, but the actual sea condition is very complicated, and the subjective selection of environmental parameters in the information table will cause large errors; second, the actual wave energy is distributed in different directions and frequency ranges. This method only considers the frequency when modeling the movement of linear waves on the ocean surface, ignoring the influence of direction, which reduces the accuracy of wave simulation.

[0004] Shanghai Jiao Tong University disclosed a three-dimensional wave height simulation method based on wave spectrum in its patent document "Three-dimensional wave simulation method based on wave spectrum" (patent application number CN 201210461513.3, application publication number CN 102982239 A). The implementation steps of the method are: (1) determine the observation time and obtain the wave height at that time; (2) take a certain time before the observation time when the sea surface is relatively calm as the observation time 0, and obtain the wave height at time 0; (3) on the basis of considering the time history of wave formation, derive the correlation function of the wave, and calculate the attenuation weight function of the wave; (4) multiply the wave height at time 0 by the attenuation weight function and superimpose it on the wave height at the observation time; (5) change the initial phase angle of the directional spectrum function of the superimposed wave to simulate the change of wind direction; (6) repeat the above steps, take the waves at multiple times and superimpose them, and finally obtain the simulation of the wave. This method reflects the influence of the waves at the previous moment on the waves at the observation moment. The simulation effect of the waves is less distorted and consistent with the actual waves. However, this method still has the following two shortcomings: First, wind and water movement are the main causes of wave formation. This method only considers the change of wind direction and ignores the influence of water movement, which reduces the accuracy of wave simulation; second, the angle initial phase parameters of wind speed and directional spectrum function need to match the actual sea conditions. Inappropriate subjectively selected simulation parameters will lead to large errors. Summary of the invention

[0005] The purpose of the present invention is to address the deficiencies of the above-mentioned prior art and to propose a wave simulation method based on model matching in marine radar, aiming to solve the problem that the wave spectrum is difficult to obtain and the wave simulation does not conform to the actual sea conditions.

[0006] The technical idea for achieving the purpose of the present invention is that the present invention uses spectral analysis to extract the on-site wave spectrum from the marine radar image sequence, and through model matching, selects the wave spectrum model for simulation to achieve wave simulation. The present invention uses marine radar images to measure the on-site wave spectrum, and the measurement results are real-time and accurate. When traditional marine radars such as buoys measure wave spectra, the measurement range is small and easily affected by sea conditions, and the deployment and maintenance at sea require a lot of manpower and material resources. Marine radar has the advantages of high temporal and spatial resolution and high cost performance. Its non-contact measurement method will not be affected by sea conditions and can operate all day and all weather, solving the problem that wave spectra are difficult to obtain. The present invention selects a wave spectrum model for wave simulation through a model matching method, and the simulation effect is real and reliable. Single subjective selection of a wave spectrum model for simulation has the problem that the wave spectrum model does not conform to the actual sea conditions, and the simulation lacks objectivity and scientificity. A wave spectrum model is constructed according to the on-site environmental parameters. The mean error and mean square error between the on-site wave spectrum and the wave spectrum model are used as the standard for selecting the model. The best wave spectrum model is matched and then the linear superposition method is used to simulate the waves, avoiding the errors caused by improper model selection.

[0007] The steps of the ocean wave simulation method of the present invention include the following:

[0008] Step 1: convert the marine radar image sequence without target pollution from the polar coordinate system to the Cartesian coordinate system, perform spectrum analysis on the radar image sequence in the Cartesian coordinate system, and obtain the image spectrum corresponding to each marine radar image;

[0009] Step 2, using the energy spectrum corresponding to each image spectrum, calculate the current velocity of the sea area corresponding to the marine radar image;

[0010] Step 3, band-pass filtering each energy spectrum to obtain an energy spectrum without wave information;

[0011] Step 4, integrating the filtered energy spectrum in the frequency domain to obtain a wave number spectrum of the sea area corresponding to the marine radar image, correcting the wave number spectrum, and calculating the on-site sea wave spectrum from the corrected wave number spectrum;

[0012] Step 5, construct each wave spectrum model using the environmental parameters measured on site, and match each wave spectrum model with the on-site wave spectrum;

[0013] Step 6, generating a simulated directional spectrum from the matched wave spectrum model and the directional distribution function, and obtaining a simulated wave according to the simulated directional spectrum and the linear superposition method.

[0014] Compared with the prior art, the present invention has the following advantages:

[0015] First, the present invention uses spectral analysis to measure the on-site wave spectrum from the marine radar image sequence. The measurement results are real-time and accurate with high temporal and spatial resolution. The marine radar can operate around the clock and in all weather conditions, overcoming the difficulty in obtaining the wave spectrum. The present invention reduces the cost of deploying and maintaining traditional measuring tools such as buoys, and can accurately measure the wave spectrum, especially under extreme sea conditions.

[0016] Second, the present invention matches the on-site wave spectrum and the wave spectrum model, uses the mean error and the mean square error as the selection criteria, objectively and scientifically determines the best wave spectrum model, models based on actual environmental parameters, and simulates waves in combination with the linear superposition method, thereby overcoming the impact of the prior art wave spectrum model and parameters not conforming to actual sea conditions, so that the present invention avoids the errors caused by improper model selection and improves the accuracy of wave simulation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a flow chart of the method of the present invention;

[0018] Figure 2 The original marine radar image according to an embodiment of the present invention;

[0019] Figure 3 is a radar image in a Cartesian coordinate system in an embodiment of the present invention;

[0020] Figure 4 It is a schematic diagram of an image spectrum, an energy spectrum before filtering, an energy spectrum after filtering, a wavenumber spectrum before correction, a corrected wavenumber spectrum, and an on-site wave spectrum in an embodiment of the present invention;

[0021] Figure 5 It is a comparison diagram of the normalized wave spectrum before the cutoff frequency between the on-site wave spectrum and the PM spectrum and the ITTC spectrum in the embodiment of the present invention;

[0022] Figure 6 Schematic diagram of ocean waves simulated by the present invention. DETAILED DESCRIPTION

[0023] The present invention is further described in detail below in conjunction with the accompanying drawings and embodiments.

[0024] Reference Figure 1 , further describing the implementation steps of the embodiment of the present invention.

[0025] Step 1: Convert the marine radar image sequence without target pollution from the polar coordinate system to the Cartesian coordinate system, perform spectrum analysis on the radar image sequence in the Cartesian coordinate system, and obtain the image spectrum corresponding to each marine radar image.

[0026] The marine radar image sequence used in the embodiment of the present invention is composed of 32 consecutive marine radar images. The image sequence was collected by a marine radar located on Yangma Island, Yantai, China. The collection time was October 2019, and the image format is mat. Figure 2 is the original radar image in the polar coordinate system, where the radial direction represents the distance in meters and the tangential direction represents the azimuth in degrees. Figure 2 The different colors in the chart represent the radar echo signal strength.

[0027] The conversion of the polar coordinate system into the Cartesian coordinate system can be achieved by using any one of the nearest neighbor interpolation method, bilinear interpolation method and bicubic interpolation method.

[0028] The embodiment of the present invention adopts the nearest neighbor interpolation method to convert the original marine radar image corresponding to the analysis sea area from the polar coordinate system to the Cartesian coordinate system.

[0029] Reference Figure 3 , the radar image in the Cartesian coordinate system of the present invention is further described.

[0030] The embodiment of the present invention obtains the radar echo signal strength corresponding to each pixel point through the nearest neighbor interpolation method. In order to facilitate the extraction of the on-site wave spectrum, a radar scanning sea area is delineated for analysis. The selected analysis sea area is free of target pollution such as ships and islands. In addition, the average period of the waves is generally 5 to 11 seconds, and at least 3 complete wave waveforms are included in the selected data analysis area. Figure 2 After the analysis sea area is determined, the original marine radar image corresponding to the analysis sea area is converted from the polar coordinate system to the Cartesian coordinate system. Figure 3 That is, a radar image in the Cartesian coordinate system of the analyzed sea area. Figure 3 The radar image of the selected analysis area in Cartesian coordinates. The horizontal axis represents the distance in the x direction in meters, and the vertical axis represents the distance in the y direction in meters. The pixel color is Figure 2 The pixel colors have the same meaning. The analysis time domain size is 768m×768m, and the radar image is evenly divided into 128×128 pixels.

[0031] Reference Figure 4 , and further describes the use of spectral analysis to extract on-site wave spectra.

[0032] Figure 4 It is the image spectrum, the energy spectrum before filtering, the energy spectrum after filtering, the wavenumber spectrum before correction, the corrected wavenumber spectrum and the on-site wave spectrum in the embodiment of the present invention. Figure 4 (a) to Figure 4 The x-axis in (e) represents the component k of the wave number in the x direction. x , the y-axis represents the component k of the wave number in the y direction y , the value range is -0.52~0.52rad / m, the z-axis represents the energy intensity and corresponds to the color.

[0033] The spectrum analysis process can be implemented by using any one of Fourier transform, wavelet transform and Nyquist transform.

[0034] The spectrum analysis processing method used in the embodiment of the present invention is three-dimensional Fourier transform. By performing Fourier transform on a radar image sequence with a time span of 80 seconds, the radar echo signal is reflected in the wavenumber frequency domain, and the angular frequency ω of the obtained image spectrum ranges from 0 to 1.256 rad / s. Figure 4 (a) is an image spectrum obtained by Fourier transform of the radar image sequence, with an angular frequency of 1.17 rad / s.

[0035] Step 2: Calculate the ocean current velocity in the sea area corresponding to the marine radar image using the energy spectrum corresponding to each image spectrum.

[0036] The energy spectrum corresponding to each image spectrum is obtained by the following formula:

[0037]

[0038] Among them, I j (·) represents the energy spectrum corresponding to the jth image spectrum in the Cartesian coordinate system, k x , k y They represent the wave number components of the waves in the x and y directions in the Cartesian coordinate system, ω represents the angular frequency component of the waves, and L x , L y , T represent the maximum values ​​of the marine radar image in the x direction, y direction and time in the Cartesian coordinate system, |·| represents the modulo operation, and F j (·) represents the image spectrum corresponding to the j-th marine radar image in the Cartesian coordinate system.

[0039] Figure 4 (b) Figure 4 (a) The energy spectrum corresponding to the image spectrum shown. Figure 4 (b) It can be seen that the energy spectrum of the unfiltered wave includes energy information of waves, harmonics and noise. The energy spectrum distribution range is scattered, and the superposition of the three parts of energy information shows a higher spectrum value.

[0040] The ocean current velocity of the sea area corresponding to the marine radar image is obtained by the following formula:

[0041]

[0042] Among them, U x , U y They represent the velocity components of the ocean current in the x and y directions in the Cartesian coordinate system, ∑· represents the summation operation, and ω j represents the angular frequency corresponding to the jth energy spectrum, g represents the local gravitational acceleration, k represents the wave number of the wave, tanh(·) represents the hyperbolic tangent function, d represents the average depth of the sea water in the sea area corresponding to the radar image, and the superscript -1 represents the inverse operation.

[0043] The core idea of ​​calculating the velocity of ocean currents is the least squares method. The least squares method is used to minimize the difference between the angular frequency component given by the theoretical dispersion relation and the angular frequency component calculated by the energy spectrum. Based on the least squares method, the velocity of the ocean surface laminar flow can be directly calculated and the actual dispersion relation satisfied by the ocean waves can be obtained.

[0044] In the embodiment of the present invention, the following is introduced Figure 4 (b) The energy spectrum represented by is used as the weight to more effectively calculate the current velocity. The current velocity calculated in the embodiment of the present invention is 1.03 m / s.

[0045] Step 3: Bandpass filter each energy spectrum to obtain an energy spectrum without wave information.

[0046] The steps of bandpass filtering each energy spectrum are as follows:

[0047] The first step is to obtain the dispersion relation based on the ocean current velocity in the sea area corresponding to the radar image according to the following formula, and calculate the wave number from the dispersion relation:

[0048]

[0049] in, Represents the wave number vector of the sea area corresponding to the radar image, Represents the ocean current velocity vector of the sea area corresponding to the radar image,

[0050] In the second step, the size of the wave number corresponding to each energy spectrum is judged according to the following formula, and the energy that the wave number does not satisfy the dispersion relationship is filtered out:

[0051]

[0052] Among them, E j (·) represents the jth energy spectrum after filtering, ω j-1 represents the angular frequency corresponding to the j-1th energy spectrum, ω j+1 Represents the angular frequency corresponding to the j+1th energy spectrum.

[0053] In the embodiment of the present invention, the analysis area is a deep water area, and the dispersion relationship can be simplified as:

[0054]

[0055] Figure 4 (c) Figure 4 (b) The energy spectrum after bandpass filtering, filtering out non-wave information. Figure 4 (c) It can be seen that after calculating the flow velocity, the energy spectrum can be separated by bandpass filtering using the actual wave dispersion relationship satisfied by the waves. Figure 4 The energy distribution in (c) is Figure 4 The energy distribution in (b) is more concentrated, and the spectrum value becomes lower because of filtering out non-wave information.

[0056] Step 4: Integrate the filtered energy spectrum in the frequency domain to obtain a wavenumber spectrum of the sea area corresponding to the marine radar image, correct the wavenumber spectrum, and calculate the on-site wave spectrum from the corrected wavenumber spectrum.

[0057] The steps of calculating the wavenumber spectrum correction are as follows:

[0058] The first step is to calculate the wave number spectrum of the filtered energy spectrum in the frequency domain according to the following formula:

[0059] H(k x ,k y ) = ∫E j (k x ,k y ,ω)dω

[0060] Where H(·) represents the wavenumber spectrum of the filtered energy spectrum in the frequency domain, and dω represents the infinitesimal element of the angular frequency ω.

[0061] Figure 4 (d) Reason Figure 4 (c) represents the wave number spectrum obtained by integrating the filtered energy spectrum in the frequency domain.

[0062] The on-site directional spectrum is calculated from the corrected wavenumber spectrum, and finally the measured on-site wave spectrum is obtained by integrating in the direction.

[0063] The second step is to correct the wave number spectrum of the waves according to the following formula:

[0064] G(k x ,k y )=M·H(k x ,k y )

[0065] Where G(·) represents the modified wave number spectrum, M represents the modulation transfer function, and M = k -1.2 .

[0066] Figure 4 (e) Reason Figure 4 (d) represents the wave number spectrum after the wave number spectrum is corrected by the modulation transfer function based on engineering experience. Figure 4 (e) It can be seen that the corrected wave number spectrum is more concentrated than that before correction, which is consistent with the actual environment that the wave frequency, main wave direction and wave wavelength of a fixed sea area will not change suddenly in a short time.

[0067] The steps of calculating the on-site wave spectrum are as follows:

[0068] The first step is to calculate the on-site directional spectrum according to the following formula:

[0069]

[0070] Among them, O(ω,θ) represents the on-site directional spectrum corresponding to the wave angular frequency ω and the wave propagation direction θ, and dk is the infinitesimal element of the wave number k.

[0071] The second step is to calculate the on-site wave spectrum according to the following formula:

[0072]

[0073] Among them, S(·) represents the on-site wave spectrum of the sea area corresponding to the radar image, and dθ is the infinitesimal element of the wave propagation direction θ.

[0074] Figure 4 (f) is the on-site wave spectrum of the sea area corresponding to the radar image sequence in the embodiment of the present invention. Figure 4 (f) It can be seen that the horizontal axis of the on-site wave spectrum represents the wave angular frequency, in rad / s. Since the radar rotation speed affects the cutoff frequency of the angular frequency, the radar rotation speed in the embodiment of the present invention is 24r / min, and the calculated cutoff frequency of the angular frequency is 1.256rad / s. Therefore, Figure 4 The horizontal axis in (f) is only from 0 to 1.256, and the vertical axis is normalized. The frequency corresponding to the maximum spectrum value of the on-site wave spectrum is 1.2 rad / s.

[0075] Step 5: construct each wave spectrum model using the environmental parameters measured on site, and match each wave spectrum model with the on-site wave spectrum.

[0076] The wave spectrum model can be any two or more of the PM spectrum, ITTC spectrum, JONSWAP spectrum, and Wen spectrum.

[0077] The embodiment of the present invention uses PM spectrum and ITTC spectrum as the wave spectrum models to be matched, and the expressions thereof are as follows:

[0078]

[0079] Among them, S PM (·) represents PM spectrum, S ITTC (·) indicates ITTC spectrum, α indicates a constant, α=8.1×10 -3 , β is a constant with a value of 0.74, U 19.5 Indicates the wind speed at 19.5 meters from the sea surface corresponding to the radar image, H s , T 0.1 They respectively represent the significant wave height and average wave period of the sea area corresponding to the radar image.

[0080] The steps of matching each wave spectrum model with the on-site wave spectrum are as follows:

[0081] The first step is to use the environmental parameters such as wind speed and effective wave height measured on site to build the wave spectrum model to be matched;

[0082] The second step is to calculate the mean error and mean square error between the on-site wave spectrum and each wave spectrum model;

[0083] The third step is to find the wave spectrum model with the smallest mean error and mean square error, and determine it as the wave spectrum model that best matches the on-site wave spectrum.

[0084] Reference Figure 5 , further describing the application of model matching.

[0085] Figure 5 It is a comparison diagram of the normalized wave spectrum before the cutoff frequency between the on-site wave spectrum and the PM spectrum and ITTC spectrum in the embodiment of the present invention. Figure 5 The horizontal axis represents the angular frequency of the waves in rad / s, which ranges from 0 to 1.256, and the vertical axis is normalized. Figure 5 The blue curve represents the on-site wave spectrum, and the frequency corresponding to its maximum spectrum value is 1.2rad / s. The red curve represents the PM spectrum, which basically tends to 0. This is because the PM spectrum is only related to the wind speed at 19.5 meters from the local sea surface. The construction of the PM spectrum only considers the influence of wind and does not consider the influence of water movement. The yellow curve represents the ITTC spectrum, and the frequency corresponding to its maximum spectrum value is 1.2rad / s. The actual environmental parameters corresponding to the analyzed sea area are a wind speed of 2.77m / s at 19.5 meters, an effective wave height of 1.07m, and an average wave period of 3.96s.

[0086] The wave spectrum model is built using the environmental parameters measured on site, and the mean error, mean square error and other indicators between the on-site wave spectrum and the wave spectrum model are calculated to match the wave spectrum model that best fits the actual sea conditions. All calculation results are plotted in Table 1:

[0087] Table 1. Matching of on-site wave spectrum and wave spectrum model in simulation experiment

[0088] index PM Spectrum ITTC Spectrum Mean difference 0.23 0.04 Mean square error 0.4 0.06

[0089] Depend on Figure 5 As well as the mean error and mean square error indicators, it can be obtained that the ITTC spectrum matches the on-site wave spectrum better than the PM spectrum. Therefore, the wave spectrum model used for simulation in the embodiment of the present invention is the ITTC spectrum.

[0090] Step 6, generating a simulated directional spectrum from the matched wave spectrum model and the directional distribution function, and obtaining a simulated wave according to the simulated directional spectrum and the linear superposition method.

[0091] The steps of generating a simulated directional spectrum from the matched wave spectrum model and directional distribution function are as follows:

[0092] In the first step, a directional distribution function is constructed according to the following formula. The directional distribution function can be any one of a cosine square directional distribution function and an optically easy directional distribution function.

[0093] In the embodiment of the present invention, the cosine square directional distribution function is selected as the directional distribution function, and its expression is as follows:

[0094] D(ω,θ)=2 / π·cos 2 (θ-θ0),|θ-θ0|<π / 2

[0095] Where D(·) represents the cosine square directional distribution function, and θ0 represents the main wave direction of the sea waves in the sea area corresponding to the radar image.

[0096] The second step is to generate the simulated directional spectrum according to the following formula:

[0097] P(ω,θ)=S best (ω)·D(ω,θ)

[0098] Where P(·) represents the simulated directional spectrum, S best (·) indicates the wave spectrum model that best matches the on-site wave spectrum.

[0099] The steps of obtaining simulated ocean waves according to the simulated directional spectrum and the linear superposition method are as follows:

[0100] The first step is to calculate the wave amplitude according to the following formula:

[0101]

[0102] Where a(·) represents the wave amplitude, Δω and Δθ represent the angular frequency and directional resolution, respectively.

[0103] In the second step, the simulated ocean waves are obtained using the linear superposition method described in the following formula:

[0104]

[0105] Where g(x,y,t) represents the wave height at the t-th moment, where the Cartesian coordinate system is located at (x,y), ω n represents the angular frequency of the nth wave, θ n represents the propagation direction of the nth wave, a n (·) represents the amplitude of the nth wave, k n represents the wave number of the nth wave, k n ,ω n Determined by the dispersion relation, ε n Represents the phase of the nth wave.

[0106] In the embodiment of the present invention, the cosine square directional distribution function is selected as the directional distribution function, and its simulation parameter is the main wave direction of 28.75° actually measured in the sea area corresponding to the radar image. The directional distribution function and the selected wave spectrum model are used to generate a simulated directional spectrum, and the wave height map is generated by the simulated directional spectrum and the linear superposition method. In the embodiment of the present invention, the random phase of the wave is uniformly distributed in [0,2π].

[0107] Reference Figure 6, the simulated ocean waves of the present invention are further described.

[0108] Depend on Figure 6 It can be seen that it is a simulated wave diagram at 2.5s, where Figure 6 (a) is a three-dimensional picture of simulated waves. Figure 6 (b) is a two-dimensional image of simulated ocean waves. Figure 6 The x-axis and y-axis represent the range of the simulated sea area, in meters, with values ​​ranging from 0 to 768 meters. The z-axis represents the simulated wave height, in meters, with values ​​ranging from -1 to 0.8 meters, and the color corresponds to the wave height.

[0109] The present invention matches the on-site wave spectrum and the wave spectrum model, uses the mean error and the mean square error as selection criteria, objectively and scientifically determines the best wave spectrum model, models based on actual environmental parameters, and simulates waves in combination with the linear superposition method, thereby overcoming the influence caused by the wave spectrum model and parameters in the prior art not conforming to the actual sea conditions, so that the present invention avoids the error caused by improper model selection and improves the accuracy of wave simulation.

Claims

1. A wave simulation method based on model matching in marine radar, characterized in that: The spectrum analysis method is used to extract the on-site wave spectrum from the marine radar image sequence, and the wave spectrum model for simulation is selected through model matching to realize wave simulation. The steps of the wave simulation method include the following: Step 1: convert the marine radar image sequence without target pollution from the polar coordinate system to the Cartesian coordinate system, perform spectrum analysis on the radar image sequence in the Cartesian coordinate system, and obtain the image spectrum corresponding to each marine radar image; Step 2, using the energy spectrum corresponding to each image spectrum, calculate the current velocity of the sea area corresponding to the marine radar image; Step 3, band-pass filtering each energy spectrum to obtain an energy spectrum without wave information; Step 4, integrating the filtered energy spectrum in the frequency domain to obtain a wave number spectrum of the sea area corresponding to the marine radar image, correcting the wave number spectrum, and calculating the on-site sea wave spectrum from the corrected wave number spectrum; Step 5, construct each wave spectrum model using the environmental parameters measured on site, and match each wave spectrum model with the on-site wave spectrum; Step 6, generating a simulated directional spectrum from the matched wave spectrum model and the directional distribution function, and obtaining a simulated wave according to the simulated directional spectrum and the linear superposition method.

2. The method for simulating ocean waves based on model matching in marine radar according to claim 1, characterized in that: The conversion of the polar coordinate system into the Cartesian coordinate system in step 1 can be implemented by any one of the nearest neighbor interpolation method, bilinear interpolation method, and bicubic interpolation method; the spectral analysis processing can be implemented by any one of the Fourier transform, wavelet transform, and Nyquist transform methods.

3. The method for simulating ocean waves based on model matching in marine radar according to claim 1, characterized in that: The energy spectrum corresponding to each image spectrum in step 2 is obtained by the following formula: Among them, I j (·) represents the energy spectrum corresponding to the jth image spectrum in the Cartesian coordinate system, k x , k y They represent the wave number components of the waves in the x and y directions in the Cartesian coordinate system, ω represents the angular frequency component of the waves, and L x , L y , T represent the maximum values ​​of the marine radar image in the x direction, y direction and time in the Cartesian coordinate system, |·| represents the modulo operation, and F j (·) represents the image spectrum corresponding to the j-th marine radar image in the Cartesian coordinate system.

4. The method for simulating ocean waves based on model matching in marine radar according to claim 3, characterized in that: The ocean current velocity of the sea area corresponding to the marine radar image in step 2 is obtained by the following formula: Among them, U x , U y They represent the current velocity components in the x and y directions in the Cartesian coordinate system, ∑· represents the summation operation, and ω j represents the angular frequency corresponding to the jth energy spectrum, g represents the gravitational acceleration, k represents the wave number of the wave, tanh(·) represents the hyperbolic tangent function, d represents the average depth of the sea water in the sea area corresponding to the radar image, and the superscript -1 represents the inverse operation.

5. The method for simulating ocean waves based on model matching in marine radar according to claim 4, characterized in that: The steps for bandpass filtering each energy spectrum described in step 3 are as follows: The first step is to obtain the dispersion relation based on the ocean current velocity in the sea area corresponding to the radar image according to the following formula, and calculate the wave number from the dispersion relation: in, Represents the wave number vector of the sea area corresponding to the radar image, Represents the ocean current velocity vector of the sea area corresponding to the radar image, In the second step, the size of the wave number corresponding to each energy spectrum is judged according to the following formula, and the energy that the wave number does not satisfy the dispersion relationship is filtered out: Among them, E j (·) represents the jth energy spectrum after filtering, ω j-1 represents the angular frequency corresponding to the j-1th energy spectrum, ω j+1 Represents the angular frequency corresponding to the j+1th energy spectrum.

6. The method for simulating ocean waves based on model matching in marine radar according to claim 5, characterized in that: The steps for calculating the wavenumber spectrum correction described in step 4 are as follows: The first step is to calculate the wave number spectrum of the filtered energy spectrum in the frequency domain according to the following formula: H(k x ,k y )=∫E j (k x ,k y ,ω)dω Where H(·) represents the wavenumber spectrum of the filtered energy spectrum in the frequency domain, and dω represents the infinitesimal element of the angular frequency ω; The second step is to correct the wave number spectrum of the waves according to the following formula: G(k x ,k y )=M·H(k x ,k y ) Where G(·) represents the modified wave number spectrum, M represents the modulation transfer function, and M = k -1.2 .

7. The method for simulating ocean waves based on model matching in marine radar according to claim 6, characterized in that: The steps for calculating the on-site wave spectrum described in step 4 are as follows: The first step is to calculate the on-site directional spectrum according to the following formula: Among them, O(ω,θ) represents the field directional spectrum corresponding to the angular frequency of the wave ω and the wave propagation direction θ, and dk is the infinitesimal element of the wave number k; The second step is to calculate the on-site wave spectrum according to the following formula: Among them, S(·) represents the on-site wave spectrum of the sea area corresponding to the radar image, and dθ is the infinitesimal element of the wave propagation direction θ.

8. The method for simulating ocean waves based on model matching in marine radar according to claim 1, characterized in that: The steps for matching each wave spectrum model with the on-site wave spectrum described in step 5 are as follows: The first step is to use the environmental parameters such as wind speed and effective wave height measured on site to build the wave spectrum model to be matched; The second step is to calculate the mean error and mean square error between the on-site wave spectrum and each wave spectrum model; The third step is to find the wave spectrum model with the smallest mean error and mean square error, and determine it as the wave spectrum model that best matches the on-site wave spectrum; The wave spectrum model can be any two or more of the PM spectrum, ITTC spectrum, JONSWAP spectrum, and Wen spectrum.

9. The method for simulating ocean waves based on model matching in marine radar according to claim 7, characterized in that: The steps of generating a simulated directional spectrum from the matched wave spectrum model and directional distribution function in step 6 are as follows: The first step is to construct a directional distribution function according to the following formula, wherein the directional distribution function can be any one of a cosine square directional distribution function and an optically easy directional distribution function; The second step is to generate the simulated directional spectrum according to the following formula: P(ω,θ)=S best (ω)·D(ω,θ) Where P·(·) represents the simulated directional spectrum, S best (·) represents the wave spectrum model that best matches the on-site wave spectrum, and D(·) represents the directional distribution function.

10. The method for simulating ocean waves based on model matching in marine radar according to claim 9, characterized in that: The steps of obtaining the simulated ocean waves according to the simulated directional spectrum and the linear superposition method described in step 6 are as follows: The first step is to calculate the wave amplitude according to the following formula: Where a(·) represents the wave amplitude, Δω and Δθ represent the angular frequency and directional resolution, respectively; In the second step, the simulated ocean waves are obtained using the linear superposition method described in the following formula: Where g(x,y,t) represents the wave height at the t-th moment, where the Cartesian coordinate system is located at (x,y), ω n represents the angular frequency of the nth wave, θ n represents the propagation direction of the nth wave, a n (·) represents the amplitude of the nth wave, k n represents the wave number of the nth wave, k n ,ω n Determined by the dispersion relation, ε n Represents the phase of the nth wave.

Citation Information

Patent Citations

  • Three-dimensional ocean wave analogy method based on ocean wave spectrum

    CN102982239A

  • Real-time sea wave simulation method for various sea conditions

    CN114818262A

Cited By

  • Wave sensor detection system based on wave spectrum and control method

    CN119879993A

  • A Wave Sensor Detection System and Control Method Based on Wave Spectrum

    CN119879993B

  • Three-dimensional rough sea surface electromagnetic reflection coefficient calculation method based on wind direction factor

    CN120065206A

  • Method for calculating electromagnetic reflection coefficient of three-dimensional rough sea surface based on wind direction factor

    CN120065206B

  • Ship structure safety forecasting method based on artificial neural network

    CN120822433A