A method for predicting the depth of an underwater moving target by combining numerical simulation and on-orbit data

Abstract

The application discloses a kind of numerical simulation and in-orbit data combined underwater moving target depth prediction method, comprising: the three-dimensional model of underwater target movement is described based on the fluid mechanics control equation of finite element analysis, and turbulence model is introduced to solve equation;Establish sea surface infrared radiation and atmospheric transmission model, solve the water surface infrared radiation characteristic map obtained by space-based detector;Combination of in-orbit satellite thermal imager observed shallow underwater moving target data contrast verification analysis, determine model parameters;According to correct model parameter simulation, induction is obtained the water surface wake map corresponding to underwater target of different depths;Establish the perfect simulation model for predicting the depth of underwater target according to the water surface wake map, and predict the corresponding depth of underwater moving target.The application can directly reflect the motion information of underwater moving target on the water surface, with the advantages of simple, intuitive and efficient, especially suitable for the application requirements of large field of view infrared detection underwater moving target depth.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of aerospace infrared photoelectric remote sensing detection, in particular to a method for predicting the depth of underwater moving targets by combining numerical simulation with on-orbit data. BACKGROUND

[0002] The depth detection of underwater moving targets is of great significance to marine science, resource development, and environmental protection. On the one hand, underwater targets such as underwater vehicles and industrial facilities generate noise when moving, which affects marine life, especially species that rely on sound waves for communication and navigation, such as whales and dolphins. Different depths of vehicles affect different species. On the other hand, with the advancement of technology, the number of underwater moving targets has increased, and the waste generated by their movement has gradually increased the pollution of the ocean. Monitoring of vehicles moving at different depths has become increasingly important.

[0003] Chinese patent document CN113379710A discloses a system and method for accurately measuring underwater targets using sonar to improve the accuracy and precision of sonar measurements. Chinese patent document CN118485908A discloses an underwater target recognition method based on multi-modal fusion, which fuses sonar data, optical data, and text description features to more comprehensively utilize information from different modalities to improve the accuracy and robustness of underwater target recognition.

[0004] In recent years, due to the application of shock absorption and noise reduction as well as stealthy shells, it has become increasingly difficult to detect underwater moving targets using traditional methods such as sonar. When underwater vehicles move in seawater, the surrounding seawater temperature rises due to heat dissipation from the power system, water disturbance from the propeller, and movement friction. These higher-temperature seawater, affected by buoyancy and propeller disturbance, will rise to the sea surface to form a thermal wake that is different from the surrounding environment. This thermal wake has a large range and a long duration, making it easy to be detected by infrared detectors. Therefore, infrared detection can be an important means for detecting underwater moving submarine targets.

[0005] Currently, due to factors such as the size of underwater moving targets and their working environment, it is difficult to conduct field experiments directly on the thermal wake generated by underwater vehicles moving at different depths, and it is difficult to accurately capture the impact of local temperature characteristics and water flow states on the wake. SUMMARY

[0006] To overcome the shortcomings of the prior art, the present application provides a method for predicting the depth of underwater moving targets by combining numerical simulation with on-orbit data, which can directly reflect the movement information of underwater moving targets on the water surface, has the advantages of simplicity, intuitiveness, and efficiency, and is particularly suitable for the application requirements of large-field infrared detection of underwater moving target depth.

[0007] A numerical simulation and in-orbit data combined underwater moving target depth prediction method, comprising the following steps:

[0008] (1) Establishing a three-dimensional model of the underwater moving target;

[0009] (2) Using a mesh division module to divide the three-dimensional model into a corresponding grid;

[0010] (3) Describing the three-dimensional model of the target movement based on the fluid dynamics equation of finite element analysis, and introducing a turbulence model to solve the equation;

[0011] (4) Configuring and initializing the turbulence model solving parameters;

[0012] (5) Using the initial parameters to iteratively solve the fluid dynamics equation until convergence, obtaining the shape, length, width, and temperature gradient characteristics of the wake of the underwater moving target at different depths;

[0013] (6) Establishing a sea surface infrared radiation model and an atmospheric transmission model, using the wake solved in step (5) as the initial value, substituting into the sea surface infrared radiation model and the atmospheric transmission model, to solve the water surface thermal wake temperature map obtained by the space-based detector and summarize and classify, extracting the image features;

[0014] (7) Combining the shallow underwater moving target data observed by the satellite thermal imager in orbit for comparative verification analysis, to determine whether the error meets the requirements, if not, perfecting the parameters of each model, if yes, determining the parameters of the turbulence model, the sea surface infrared radiation model, and the atmospheric transmission model;

[0015] (8) Using the turbulence model, the sea surface infrared radiation model, and the atmospheric transmission model with the determined parameters to perform simulation, and inducing the water surface thermal wake map corresponding to the underwater moving target at different depths; establishing a perfect simulation model for predicting the depth of the underwater target according to the water surface thermal wake image, to predict the corresponding depth of the underwater moving target.

[0016] In step (2), the mesh division includes different types, including overlapping mesh, MRF mesh, sliding mesh, and adaptive mesh.

[0017] In step (3), the control equation includes continuity equation, momentum conservation equation, and energy conservation equation; the turbulence model includes Spalart-Allmaras model, k-ε model (divided into Standard k-ε, RNG k-ε, and Realizable k-ε), k-ω model (divided into k-ω, BSL k-ω, and SST k-ω), etc.

[0018] In step (4), the solving parameters of the turbulence model are configured and initialized, including multiphase flow settings, region condition settings, boundary condition settings, dynamic mesh settings, solver option settings, convergence condition settings, and initialization settings;

[0019] Wherein, the multiphase flow includes gas-liquid two-phase flow; the region condition includes a background region and a target moving region; the boundary condition includes an inlet, an outlet, a model boundary, and a background region boundary; the dynamic mesh includes an overlapping mesh, an MRF mesh, a sliding mesh, and an adaptive mesh; the solver option includes control equation selection, turbulence equation selection, and data monitoring; the convergence condition includes parameter precision and convergence range; and the initialization setting includes inlet velocity and pressure setting of a fluid environment, and velocity and depth setting of an underwater moving target.

[0020] In step (6), the sea surface infrared radiation model includes a sea surface height field sea spectrum model, a sea surface spontaneous radiation and an environmental reflection model, and the atmospheric transmission model includes an atmospheric absorption, scattering, and radiation model.

[0021] In step (6), the image features include shape, length, width, and temperature gradient features of the thermal tail.

[0022] In step (7), the various model parameters include three-dimensional model parameters, mesh division parameters, turbulence model solving parameters, initialization configuration parameters, sea surface infrared radiation model parameters, and atmospheric transmission model parameters.

[0023] In step (7), the error requirement is that the temperature difference of the thermal tail is not more than 10 mK, and the shape, length, and width errors of the thermal tail are not more than 5%.

[0024] Compared with the prior art, the present application has the following beneficial effects:

[0025] The present application does not need to perform complex and difficult overseas experiments, and initiatively combines in-orbit data to establish a model, so that the accuracy is greatly improved, so that the motion information of an underwater moving target can be directly reflected on the water surface, and the present application has the advantages of simplicity, directness, high efficiency, and accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0026] Figure 1 A numerical simulation and in-orbit data combined underwater moving target depth prediction method flowchart of an embodiment of the present application;

[0027] Figure 2 A three-dimensional model schematic diagram constructed by an embodiment of the present application. DETAILED DESCRIPTION

[0028] The present application will be further described in detail below in combination with the drawings and embodiments, and it should be pointed out that the following embodiments are intended to facilitate the understanding of the present application and do not have any limiting effect on the present application.

[0029] In view of the problems of the volume of the ship body and the working environment restricting the in-situ detection of the underwater ship body wake, an underwater moving target depth prediction method combining numerical simulation and in-orbit data is designed, and the specific implementation steps are as follows:

[0030] (1) The geometric shape and length-diameter ratio of the underwater target are determined by taking the SUBOFF model as a standard to establish a scaled model. SUBOFF is a submarine test model designed by the Defense Advanced Research Projects Agency (DARPA) and the David Taylor Research Center (DTRC). A large number of scholars have used the SUBOFF model for extensive simulation and experimental simulation, therefore, the SUBOFF submarine can be used as a standard general calculation model, and the main geometric parameters and physical parameters of the submarine are shown in Table 1.

[0031] Table 1 Main dimension parameters of SUBOFF submarine model

[0032]

[0033] (2) The structured overlapping grid division method is adopted, and the calculation domain is divided into three parts by using a grid division module, including a background grid, a grid around the target ship body, a grid around the propeller, and a grid in the water discharge area.

[0034] (3) Large-scale vortex shedding phenomenon occurs during the submarine sailing movement in water. In order to capture the fine flow field characteristics, the LES method is adopted. The basic RANS closure model is used to solve the boundary layer and vortex-free area, and the LES sub-grid scale model is applied to the unsteady separation area. The control equations include the continuity equation, the momentum conservation equation and the energy conservation equation, which are respectively:

[0035] Continuity equation

[0036]

[0037] Momentum conservation equation

[0038]

[0039] Energy conservation equation

[0040]

[0041] Wherein, p is the fluid density, t is the time, ρ is the velocity vector, is the pressure, τ is the viscous stress, k eff is the effective thermal conductivity, S h is the source term, including radiation and other volume heat sources, and E is the total energy in the unit.

[0042] For moving ships, the thermal wake is typically caused by the propeller's disturbance of the water, friction between the hull and the water, and the high-temperature cooling water discharged during movement. These processes can all be categorized as two-phase flow problems. To clarify the heat and mass transfer behavior of water during buoyancy and the resulting sea surface temperature characteristics, it is necessary to accurately capture the gas-liquid interface present during buoyancy. The commonly used interface tracking model is the VOF model. The VOF model is a surface tracking method under a fixed Eulerian grid. Different fluid components share a set of momentum equations. During calculation, the volume fraction α of each fluid component is recorded in each cell within the flow field. The sum of the phase volume fractions in each cell is 1, and all variables and attributes are shared within each cell. This model is a gas-liquid two-phase flow. If water is defined as the m-th phase and air as the n-th phase, then: α m =0, the cell is entirely water; α m =1, the cell is entirely filled with air; 0 < α m <1, the unit is a gas-liquid interface.

[0043] In the VOF model, all phases share a set of momentum equations, and the velocity field obtained by solving them is shared by all phases. The momentum equations in the VOF model are:

[0044]

[0045] Where μ is the average viscosity. For pressure, This is the acceleration due to gravity.

[0046] (4) Configure and initialize the model solution parameters. The parameter configuration includes multiphase flow settings, region condition settings, boundary condition settings, dynamic mesh settings, solver option settings, convergence condition settings, and initialization settings.

[0047] The multiphase flow is set as a two-phase flow of air and seawater, with uniform air density and seawater density set as density stratification, where the density increases with increasing seawater depth;

[0048] The area condition is set to make the propeller rotation area a rotating area;

[0049] Boundary condition settings include appendices Figure 2 The boundary conditions shown are: wall boundary conditions, air-water interface boundary conditions, propeller rotation region boundary conditions, and boundary conditions near the drain outlet.

[0050] The dynamic mesh is set as an overlapping mesh that distinguishes between the background mesh and the component mesh;

[0051] The solver settings are used to select turbulence model parameters, wall treatment, and time step.

[0052] Convergence condition is set to monitor the calculation of residual, when the residual meets the condition, the iteration is calculated; initialization is set to limit the parameters of each phase to the maximum extent to approach the real physical field.

[0053] (5) The solver is used to solve the flow field of the underwater moving target on the sea surface.

[0054] (6) The sea surface infrared radiation and atmospheric transmission model is established, the water surface thermal wake temperature map obtained by the space-based detector is solved, and the image features are extracted. The model is as follows:

[0055] For the simulated sea surface infrared temperature, on the one hand, it will be affected by the emissivity of seawater and will not be 100% radiated, on the other hand, the energy radiated will be absorbed and scattered by the atmosphere before it reaches the space-based detector. In addition, the detection probability will also be affected by the performance of the infrared detector itself, so the three problems need to be considered comprehensively.

[0056] For the thermal wake in the form of surface source, the minimum resolvable temperature difference method is usually used to estimate the action distance of the infrared detector. Specifically, assuming that the minimum temperature difference ΔT that can be distinguished by the infrared detector is calculated by the following formula:

[0057]

[0058] Where SNR d is the signal-to-noise ratio, a and b are the horizontal and vertical angles of the instantaneous field of view, t0 is the scanning dwell time, t i is the integration time of the human eye, f is the spatial frequency, f p is the frame rate, Δf is the equivalent noise bandwidth, MTF(f) is the modulation transfer function, τ s is the optical system transmittance, NETD is the noise equivalent temperature difference, n e is the equivalent strip logarithm corresponding to a certain detection probability, α0 is the aspect ratio of the target, R is the distance between the infrared radiation source and the detector, τ(R) is the average atmospheric transmittance, and ∈ is the emissivity of seawater. According to Johnson's criterion, the relationship between the detection probability and the equivalent strip logarithm can be obtained, and the condition for the critical width of the detectable thermal wake is Only by satisfying the equation condition, the unique corresponding detection distance R can be obtained.

[0059] According to the above analysis, when the infrared radiation value of the sea surface thermal wake is transmitted to the detector and is greater than the minimum resolvable temperature difference (MRTD) of the detector, and the angular of the target to the system is greater than the minimum resolution angle of the system, it is considered that the detector can detect the thermal wake.

[0060] (7) The data of the satellite thermal imager for shallow underwater moving target is verified, and the error meets the requirements. The specific error requirements are as follows: the temperature difference of the wake is 10mK, the error of the shape, length and width of the wake is not more than 5%; if the requirements are not met, the model parameters are improved, and if the requirements are met, the model parameters are determined. The specific model parameters include the above grid volume, quantity, boundary conditions, turbulence equation, relaxation factor and the like.

[0061] (8) After the error meets the requirements, a perfect simulation model for predicting the depth of the underwater target according to the water surface image is established. According to the correct model parameters, the simulation is performed, and the water surface wake image corresponding to the underwater target at different depths is induced. The water surface thermal wake temperature image obtained by solving is summarized and classified, and the image features are extracted, including the shape, length, width and temperature difference of the thermal wake.

[0062] (9) According to the improved model, the depth of the underwater moving target is predicted.

[0063] The above embodiments have described the technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application, and is not used to limit the present application. Any modification, supplement and equivalent replacement made within the principle range of the present application should be included in the protection range of the present application.

Claims

1. A method for underwater moving target depth prediction combining numerical simulation and on-orbit data, characterized in that, The method comprises the following steps: (1) establishing a three-dimensional model of the underwater moving target; (2) performing mesh division on the three-dimensional model by using a mesh division module to generate corresponding meshes; (3) describing the three-dimensional model of the target movement based on a fluid dynamics equation of finite element analysis, and introducing a turbulence model to solve the equation; (4) configuring and initializing parameters for solving the turbulence model; (5) iteratively solving the fluid dynamics equation by using initial parameters until convergence is achieved, so as to obtain the shape, length, width and temperature gradient characteristics of the wake corresponding to the underwater moving target at different depths; (6) establishing a sea surface infrared radiation model and an atmospheric transmission model, taking the wake solved in step (5) as an initial value, and substituting the wake into the sea surface infrared radiation model and the atmospheric transmission model, so as to solve the water surface thermal wake temperature map obtained by the space-based detector and perform summarization and classification, and extract image features; (7) combining the shallow underwater moving target data observed by the satellite thermal imager in orbit for comparison and verification analysis, judging whether the error meets the requirements, if not, perfecting the parameters of each model, and if yes, determining the parameters of the turbulence model, the sea surface infrared radiation model and the atmospheric transmission model; (8) performing simulation by using the turbulence model, the sea surface infrared radiation model and the atmospheric transmission model with the determined parameters, and inductively obtaining the water surface thermal wake map corresponding to the underwater moving target at different depths; establishing a perfect simulation model for predicting the depth of the underwater target according to the water surface thermal wake image, and predicting the corresponding depth of the underwater moving target.

2. The method of claim 1, wherein, In step (2), the mesh division includes different types, including overlapping meshes, MRF meshes, sliding meshes and adaptive meshes.

3. The method of claim 1, wherein, In step (3), the control equation includes continuity equation, momentum conservation equation and energy conservation equation; the turbulence model includes Spalart-Allmaras model, k-ε model and k-ω model.

4. The method of claim 1, wherein, In step (4), the solving parameters of the turbulence model are configured and initialized, including multiphase flow setting, region condition setting, boundary condition setting, dynamic mesh setting, solver option setting, convergence condition setting and initialization setting; wherein the multiphase flow includes gas-liquid two-phase flow; the region condition includes background region and target movement region; the boundary condition includes inlet, outlet, model boundary and background region boundary; the dynamic mesh includes overlapping mesh, MRF mesh, sliding mesh and adaptive mesh; the solver option includes control equation selection, turbulence equation selection and data monitoring; the convergence condition includes parameter precision and convergence range; the initialization setting includes inlet velocity and pressure setting of the fluid environment, and velocity and depth setting of the underwater moving target.

5. The method of claim 1, wherein, In step (6), the sea surface infrared radiation model includes a sea surface height field sea spectrum model, a sea surface spontaneous radiation and an environmental reflection model; and the atmospheric transmission model includes an atmospheric absorption, scattering and radiation model.

6. The method of claim 1, wherein, In step (6), the image features include the shape, length, width and temperature gradient characteristics of the thermal wake.

7. The method of claim 1, wherein, In step (7), the model parameters include three-dimensional model parameters, mesh division parameters, turbulence model solving parameters, initialization configuration parameters, sea surface infrared radiation model parameters and atmospheric transmission model parameters.

8. The method of claim 1, wherein, In step (7), the error requirement is: the temperature difference of the hot tail is not more than 10 mK, and the shape, length and width error of the hot tail is not more than 5%.

Images