A laser heat halo effect simulation method and system

Abstract

This application provides a laser thermal coma effect simulation method and system, belonging to the field of laser thermal coma effect simulation technology, including the following steps: dividing the atmospheric space along the laser transmission path into a two-layer structure of macroscopic and microscopic grids; calculating the absorbed laser power density in each macroscopic grid cell based on the initial temperature field at the current moment, and transferring the laser power density in the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid; within the activated microscopic grid region, using the received laser power density as the heat source, solving the heat-fluid coupling equation system to obtain local temperature field data, and returning the local temperature field data to the macroscopic grid node through volume weighted averaging to update the temperature field at the current moment; based on the updated temperature field, using pre-determined spatial basis functions, obtaining the modal coefficients at the current moment through eigenorthogonal decomposition; achieving improved computational efficiency while ensuring physical accuracy.

Description

Technical Field

[0001] This application relates to the field of laser thermal coma effect simulation technology, and in particular to a laser thermal coma effect simulation method and system. Background Technology

[0002] When high-power lasers propagate through the atmosphere, the absorption of laser energy by atmospheric molecules and aerosols causes heating and density changes in the air along the transmission path, which in turn alters the refractive index distribution, resulting in a thermal coma effect. This thermal coma effect leads to laser beam expansion, distortion, and focal shift, severely limiting the laser power density to the target and impacting the effectiveness of engineering applications such as laser anti-drone operations, laser communication, and laser energy transfer.

[0003] Currently, simulation methods for laser thermal coma effects are mainly divided into two categories: one is high-precision numerical simulation methods, such as the finite element method, which obtains relatively accurate temperature and flow field distributions by dividing the fluid region into infinitesimal elements and solving the thermo-fluid coupling equations. However, this type of method involves huge computational loads, with a single simulation typically taking several hours to several days, which cannot meet the real-time requirements of dynamic scenarios such as laser anti-drone operations. The other category is fast calculation methods, such as the phase screen method and the iterative scaling law method, which improve calculation speed by simplifying the physical model or introducing empirical parameters, but often at the cost of accuracy, making it difficult to accurately reflect the dynamic evolution of the thermal coma effect, and its applicability is limited.

[0004] Therefore, the core technical problem of existing laser thermal coma effect simulation technology is that computational efficiency and physical accuracy cannot be balanced. High-precision methods have too much computational load and are difficult to apply in real time, while fast calculation methods are not accurate enough due to oversimplification and cannot meet the needs of actual engineering. Summary of the Invention

[0005] To address the aforementioned issues, this application provides a laser thermal coma effect simulation method and system that can improve computational efficiency while ensuring physical accuracy.

[0006] This application provides a method for simulating laser thermal corona effect, including the following steps: Step S1: Divide the atmospheric space along the laser transmission path into a two-layer structure of macroscopic grid and microscopic grid. The microscopic grid is dynamically activated according to the laser intensity and temperature change rate meeting preset conditions. Step S2: Calculate the absorbed laser power density on each macroscopic grid cell based on the initial temperature field at the current moment, and transfer the laser power density on the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid. Step S3: Within the activated microgrid region, using the received laser power density as the heat source, solve the heat-fluid coupling equations to obtain local temperature field data. Then, return the local temperature field data to the macrogrid nodes using a volume-weighted average method to update the temperature field at the current moment. Step S4: Based on the updated temperature field, the modal coefficients at the current time are obtained through eigenorthogonal decomposition using the pre-determined spatial basis functions; Step S5: Based on the modal coefficients at the current moment, reconstruct the temperature field at the current moment using the spatial basis functions, and calculate the refractive index field at the current moment based on the reconstructed temperature field; Step S6: Using the refractive index field at the current moment as input, calculate the laser transmission process using the step-by-step Fourier method, and output the first thermal focal length at the current moment; Step S7: Construct a physical information neural network, using the modal coefficients, laser parameters, and environmental parameters at the current moment as inputs, to predict the initial modal coefficients at the next moment, and use the spatial basis functions to reconstruct the initial temperature field at the next moment as input to step S2 at the next moment.

[0007] In some embodiments, in step S1, the activation condition of the microgrid is: laser intensity And the rate of temperature change The micromesh is activated at time, where, This represents the maximum laser intensity at the current moment.

[0008] In some embodiments, in step S2, the absorbed laser power density ,in, For laser intensity, The atmospheric absorption coefficient is derived from the molecular absorption coefficient. Aerosol absorption coefficient and nonlinear absorption coefficient Superimposed; the molecular absorption coefficient The aerosol absorption coefficient is calculated based on real-time temperature and pressure and spectral databases, and depends on the temperature field at the current moment. The nonlinear absorption coefficient was calculated based on light scattering theory using aerosol particle size distribution and complex refractive index; Calculations were performed based on a plasma absorption model when the laser power density exceeded the ionization threshold. For macroscopic grid cells containing activated microgrids, the calculated laser power density is transferred to the corresponding microgrid nodes through an interpolation method, serving as the heat source input for fine-grained microscale solutions.

[0009] In some embodiments, step S3 includes: Step S31: Within the activated microgrid region, construct and solve the heat-fluid coupling equations, which include: Energy equation: ; Momentum equation: ; Continuity equation: ; in, For laser power density, For density, For isobaric heat capacity, For temperature, It is a velocity vector. Thermal conductivity, For pressure, For dynamic viscosity, It is the acceleration due to gravity. The coefficient of thermal expansion is For ambient reference temperature; Step S32: The heat-fluid coupling equations are discretized and solved using the finite volume method to obtain local temperature field data on the micro-grid cells. ,in, Index for micro-grid cells; Step S33: Feed back the microscopic temperature information to the macroscopic grid nodes using a volume-weighted average method to update the macroscopic temperature field. ; in, For the new macro grid nodes Temperature value, For macro grid nodes The associated set of micro-grid units, The volume of a micro-grid cell.

[0010] In some embodiments, step S4 includes: Step S41: Construct the snapshot matrix ,in For the first Temperature field vector at time 1 This represents the number of macroscopic grid nodes. Number of snapshots; Step S42: Calculate the average temperature field Singular value decomposition is performed on the snapshot matrix after subtracting the average temperature field. Take the front The left singular vectors are used as basis functions in the space. , Meet the energy ratio ,in It is a singular value; Step S43: Update the temperature field Subtract the average temperature field Then, with each spatial basis function Perform the inner product to obtain the modal coefficients at the current time step. .

[0011] In some embodiments, step S5 includes: Step S51: Based on the modal coefficients at the current time and space basis functions Reconstruct the temperature field at the current moment: ; Step S52: Based on the reconstructed temperature field The refractive index field at the current moment is calculated using the Edren formula: ; in, For refractive index, For pressure, For the reconstructed temperature, For water vapor partial pressure, This is due to refractive index fluctuations caused by turbulence. This refers to the refractive index change caused by laser-induced plasma.

[0012] In some embodiments, step S6 includes: Step S61: Divide the laser transmission path into multiple thin layers, each with a thickness of... satisfy ,in The radius of the beam. The wavelength of the laser; Step S62: Perform vacuum diffraction and phase modulation calculations for each layer sequentially; multiply the vacuum diffraction by the transfer function in the frequency domain. ,in For spatial frequency, The imaginary unit; phase modulation multiplied by the phase screen in the spatial domain ,in For phase delay, For wave number, The refractive index; Step S63: Repeat until the transmission endpoint to obtain the far-field light intensity distribution. Perform Zernike polynomial fitting on the phase distortion of the far-field light intensity distribution, extract the defocus term coefficients, and calculate the first thermal focal length at the current moment based on the defocus term coefficients. (t).

[0013] In some embodiments, step S7 includes: Step S71: Construct a physical information neural network, the input of which includes: Modal coefficients at the current moment ; Laser parameters: laser power The radius of the light spot at the current moment Time step ; Environmental parameters: wind speed vector Atmospheric pressure Water vapor partial pressure turbulence intensity ; The output is the initial mode coefficients for the next time step. ; Step S72: Construct the loss function for the physical information neural network: ; in: For data fitting terms, For PDE residuals, For energy conservation terms, , These are weight parameters; Step S73: After predicting the initial modal coefficients for the next time step, reconstruct the initial temperature field for the next time step using spatial basis functions: ; The initial temperature field for the next time step will be input into step S2.

[0014] In some embodiments, step S8 is also included: Step S81: Based on the initial mode coefficients predicted by the physical information neural network for the next moment, the second thermal focal length for the next moment is calculated through steps S5 and S6. ; Step S82: Determine the second thermal focal length With the first thermal focal length absolute difference When the absolute difference Greater than the preset threshold When this happens, it is determined that the deviation exceeds the standard; Step S83: In response to the deviation exceeding the limit, the parameters of the physical information neural network are corrected online using an extended Kalman filter, wherein the state vector is the modal coefficient vector. Observation vector The Zernike coefficients are extracted from measured wavefront data; Step S84: Calculate the cumulative deviation index ,in To map the modal coefficients to the Zernike coefficients in the observation matrix, For the integration time window, For the observation vector at time... The value, The modal coefficient vector at time t The value; when And the duration exceeds At each time step, a spatial basis function update is triggered, where... The preset threshold; Step S85: Collect the latest measured temperature field data, construct a new snapshot matrix, and perform singular value decomposition again to update the spatial basis functions. and average temperature field .

[0015] This invention provides a laser thermal coma effect simulation system, comprising: The grid generation module is used to divide the atmospheric space along the laser transmission path into a two-layer structure of macro-grid and micro-grid. The micro-grid is dynamically activated according to the laser intensity and temperature change rate meeting preset conditions. The laser power density calculation module is used to calculate the absorbed laser power density on each macroscopic grid cell based on the initial temperature field at the current moment, and to transfer the laser power density on the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid. The microgrid solving module is used to solve the heat-fluid coupling equations within the activated microgrid region, using the received laser power density as the heat source, to obtain local temperature field data. The local temperature field data is then returned to the macrogrid nodes through a volume-weighted average to update the temperature field at the current moment. The intrinsic orthogonal decomposition module is used to obtain the modal coefficients at the current time through intrinsic orthogonal decomposition based on the updated temperature field and using pre-determined spatial basis functions. The reconstruction module is used to reconstruct the temperature field at the current moment based on the modal coefficients at the current moment using the spatial basis functions, and to calculate the refractive index field at the current moment based on the reconstructed temperature field. The wavefront propagation simulation module uses the current refractive index field as input, employs a split-step Fourier method to calculate the laser propagation process, and outputs the first thermal focal length at the current moment. The physical information neural network module is used to construct a physical information neural network. It uses the modal coefficients, laser parameters, and environmental parameters at the current moment as inputs to predict the initial modal coefficients at the next moment. It also uses the spatial basis functions to reconstruct the initial temperature field at the next moment, which serves as the input to the laser power density calculation module at the next moment.

[0016] Compared with the prior art, the present invention has the following beneficial effects: By using a two-layer partitioning of macro-grids and micro-grids, high-resolution micro-grids are used to accurately analyze small-scale physical processes in areas with significant thermal halo effects, while macro-grids are used in other areas to reduce computational load. Combined with intrinsic orthogonal decomposition dimensionality reduction technology, high-dimensional partial differential equations are transformed into low-dimensional ordinary differential equations, reducing computational load while preserving the dominant physical processes, thus achieving an optimized balance between computational efficiency and accuracy. Attached Figure Description

[0017] The embodiments of the present invention will be further described below with reference to the accompanying drawings: Figure 1 A schematic diagram illustrating the implementation process of a laser thermal coma effect simulation method provided in this application embodiment; Figure 2 A schematic diagram of a laser thermal coma effect simulation system provided in this application embodiment; Figure 3 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0020] If the application documents contain similar descriptions such as "first, second, third", the following explanation shall be added: In the following description, the terms "first, second, third" are used only to distinguish similar objects and do not represent a specific order of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in an order other than that illustrated or described herein.

[0021] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0022] To address the problems existing in related technologies, this application provides a method for simulating laser thermal coma effects. The subject executing this simulation method can be an electronic device. The electronic device can be various types of terminals such as laptops, tablets, desktop computers, set-top boxes, and mobile devices (e.g., mobile phones, portable music players, personal digital assistants, dedicated messaging devices, portable gaming devices), or it can be implemented as a server. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0023] In some embodiments, the functions implemented by the simulation method provided in this application can be achieved by the processor of an electronic device calling program code, wherein the program code can be stored in a computer storage medium.

[0024] This application provides a method for simulating laser thermal corona effect. Figure 1 This is a schematic diagram illustrating the implementation process of a laser thermal coma effect simulation method provided in an embodiment of this application, as shown below. Figure 1 As shown, it includes the following steps: Step S1: Divide the atmospheric space along the laser transmission path into a two-layer structure of macroscopic grid and microscopic grid. The microscopic grid is dynamically activated according to the laser intensity and temperature change rate meeting preset conditions. Step S2: Calculate the absorbed laser power density on each macroscopic grid cell based on the initial temperature field at the current moment, and transfer the laser power density on the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid. Step S3: Within the activated microgrid region, using the received laser power density as the heat source, solve the heat-fluid coupling equations to obtain local temperature field data. Then, return the local temperature field data to the macrogrid nodes using a volume-weighted average method to update the temperature field at the current moment. Step S4: Based on the updated temperature field, the modal coefficients at the current time are obtained through eigenorthogonal decomposition using the pre-determined spatial basis functions; Step S5: Based on the modal coefficients at the current moment, reconstruct the temperature field at the current moment using the spatial basis functions, and calculate the refractive index field at the current moment based on the reconstructed temperature field; Step S6: Using the refractive index field at the current moment as input, calculate the laser transmission process using the step-by-step Fourier method, and output the first thermal focal length at the current moment; Step S7: Construct a physical information neural network, using the modal coefficients, laser parameters, and environmental parameters at the current moment as inputs, to predict the initial modal coefficients at the next moment, and use the spatial basis functions to reconstruct the initial temperature field at the next moment as input to step S2 at the next moment.

[0025] In this embodiment of the invention, before the simulation begins, an initial temperature field needs to be constructed as the starting point for the entire time-stepping simulation. The initial temperature field is constructed as follows: Measured atmospheric temperature data at the initial moment are collected, including ground meteorological station temperatures, radiosonde vertical temperature profiles, and lidar temperature data. Kriging interpolation is used to map discrete measured temperature points onto macroscopic grid nodes. The interpolation results are checked for reasonableness, outliers are removed, and the initial temperature field is ensured to be within a physically reasonable range. The constructed initial temperature field is used as the input for step S2, and the time-stepping simulation begins. Step S1 first divides the computational domain into a two-layer structure: a macroscopic grid and a microscopic grid. The macroscopic grid covers the entire transmission path, while the microscopic grid is dynamically activated only in areas with significant thermal halo effects (where laser intensity is high and temperature changes drastically). This layered strategy allows computational resources to be focused on key areas, avoiding the huge overhead of a full-domain high-resolution grid. Step S2 calculates the absorbed laser power density based on the current temperature field and transfers the heat source on the macroscopic grid cell containing the activated microgrid to the corresponding microgrid, realizing the transfer of macroscopic heat source information to the microscale. Step S3 obtains the current temperature field through fine solving of the microgrid. Steps S4 and S5 do not update the temperature field again, but perform feature extraction and data transformation based on the temperature field: Step S4 projects it to a low-dimensional modal coefficient space through intrinsic orthogonal decomposition to obtain modal coefficients; Step S5 reconstructs the temperature field using the modal coefficients and calculates the refractive index field based on the reconstruction result, providing necessary input parameters for subsequent wavefront transmission (step S6) and future time prediction (step S7). This process retains the high-precision temperature field information and achieves dimensionality reduction representation, laying the foundation for subsequent rapid prediction. Step S6 uses the refractive index field at the current time as input, uses the step-by-step Fourier method to calculate the laser transmission process, obtains the first thermal focal length at the current time by fitting the quadratic term of phase distortion, and outputs the key engineering indicators of thermal halo effect. By directly outputting the thermal focal length, a key engineering indicator, laser system designers can assess the impact of thermal corona and optimize emission parameters. Step S7 constructs a physical information neural network, using the current modal coefficients, laser parameters, and environmental parameters as inputs to predict the initial modal coefficients for the next time step. The loss function of this neural network integrates data fitting terms, dimensionality reduction equation residual terms, and energy conservation terms, ensuring that the prediction results conform to both data patterns and physical laws. After prediction, the initial temperature field for the next time step is reconstructed using spatial basis functions, serving as the initial temperature field for the next time step in step S2, thus achieving rapid time progression in a low-dimensional space.

[0026] In some embodiments, in step S1, the activation condition of the microgrid is: laser intensity And the rate of temperature change The micromesh is activated at time, where, This represents the maximum laser intensity at the current moment.

[0027] In this embodiment of the invention, sufficiently high laser intensity is required to generate significant heating, and a sufficiently large rate of temperature change indicates that the thermal halo effect is developing rapidly. When both conditions are met simultaneously, the microgrid is activated, ensuring that computational resources are focused on the spatiotemporal region where the thermal halo effect is most pronounced.

[0028] In some embodiments, based on turbulent dissipation rate Dynamically adjust the micromesh size To satisfy ,in Kolmogorov scale, For kinematic viscosity; simultaneously, based on laser intensity gradient and wind speed gradient Perform adaptive encryption when or At that time, the grid cell is refined.

[0029] In this embodiment of the invention, the Kolmogorov scale is the characteristic scale of the smallest eddy in turbulence, representing the smallest spatial scale of temperature fluctuations. Limiting the microgrid size to below the Kolmogorov scale ensures that the microgrid can resolve small-scale temperature fluctuations caused by turbulence, which is crucial for accurately simulating the turbulence-thermal co-existence effect in thermal coma. Regions with large laser intensity gradients indicate drastic changes in light intensity distribution and uneven heat source distribution; regions with large wind speed gradients indicate strong flow shear and complex convective heat transfer. Refining the mesh in these two types of regions improves the accuracy of heat source calculations and flow field simulations.

[0030] In some embodiments, in step S2, the absorbed laser power density ,in, For laser intensity, The atmospheric absorption coefficient is derived from the molecular absorption coefficient. Aerosol absorption coefficient and nonlinear absorption coefficient Superimposed; the molecular absorption coefficient The aerosol absorption coefficient is calculated based on real-time temperature and pressure and spectral databases, and depends on the temperature field at the current moment. The nonlinear absorption coefficient was calculated based on light scattering theory using aerosol particle size distribution and complex refractive index; Calculations were performed based on a plasma absorption model when the laser power density exceeded the ionization threshold. For macroscopic grid cells containing activated microgrids, the calculated laser power density is transferred to the corresponding microgrid nodes through an interpolation method, serving as the heat source input for fine-grained microscale solutions.

[0031] In this embodiment of the invention, the laser power density is calculated based on the differential form of the Beer-Lambert law: The absorption coefficient It integrates three physical mechanisms: molecular absorption, aerosol absorption, and nonlinear absorption. Molecular absorption coefficient. Calculated based on real-time temperature and pressure and spectral databases (such as HITRAN), and dependent on the current temperature field because molecular spectral line intensities are temperature-dependent. Aerosol absorption coefficient. The absorption characteristics of laser energy by aerosols are calculated using Mie scattering theory based on aerosol particle size distribution and complex refractive index. The nonlinear absorption coefficient is also included. Activated when the laser power density exceeds the ionization threshold, the energy deposition mechanism after strong laser-induced ionization is characterized based on plasma absorption models (such as inverse bremsstrahlung absorption). For macroscopic grid cells containing activated microgrids, the calculated laser power density is transferred to the corresponding microgrid nodes via interpolation, serving as the heat source input for fine-grained microscale solutions. This transfer ensures that the microgrid can perform fine-grained solutions based on heat sources calculated at the macroscopic scale, forming a flow of heat source information between the macroscopic and microscopic scales.

[0032] In some embodiments, step S3 includes: Step S31: Within the activated microgrid region, construct and solve the heat-fluid coupling equations, which include: Energy equation: ; Momentum equation: ; Continuity equation: ; in, For laser power density, For density, For isobaric heat capacity, For temperature, It is a velocity vector. Thermal conductivity, For pressure, For dynamic viscosity, It is the acceleration due to gravity. The coefficient of thermal expansion is For ambient reference temperature; Step S32: The heat-fluid coupling equations are discretized and solved using the finite volume method to obtain local temperature field data on the micro-grid cells. ,in, Index for micro-grid cells; Step S33: Feed back the microscopic temperature information to the macroscopic grid nodes using a volume-weighted average method to update the macroscopic temperature field. ; in, For the new macro grid nodes Temperature value, For macro grid nodes The associated set of micro-grid units, The volume of a micro-grid cell.

[0033] In this embodiment of the invention, the energy equation describes the transport and dissipation of thermal energy: the left side represents the time-varying rate of temperature and the convection term, while the right side represents the heat conduction term and the laser heat source term. This equation converts the energy absorbed by the laser into a temperature increase, which propagates through the air via heat conduction and convection. The momentum equation uses the Boussinesq approximation to consider the buoyancy effect: the left side represents the time-varying rate of velocity and the convection term, while the right side represents the pressure gradient term, the viscous dissipation term, and the buoyancy term. (Buoyancy term...) This demonstrates the impact of density changes caused by temperature differences on flow, which is the driving force of natural convection. The continuity equation ensures mass conservation and adopts an incompressible form for low-velocity flows. After solving the local temperature field using a microgrid, a volume-weighted average is returned to the macrogrid nodes: the temperature value of each macrogrid node is equal to the volume-weighted average temperature of its associated microgrid cells. This weighted averaging method ensures energy conservation, meaning the total thermal energy of the microregion is equal to the total thermal energy of the region represented by the returned macrogrid node.

[0034] In some embodiments, step S4 includes: Step S41: Construct the snapshot matrix ,in For the first Temperature field vector at time 1 This represents the number of macroscopic grid nodes. Number of snapshots; Step S42: Calculate the average temperature field Singular value decomposition is performed on the snapshot matrix after subtracting the average temperature field. Take the front The left singular vectors are used as basis functions in the space. , Meet the energy ratio ,in It is a singular value; Step S43: Update the temperature field Subtract the average temperature field Then, with each spatial basis function Perform the inner product to obtain the modal coefficients at the current time step. .

[0035] In this embodiment of the invention, a snapshot matrix is ​​first constructed. ,collect Temperature field vectors at typical moments, each vector length being the number of macroscopic grid nodes. Calculate the average temperature field. The pulsating temperature field matrix is ​​obtained by subtracting it from the snapshot matrix. Singular value decomposition (SVD) is then performed on the pulsating temperature field matrix to obtain the left singular vector matrix. Singular value diagonal matrix and right singular vector matrix .forward The left singular vectors are the basis functions of the space. , The selection is based on the energy percentage criterion: [Previous] The sum of squares of the singular values ​​must account for at least 99% of the total sum of squares of singular values ​​to ensure that the basis functions can capture more than 99% of the energy (variance) of the temperature field. After obtaining the spatial basis functions, the average temperature field is subtracted from the current temperature field, and then the inner product is taken with each spatial basis function to obtain the modal coefficients at the current time. This process essentially projects a high-dimensional temperature field onto a low-dimensional basis function space, realizing the transfer of temperature data from grid nodes (…). (dimensional) to modal coefficients ( Dimensionality reduction representation of (dimensionality).

[0036] In some embodiments, step S5 includes: Step S51: Based on the modal coefficients at the current time and space basis functions Reconstruct the temperature field at the current moment: ; Step S52: Based on the reconstructed temperature field The refractive index field at the current moment is calculated using the Edren formula: ; in, For refractive index, For pressure, For the reconstructed temperature, For water vapor partial pressure, This is due to refractive index fluctuations caused by turbulence. This refers to the refractive index change caused by laser-induced plasma.

[0037] In this embodiment of the invention, temperature field reconstruction is the inverse operation of the dimensionality reduction process: the modal coefficients at the current moment are... with space basis functions Linear combination, plus average temperature field To obtain the high-dimensional temperature field at the current moment. This reconstruction restores the low-dimensional abstract representation to a high-dimensional temperature field with clear physical meaning, providing input for subsequent physical calculations. The refractive index field is calculated using the Edren formula, a universally accepted empirical formula for the air refractive index in optics and atmospheric transmission. The first term on the right-hand side of the formula is the vacuum refractive index, the second is the contribution from dry air (related to pressure and temperature), the third is the contribution from water vapor (related to water vapor partial pressure and temperature), the fourth is the refractive index fluctuation caused by turbulence (generated by the Kolmogorov spectral model), and the fifth is the refractive index change caused by laser-induced plasma (activated only above the ionization threshold). The Edren formula directly correlates the temperature field with the refractive index field, reflecting the influence of temperature changes on the optical properties of the light transmission medium.

[0038] In some embodiments, step S6 includes: Step S61: Divide the laser transmission path into multiple thin layers, each with a thickness of... satisfy ,in The radius of the beam. The wavelength of the laser; Step S62: Perform vacuum diffraction and phase modulation calculations for each layer sequentially; multiply the vacuum diffraction by the transfer function in the frequency domain. ,in For spatial frequency, The imaginary unit; phase modulation multiplied by the phase screen in the spatial domain ,in For phase delay, For wave number, The refractive index; Step S63: Repeat until the transmission endpoint to obtain the far-field light intensity distribution. Perform Zernike polynomial fitting on the phase distortion of the far-field light intensity distribution, extract the defocus term coefficients, and calculate the first thermal focal length at the current moment based on the defocus term coefficients. (t).

[0039] In this embodiment of the invention, the split-step Fourier method divides the laser transmission path into multiple thin layers, each with a thickness of... satisfy This condition ensures that the diffraction effect within each layer can be approximated as vacuum diffraction, while phase accumulation can be approximated as a thin-screen approximation. Two calculations are performed sequentially for each layer: first, vacuum diffraction calculations are performed, transforming the light field to the frequency domain and multiplying it by the transfer function. Then, the transformation is performed back to the spatial domain. This operation simulates the diffraction and propagation of light waves in free space. Phase modulation calculations are then performed, and the phase delay is calculated based on the refractive index distribution of this layer. Multiply by the phase screen in the airspace The phase distortion caused by refractive index inhomogeneity was simulated. Layer-by-layer calculations were repeated until the transmission endpoint to obtain the far-field light intensity distribution. Zernike polynomial fitting was performed on the phase distortion of the far-field light intensity distribution to extract the defocus term coefficients, and the first thermal focal length at the current moment was calculated based on these defocus term coefficients. (t).

[0040] In some embodiments, step S7 includes: Step S71: Construct a physical information neural network, the input of which includes: Modal coefficients at the current moment ; Laser parameters: laser power The radius of the light spot at the current moment Time step ; Environmental parameters: wind speed vector Atmospheric pressure Water vapor partial pressure turbulence intensity ; The output is the initial mode coefficients for the next time step. ; Step S72: Construct the loss function for the physical information neural network: ; in: For data fitting terms, For PDE residuals, For energy conservation terms, , These are weight parameters; Step S73: After predicting the initial modal coefficients for the next time step, reconstruct the initial temperature field for the next time step using spatial basis functions: ; The initial temperature field for the next time step will be input into step S2.

[0041] In this embodiment of the invention, the input to the neural network includes three types of information: the modal coefficients at the current moment (representing the current temperature field state), laser parameters (laser power, spot radius, time step), and environmental parameters (wind speed vector, atmospheric pressure, water vapor partial pressure, turbulence intensity). The output is the initial modal coefficients for the next moment. This input-output design enables the network to predict future states based on the current state and external conditions, achieving an end-to-end mapping from data to prediction. The loss function consists of three parts: a data fitting term... PDE residuals and energy conservation terms The data fitting term measures the deviation between the network's predicted values ​​and the actual training data, reflecting the data-driven nature of neural networks. The PDE residual term is obtained by substituting the predicted mode coefficients into the system of ordinary differential equations obtained after dimensionality reduction of the heat conduction equation using Galerkin projection. ,in , Given the laser spot radius, calculate the residuals to force the network predictions to satisfy the dimensionality-reduced physical equations. The energy conservation term calculates the deviation from the global energy conservation, ensuring the prediction results satisfy the balance between laser energy absorption, heat storage from temperature rise, and convective heat dissipation. and The introduction of this technology transforms the network into a "physical information" neural network, combining the advantages of both data-driven and physics-driven approaches. After predicting the initial modal coefficients for the next time step, the temperature field for that next time step is reconstructed using spatial basis functions, providing the initial temperature field for the simulation at the next time step and achieving a closed loop for time progression.

[0042] In some embodiments, step S8 is also included: Step S81: Based on the initial mode coefficients predicted by the physical information neural network for the next moment, the second thermal focal length for the next moment is calculated through steps S5 and S6. ; Step S82: Determine the second thermal focal length With the first thermal focal length absolute difference When the absolute difference Greater than the preset threshold When this happens, it is determined that the deviation exceeds the standard; Step S83: In response to the deviation exceeding the limit, the parameters of the physical information neural network are corrected online using an extended Kalman filter, wherein the state vector is the modal coefficient vector. Observation vector The Zernike coefficients are extracted from measured wavefront data; Step S84: Calculate the cumulative deviation index ,in To map the modal coefficients to the Zernike coefficients in the observation matrix, For the integration time window, For the observation vector at time... The value, The modal coefficient vector at time t The value; when And the duration exceeds At each time step, a spatial basis function update is triggered, where... The preset threshold; Step S85: Collect the latest measured temperature field data, construct a new snapshot matrix, and perform singular value decomposition again to update the spatial basis functions. and average temperature field .

[0043] In this embodiment of the invention, the second thermal focal length predicted by the physical information neural network is compared with the first thermal focal length obtained through high-precision simulation at the current moment. This is because the time step is short. Within the thermal focal length, changes are typically gradual. The high-precision calculation result at the current moment can serve as an approximate reference for the predicted value at the next moment. Therefore, when the deviation exceeds a threshold, it indicates that the neural network's prediction has deviated from the expected physical trend, requiring online correction. Online correction involves absorbing measured wavefront data in real time and dynamically adjusting the neural network parameters using an extended Kalman filter to make the model output approximate the actual physical process. The extended Kalman filter can fuse model predictions and observational data, providing the optimal state estimate in the sense of minimum variance, and simultaneously correcting model parameters. Spatial basis functions are fixed patterns extracted from offline snapshot data through intrinsic orthogonal decomposition, representing the dominant morphology of the temperature field. When environmental conditions change drastically (such as sudden changes in wind speed, significant adjustments in laser power, or the emergence of new atmospheric phenomena), the original basis functions may fail to effectively represent the new temperature field distribution, leading to distortion of the dimensionality-reduced model. In this case, even fine-tuning the neural network parameters is insufficient to compensate for the lack of representational power. Therefore, it is necessary to re-perform intrinsic orthogonal decomposition to extract basis functions from the latest measured data that better characterize the current physical process, fundamentally restoring the model's representational capability. Thermal focal length is a macroscopic indicator of the thermal halo effect, and its changes directly reflect the accuracy of model predictions. A threshold of 1%-5% of the first thermal focal length can be set to avoid frequent triggering caused by noise while ensuring timely response to significant deviations. Using an integral form of accumulated deviation smooths out single-point noise and more reliably reflects the model's systematic deviations. When the limits are consistently exceeded, it indicates that the model can no longer adapt to environmental changes and the basis functions need to be updated. Duration This ensures that basis function updates are triggered only when the bias actually persists, avoiding false triggers caused by transient perturbations. After the new basis function is generated, the input-output space of the neural network changes (the physical meaning of the modal coefficients changes). Fine-tuning training with a low learning rate allows for rapid adaptation to the new basis function representation while retaining the original network knowledge, achieving a smooth transition of the model.

[0044] Based on the foregoing embodiments, this application provides a laser thermal coma effect simulation system. The modules and units included in the system can be implemented by a processor in a computer device; of course, they can also be implemented by specific logic circuits. In the implementation process, the processor can be a central processing unit (CPU), a microprocessor (MPU), a digital signal processor (DSP), or a field programmable gate array (FPGA), etc.

[0045] This application provides a laser thermal coma effect simulation system. Figure 2 A schematic diagram of a laser thermal coma effect simulation system provided in this application embodiment is shown below. Figure 2 As shown, it includes: The grid generation module is used to divide the atmospheric space along the laser transmission path into a two-layer structure of macro-grid and micro-grid. The micro-grid is dynamically activated according to the laser intensity and temperature change rate meeting preset conditions. The laser power density calculation module is used to calculate the absorbed laser power density on each macroscopic grid cell based on the initial temperature field at the current moment, and to transfer the laser power density on the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid. The microgrid solving module is used to solve the heat-fluid coupling equations within the activated microgrid region, using the received laser power density as the heat source, to obtain local temperature field data. The local temperature field data is then returned to the macrogrid nodes through a volume-weighted average to update the temperature field at the current moment. The intrinsic orthogonal decomposition module is used to obtain the modal coefficients at the current time through intrinsic orthogonal decomposition based on the updated temperature field and using pre-determined spatial basis functions. The reconstruction module is used to reconstruct the temperature field at the current moment based on the modal coefficients at the current moment using the spatial basis functions, and to calculate the refractive index field at the current moment based on the reconstructed temperature field. The wavefront propagation simulation module uses the current refractive index field as input, employs a split-step Fourier method to calculate the laser propagation process, and outputs the first thermal focal length at the current moment. The physical information neural network module is used to construct a physical information neural network. It uses the modal coefficients, laser parameters, and environmental parameters at the current moment as inputs to predict the initial modal coefficients at the next moment. It also uses the spatial basis functions to reconstruct the initial temperature field at the next moment, which serves as the input to the laser power density calculation module at the next moment.

[0046] In some embodiments, the activation condition for the microgrid in the mesh generation module is: laser intensity. And the rate of temperature change The micromesh is activated at time, where, This represents the maximum laser intensity at the current moment.

[0047] In some embodiments, the absorbed laser power density in the laser power density calculation module ,in, For laser intensity, The atmospheric absorption coefficient is derived from the molecular absorption coefficient. Aerosol absorption coefficient and nonlinear absorption coefficient Superimposed; the molecular absorption coefficient The aerosol absorption coefficient is calculated based on real-time temperature and pressure and spectral databases, and depends on the temperature field at the current moment. The nonlinear absorption coefficient was calculated based on light scattering theory using aerosol particle size distribution and complex refractive index; Calculations were performed based on a plasma absorption model when the laser power density exceeded the ionization threshold. For macroscopic grid cells containing activated microgrids, the calculated laser power density is transferred to the corresponding microgrid nodes through an interpolation method, serving as the heat source input for fine-grained microscale solutions.

[0048] In some embodiments, the microgrid solving module includes: Solver cells are used to construct and solve a set of heat-fluid coupling equations within an activated microgrid region, the set of equations including: Energy equation: ; Momentum equation: ; Continuity equation: ; in, For laser power density, For density, For isobaric heat capacity, For temperature, It is a velocity vector. Thermal conductivity, For pressure, For dynamic viscosity, It is the acceleration due to gravity. The coefficient of thermal expansion is For ambient reference temperature; Discrete solution elements are used to discretely solve the thermal-fluid coupled equations using the finite volume method, obtaining local temperature field data on the micro-grid elements. ,in, Index for micro-grid cells; The update unit is used to feed back microscopic temperature information to macroscopic grid nodes through volume-weighted averaging, thereby updating the macroscopic temperature field. ; in, For the new macro grid nodes Temperature value, For macro grid nodes The associated set of micro-grid units, The volume of a micro-grid cell.

[0049] In some embodiments, the intrinsic orthogonal decomposition module includes: The first building unit is used to construct the snapshot matrix. ,in For the first Temperature field vector at time 1 This represents the number of macroscopic grid nodes. Number of snapshots; The first calculation unit is used to calculate the average temperature field. Singular value decomposition is performed on the snapshot matrix after subtracting the average temperature field. Take the front The left singular vectors are used as basis functions in the space. , Meet the energy ratio ,in It is a singular value; Inner product elements are used to update the temperature field. Subtract the average temperature field Then, with each spatial basis function Perform the inner product to obtain the modal coefficients at the current time step. .

[0050] In some embodiments, the reconstruction module includes: The first reconstruction unit is used to reconstruct the modal coefficients based on the current time step. and space basis functions Reconstruct the temperature field at the current moment: ; The second computational unit is used to calculate the temperature field obtained from the reconstruction. The refractive index field at the current moment is calculated using the Edren formula: ; in, For refractive index, For pressure, For the reconstructed temperature, For water vapor partial pressure, This is due to refractive index fluctuations caused by turbulence. This refers to the refractive index change caused by laser-induced plasma.

[0051] In some embodiments, the wavefront transmission simulation module includes: Dividing units are used to divide the laser transmission path into multiple thin layers, each with a thickness of [missing information]. satisfy ,in The radius of the beam. The wavelength of the laser; The third computational unit performs vacuum diffraction and phase modulation calculations sequentially for each layer; the vacuum diffraction is multiplied by the transfer function in the frequency domain. ,in For spatial frequency, The imaginary unit; phase modulation multiplied by the phase screen in the spatial domain ,in For phase delay, For wave number, The refractive index; The fourth calculation unit is used to repeat the process until the transmission endpoint to obtain the far-field light intensity distribution. It then performs Zernike polynomial fitting on the phase distortion of the far-field light intensity distribution, extracts the defocus term coefficients, and calculates the first thermal focal length at the current moment based on these coefficients. (t).

[0052] In some embodiments, the physical information neural network module includes: The second building unit is used to construct the physical information neural network, and its input includes: Modal coefficients at the current moment ; Laser parameters: laser power The radius of the light spot at the current moment Time step ; Environmental parameters: wind speed vector Atmospheric pressure Water vapor partial pressure turbulence intensity ; The output is the initial mode coefficients for the next time step. ; The third building unit is used to construct the loss function of the physical information neural network: ; in: For data fitting terms, For PDE residuals, For energy conservation terms, , These are weight parameters; The second reconstruction unit is used to reconstruct the initial temperature field at the next time step after predicting the initial modal coefficients at the next time step using spatial basis functions. ; The initial temperature field for the next time step will be input into step S2.

[0053] In some embodiments, a self-calibration module is also included: The fifth calculation unit is used to calculate the second thermal focal length at the next moment based on the initial mode coefficients predicted by the physical information neural network, through the reconstruction module and the wavefront propagation simulation module. ; Determining unit, used to determine the second thermal focal length With the first thermal focal length absolute difference When the absolute difference Greater than the preset threshold When this happens, it is determined that the deviation exceeds the standard; A correction unit is used to perform online correction of the parameters of the physical information neural network using an extended Kalman filter in response to the deviation exceeding the limit, wherein the state vector is the modal coefficient vector. Observation vector The Zernike coefficients are extracted from measured wavefront data; The sixth calculation unit is used to calculate the cumulative deviation index. ,in To map the modal coefficients to the Zernike coefficients in the observation matrix, For the integration time window, For the observation vector at time... The value, The modal coefficient vector at time t The value; when And the duration exceeds At each time step, a spatial basis function update is triggered, where... The preset threshold; The update unit is used to collect the latest measured temperature field data, construct a new snapshot matrix, and perform singular value decomposition again to update the spatial basis functions. and average temperature field .

[0054] It should be noted that, in the embodiments of this application, if the above simulation method is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), magnetic disks, or optical disks. Thus, the embodiments of this application are not limited to any specific hardware and software combination.

[0055] Accordingly, this application provides a storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the steps in the simulation method provided in the above embodiments.

[0056] This application provides an electronic device; Figure 3 This is a schematic diagram of the composition structure of the electronic device provided in the embodiments of this application, such as... Figure 3 As shown, the electronic device 400 includes: a processor 401, at least one communication bus 402, a user interface 403, at least one external communication interface 404, and a memory 405. The communication bus 402 is configured to enable communication between these components. The user interface 403 may include a display screen, and the external communication interface 404 may include standard wired and wireless interfaces. The processor 401 is configured to execute a program of a simulation method stored in the memory to implement the steps of the simulation method provided in the above embodiment.

[0057] It should be noted that the descriptions of the storage media and electronic device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage media and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0058] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above-described processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above-described embodiments are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0059] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, object, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, object, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, object, or apparatus that includes that element.

[0060] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0061] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0062] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0063] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0064] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a controller to execute all or part of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0065] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for simulating laser thermal corona effect, characterized in that, Includes the following steps: Step S1: Divide the atmospheric space along the laser transmission path into a two-layer structure of macroscopic grid and microscopic grid. The microscopic grid is dynamically activated according to the laser intensity and temperature change rate meeting preset conditions. Step S2: Calculate the absorbed laser power density on each macroscopic grid cell based on the initial temperature field at the current moment, and transfer the laser power density on the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid. Step S3: Within the activated microgrid region, using the received laser power density as the heat source, solve the heat-fluid coupling equations to obtain local temperature field data. Then, return the local temperature field data to the macrogrid nodes using a volume-weighted average method to update the temperature field at the current moment. Step S4: Based on the updated temperature field, the modal coefficients at the current time are obtained through eigenorthogonal decomposition using the pre-determined spatial basis functions; Step S5: Based on the modal coefficients at the current moment, reconstruct the temperature field at the current moment using the spatial basis functions, and calculate the refractive index field at the current moment based on the reconstructed temperature field; Step S6: Using the refractive index field at the current moment as input, calculate the laser transmission process using the step-by-step Fourier method, and output the first thermal focal length at the current moment; Step S7: Construct a physical information neural network, using the modal coefficients, laser parameters, and environmental parameters at the current moment as inputs, to predict the initial modal coefficients at the next moment, and use the spatial basis functions to reconstruct the initial temperature field at the next moment as input to step S2 at the next moment.

2. The method according to claim 1, characterized in that, In step S1, the activation condition for the microgrid is: laser intensity. And the rate of temperature change The micromesh is activated at time, where, This represents the maximum laser intensity at the current moment.

3. The method according to claim 1, characterized in that, In step S2, the absorbed laser power density ,in, For laser intensity, The atmospheric absorption coefficient is derived from the molecular absorption coefficient. Aerosol absorption coefficient and nonlinear absorption coefficient Superimposed; the molecular absorption coefficient The aerosol absorption coefficient is calculated based on real-time temperature and pressure and spectral databases, and depends on the temperature field at the current moment. The nonlinear absorption coefficient was calculated based on light scattering theory using aerosol particle size distribution and complex refractive index; Calculations were performed based on a plasma absorption model when the laser power density exceeded the ionization threshold. For macroscopic grid cells containing activated microgrids, the calculated laser power density is transferred to the corresponding microgrid nodes through an interpolation method, serving as the heat source input for fine-grained microscale solutions.

4. The method according to claim 1, characterized in that, Step S3 includes: Step S31: Within the activated microgrid region, construct and solve the heat-fluid coupling equations, which include: Energy equation: ； Momentum equation: ； Continuity equation: ； in, For laser power density, For density, For isobaric heat capacity, For temperature, It is a velocity vector. Thermal conductivity, For pressure, For dynamic viscosity, It is the acceleration due to gravity. The coefficient of thermal expansion is For ambient reference temperature; Step S32: The heat-fluid coupling equations are discretized and solved using the finite volume method to obtain local temperature field data on the micro-grid cells. ,in, Index for micro-grid cells; Step S33: Feed back the microscopic temperature information to the macroscopic grid nodes using a volume-weighted average method to update the macroscopic temperature field. ； in, For the new macro grid nodes Temperature value, For macro grid nodes The associated set of micro-grid units, The volume of a micro-grid cell.

5. The method according to claim 1, characterized in that, Step S4 includes: Step S41: Construct the snapshot matrix ,in For the first Temperature field vector at time 1 This represents the number of macroscopic grid nodes. Number of snapshots; Step S42: Calculate the average temperature field Singular value decomposition is performed on the snapshot matrix after subtracting the average temperature field. Take the front The left singular vectors are used as basis functions in the space. , Meet the energy ratio ,in It is a singular value; Step S43: Update the temperature field Subtract the average temperature field Then, with each spatial basis function Perform the inner product to obtain the modal coefficients at the current time step. .

6. The method according to claim 5, characterized in that, Step S5 includes: Step S51: Based on the modal coefficients at the current time and space basis functions Reconstruct the temperature field at the current moment: ； Step S52: Based on the reconstructed temperature field The refractive index field at the current moment is calculated using the Edren formula: ； in, For refractive index, For pressure, For the reconstructed temperature, For water vapor partial pressure, This is due to refractive index fluctuations caused by turbulence. This refers to the refractive index change caused by laser-induced plasma.

7. The method according to claim 1, characterized in that, Step S6 includes: Step S61: Divide the laser transmission path into multiple thin layers, each with a thickness of... satisfy ,in The radius of the beam. The wavelength of the laser; Step S62: Perform vacuum diffraction and phase modulation calculations for each layer sequentially; multiply the vacuum diffraction by the transfer function in the frequency domain. ,in For spatial frequency, The imaginary unit; phase modulation multiplied by the phase screen in the spatial domain ,in For phase delay, For wave number, The refractive index; Step S63: Repeat until the transmission endpoint to obtain the far-field light intensity distribution. Perform Zernike polynomial fitting on the phase distortion of the far-field light intensity distribution, extract the defocus term coefficients, and calculate the first thermal focal length at the current moment based on the defocus term coefficients. (t).

8. The method according to claim 1, characterized in that, Step S7 includes: Step S71: Construct a physical information neural network, the input of which includes: Modal coefficients at the current moment ; Laser parameters: laser power The radius of the light spot at the current moment Time step ; Environmental parameters: wind speed vector Atmospheric pressure Water vapor partial pressure turbulence intensity ; The output is the initial mode coefficients for the next time step. ; Step S72: Construct the loss function for the physical information neural network: ； in: For data fitting terms, For PDE residuals, For energy conservation terms, , These are weight parameters; Step S73: After predicting the initial modal coefficients for the next time step, reconstruct the initial temperature field for the next time step using spatial basis functions: ； The initial temperature field for the next time step will be input into step S2.

9. The method according to claim 1, characterized in that, It also includes step S8: Step S81: Based on the initial mode coefficients predicted by the physical information neural network for the next moment, the second thermal focal length for the next moment is calculated through steps S5 and S6. ; Step S82: Determine the second thermal focal length With the first thermal focal length absolute difference When the absolute difference Greater than the preset threshold When this happens, it is determined that the deviation exceeds the standard; Step S83: In response to the deviation exceeding the limit, the parameters of the physical information neural network are corrected online using an extended Kalman filter, wherein the state vector is the modal coefficient vector. Observation vector The Zernike coefficients are extracted from measured wavefront data; Step S84: Calculate the cumulative deviation index ,in To map the modal coefficients to the Zernike coefficients in the observation matrix, For the integration time window, For the observation vector at time... The value, The modal coefficient vector at time t The value; when And the duration exceeds At each time step, a spatial basis function update is triggered, where... The preset threshold; Step S85: Collect the latest measured temperature field data, construct a new snapshot matrix, and perform singular value decomposition again to update the spatial basis functions. and average temperature field .

10. A laser thermal coma effect simulation system, characterized in that, include: The grid generation module is used to divide the atmospheric space along the laser transmission path into a two-layer structure of macro-grid and micro-grid. The micro-grid is dynamically activated according to the laser intensity and temperature change rate meeting preset conditions. The laser power density calculation module is used to calculate the absorbed laser power density on each macroscopic grid cell based on the initial temperature field at the current moment, and to transfer the laser power density on the macroscopic grid cell containing the activated microscopic grid to the corresponding microscopic grid. The microgrid solving module is used to solve the heat-fluid coupling equations within the activated microgrid region, using the received laser power density as the heat source, to obtain local temperature field data. The local temperature field data is then returned to the macrogrid nodes through a volume-weighted average to update the temperature field at the current moment. The intrinsic orthogonal decomposition module is used to obtain the modal coefficients at the current time through intrinsic orthogonal decomposition based on the updated temperature field and using pre-determined spatial basis functions. The reconstruction module is used to reconstruct the temperature field at the current moment based on the modal coefficients at the current moment using the spatial basis functions, and to calculate the refractive index field at the current moment based on the reconstructed temperature field. The wavefront propagation simulation module is used to calculate the laser propagation process using the current refractive index field as input and the step-by-step Fourier method, and outputs the first thermal focal length at the current moment. The physical information neural network module is used to construct a physical information neural network. It uses the modal coefficients, laser parameters, and environmental parameters at the current moment as inputs to predict the initial modal coefficients at the next moment. It also uses the spatial basis functions to reconstruct the initial temperature field at the next moment, which serves as the input to the laser power density calculation module at the next moment.

Images