Laser-induced coloring temperature field prediction method, device, equipment and medium

Abstract

The invention provides a laser-induced coloring temperature field prediction method, device, equipment and medium, and belongs to the technical field of laser marking coloring, and the method comprises the steps: constructing a laser-induced coloring photo-thermal response physical model; a neural network model with time and space positions as input and temperature as output is constructed, a loss function of the neural network model is constructed based on the photo-thermal response physical model, iterative optimization is performed on the neural network model based on the loss function and a training set, and a temperature prediction model with embedded physical information is obtained; and obtaining target time and a target spatial position of a laser-induced coloring process under a target process parameter combination, and determining a predicted temperature field under the target process parameter combination based on the target time, the target spatial position and the temperature prediction model embedded with the physical information. According to the method, the photo-thermal response physical model is taken as priori knowledge, so that the determination efficiency and precision of the predicted temperature field are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of laser marking and coloring, and in particular to a method, device, equipment and medium for predicting a temperature field of laser-induced coloring. Background Art

[0002] Laser induced coloring refers to coloring materials through laser beams. The basic principle is that the laser beam locally heats the surface of the material, which can cause phase change, oxidation or other chemical reactions on the surface of the material, thereby changing its optical properties and making the material appear in different colors. Color uniformity is a major factor that restricts the application prospects of this process.

[0003] The uniformity of color is closely related to the distribution of temperature field, and accurate temperature compensation can provide a basis for color distribution. The traditional method of analyzing the temperature field of laser-induced coloring is generally FEM finite element simulation. For different parameter combinations, each finite element simulation result needs to be processed in a complicated way, which requires a large number of iterative solutions, which is very time-consuming. In addition, the solution accuracy is easily affected by the grid accuracy, resulting in low efficiency and accuracy in determining the temperature field of laser-induced coloring.

[0004] Therefore, there is an urgent need to provide a method, device, equipment and medium for predicting the laser-induced coloring temperature field to improve the efficiency and accuracy of determining the laser-induced coloring temperature field. Summary of the invention

[0005] In view of this, it is necessary to provide a laser induced coloring temperature field prediction method, device, equipment and medium to solve the technical problem in the prior art of determining the laser induced coloring temperature field based on the finite element simulation method, resulting in low efficiency and accuracy in determining the temperature field.

[0006] On the one hand, in order to solve the above technical problems, the present invention provides a method for predicting a laser-induced coloring temperature field, comprising: Construct a physical model of the photothermal response of laser-induced coloration; Constructing a neural network model with time and spatial position as input and temperature as output, and constructing a loss function of the neural network model based on the photothermal response physical model, iteratively optimizing the neural network model based on the loss function and the training set to obtain a temperature prediction model with embedded physical information; The target time and target spatial position of the laser induced coloring process under the target process parameter combination are obtained, and the predicted temperature field under the target process parameter combination is determined based on the target time, the target spatial position and the temperature prediction model of the embedded physical information.

[0007] In a possible implementation, the photothermal response physical model includes an overall heat balance control equation, a boundary condition equation, and an initial condition equation. Then, the loss function of the neural network model constructed based on the photothermal response physical model includes: Determine the laser scanning partial differential loss term based on the overall heat balance control equation, determine the boundary condition loss term based on the boundary condition equation, and determine the initial condition equation loss term based on the initial condition; The weighted sum of the laser scanning partial differential loss term, the boundary condition loss term and the initial condition equation loss term is used as the loss function.

[0008] In a possible implementation, the overall heat balance control equation is:

[0009] The boundary condition equation is: ,

[0010]

[0011]

[0012] The initial condition equation is:

[0013] In the formula, ρ is the density of the laser-induced coloring target; C p is the specific heat capacity of the target; is the temperature field vector; t For time; k is the heat transfer efficiency; x is the horizontal coordinate value; is the energy generated per unit volume of the laser source; is the total horizontal length of a single-line scanning track; T is temperature; For instantaneous moment; Q v The heat from the internal heat source; n is the temperature gradient; h is the convective heat transfer coefficient; is the emissivity; is the Stefan-Boltzmann constant; T 0 is the initial temperature; At time 0, the horizontal coordinate x The temperature field at is the initial temperature field.

[0014] In a possible implementation, the loss function is:

[0015]

[0016]

[0017]

[0018]

[0019]

[0020]

[0021] In the formula, is the weight of the partial differential loss term of laser scanning; is the partial differential loss term of laser scanning; is the weight of the boundary condition loss term; is the boundary condition loss term; is the weight of the loss term in the initial condition equation; is the loss term of the initial condition equation; N pde is the number of sample points of the laser scanning partial differential loss term; N bc is the number of sample points of the boundary condition loss term; N init is the number of sample points of the loss term of the initial condition equation.

[0022] In a possible implementation, after obtaining the temperature prediction model with embedded physical information, the method further includes: Obtain a validation set, which includes multiple groups of validation sample data with simulated temperature labels; Inputting the spatiotemporal coordinate points in the verification sample data into the temperature prediction model embedded with physical information to obtain a predicted temperature; The accuracy of the temperature prediction model of the embedded physical information is verified based on the temperature difference between the simulated temperature tag and the predicted temperature.

[0023] In one possible implementation, the neural network model includes a spatiotemporal data feature extraction module, multiple parallel self-attention heads, and a temperature prediction module.

[0024] In a possible implementation, the method further includes: Obtaining a migration training set of the laser induced coloring process under an adjusted process parameter combination; Performing transfer learning on the temperature prediction model of the embedded physical information based on the transfer training set to obtain a transfer temperature prediction model; The temperature field under the adjustment process parameter combination is predicted based on the migration temperature prediction model.

[0025] On the other hand, the present invention also provides a laser induced coloring temperature field prediction device, comprising: A physical model building unit, used to build a photothermal response physical model of laser-induced coloration; A model building and optimization unit, used to build a neural network model with time and spatial position as input and temperature as output, and to build a loss function of the neural network model based on the photothermal response physical model, and to iteratively optimize the neural network model based on the loss function and the training set to obtain a temperature prediction model with embedded physical information; The temperature field prediction unit is used to obtain the target time and target spatial position of the laser induced coloring process under the target process parameter combination, and determine the predicted temperature field under the target process parameter combination based on the target time, the target spatial position and the temperature prediction model of the embedded physical information.

[0026] On the other hand, the present invention also provides a laser induced coloring temperature prediction device, comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the laser induced coloring temperature prediction method described in any one of the possible implementations above.

[0027] On the other hand, the present invention also provides a computer-readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the laser-induced coloring temperature prediction method described in any possible implementation method mentioned above are implemented.

[0028] The beneficial effects of the present invention are as follows: the laser induced coloring temperature prediction method provided by the present invention constructs the loss function of the neural network model based on the constructed photothermal response physical model. In the iterative optimization process of the neural network model, the photothermal response physical model is used as prior knowledge to help the neural network model better follow the laws of physics, thereby improving the accuracy of the obtained predicted temperature field. In addition, no complicated pre-processing and post-processing processes are required, thereby improving the determination efficiency of the predicted temperature field.

[0029] Moreover, the loss function in the present invention is constructed according to the physical model of photothermal response, and is not a loss function related to label value and prediction value in traditional neural network. That is, the present invention can realize the training of temperature prediction model with embedded physical information without obtaining label data, which reduces the dependence on label data and further improves the applicability and prediction efficiency of the laser induced coloring temperature field prediction method.

[0030] Furthermore, the present invention uses time and spatial position as input, and can obtain predicted temperatures at different times and locations, further improving the precision and accuracy of the predicted temperature field. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 A schematic flow chart of an embodiment of the method for predicting the temperature field of laser induced coloring provided by the present invention; Figure 2 For the present invention Figure 1 A schematic diagram of an embodiment of constructing a loss function in step S102; Figure 3 A schematic diagram of a flow chart of an embodiment of verifying a temperature prediction model with embedded physical information provided by the present invention; Figure 4 A schematic diagram of an embodiment of the present invention comparing the actual temperature and the predicted temperature that increase with the time frame; Figure 5 A schematic diagram of an embodiment of the present invention comparing the actual temperature and the predicted temperature that increase along with the laser scanning moving direction; Figure 6 A schematic diagram of the structure of an embodiment of a neural network model provided by the present invention; Figure 7 A schematic diagram of a flow chart of an embodiment of transfer learning of a temperature prediction model with embedded physical information provided by the present invention; Figure 8 A schematic diagram of the structure of an embodiment of the laser induced coloring temperature field prediction device provided by the present invention; Fig. 9 A schematic structural diagram of an embodiment of the laser induced coloring temperature field prediction device provided by the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0034] It should be understood that the schematic drawings are not drawn to scale. The flowchart used in the present invention shows the operations implemented according to some embodiments of the present invention. It should be understood that the operations of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art, under the guidance of the content of the present invention, can add one or more other operations to the flowchart, and can also remove one or more operations from the flowchart. Some of the block diagrams shown in the accompanying drawings are functional entities, which do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor systems and / or microcontroller systems.

[0035] Reference to an "embodiment" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiment may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0036] The present invention provides a method, device, equipment and medium for predicting a laser induced coloring temperature field, which are described below respectively.

[0037] Figure 1 A schematic flow chart of an embodiment of the method for predicting the temperature field of laser induced coloring provided by the present invention is shown in FIG. Figure 1 As shown, the laser induced coloring temperature field prediction method includes: S101. Constructing a photothermal response physical model of laser-induced coloration; S102, constructing a neural network model with time and spatial position as input and temperature as output, and constructing a loss function of the neural network model based on the photothermal response physical model, iteratively optimizing the neural network model based on the loss function and the training set, and obtaining a temperature prediction model with embedded physical information; S103, obtaining the target time and target spatial position of the laser induced coloring process under the target process parameter combination, and determining the predicted temperature field under the target process parameter combination based on the target time, target spatial position and a temperature prediction model with embedded physical information.

[0038] It should be noted that in order to reduce the computational cost while ensuring the training efficiency of the neural network model, the photothermal response physical model of step S101 is constructed with certain assumptions, as follows: 1. The laser scanning process is divided into two parts: heating and cooling. Only the temperature field in the laser scanning temperature rising stage is considered; 2. The laser power of the Laser Induced Coloration (LIC) process is not enough to remove a large amount of material from the target, so the mass loss of the target caused by the LIC process is negligible.

[0039] 3. Considering the computational efficiency and the difficulty of simulation, only the thermal accumulation of the temperature field of a single line scan during the LIC laser scanning coloring process is considered, and the thermal accumulation effect in the vertical direction of the scan is not considered. Under the action of a single line laser scan, the temperature distribution on the target surface satisfies the Gaussian distribution. When the single pulse laser energy is 40μm away from the origin, the temperature rise on the surface of the material can be ignored. The laser power has a decisive effect on the maximum temperature of the material surface. When the laser power drops to 1.19W, the maximum surface temperature drops to 1685K, which is lower than the melting point of the target.

[0040] 4. The target material of LIC process has uniform texture and isotropic characteristics.

[0041] 5. Thermal material properties are assumed to be quadratic functions of temperature.

[0042] 6. Only heat conduction, heat convection and heat radiation are considered, and the potential thermal phase change and heat loss during laser scanning are ignored.

[0043] 7. The laser heat source presents a Gaussian distribution and is represented by an ellipsoid model.

[0044] The training set in step S102 includes a plurality of time values ​​and a plurality of spatial position values, which can be obtained based on a COMSOL finite element simulation method.

[0045] The method based on COMSOL finite element simulation to obtain the above-mentioned multiple time values ​​and multiple spatial position values ​​is specifically as follows: first, complete the setting of the laser source of the laser induced coloring process, and set the relevant parameters of the Gaussian laser source; then, set the target material; secondly, select the solid and fluid heat transfer calculation module, select multi-physical field coupling and perform finite element mesh division, and obtain multiple spatial position values ​​corresponding to a certain time value. As the time step changes, multiple spatial position values ​​corresponding to multiple time values ​​are obtained.

[0046] It should be understood that the iterative optimization process of the neural network model in step S102 is as follows: after the network parameters of the neural network model are randomly initialized, the Adam optimizer is first used to optimize the network parameters with a speed of 10e. -3 The model is trained for 10000 epochs with a learning rate of , and then the L-BFGS optimizer performs forward propagation optimization training on the model until the loss value of the loss function reaches the convergence requirement or the maximum number of iterations is reached. The iteration is stopped to obtain a temperature prediction model with embedded physical information.

[0047] Compared with the prior art, the laser-induced coloring temperature prediction method provided by the embodiment of the present invention constructs the loss function of the neural network model based on the constructed photothermal response physical model. In the iterative optimization process of the neural network model, the photothermal response physical model is used as prior knowledge to help the neural network model better follow the laws of physics and improve the accuracy of the obtained predicted temperature field. In addition, there is no need for complex pre-processing and post-processing processes, which improves the determination efficiency of the predicted temperature field.

[0048] Moreover, the loss function in the embodiment of the present invention is constructed according to the photothermal response physical model, and is not a loss function related to label values ​​and predicted values ​​in traditional neural networks. That is, the present invention can realize the training of the temperature prediction model with embedded physical information without obtaining label data, thereby reducing the dependence on label data and further improving the applicability and prediction efficiency of the laser induced coloring temperature field prediction method.

[0049] Furthermore, the embodiment of the present invention uses time and spatial position as input, and can obtain predicted temperatures at different times and locations, further improving the precision and accuracy of the predicted temperature field.

[0050] In some embodiments of the present invention, the photothermal response physical model includes an overall heat balance control equation, a boundary condition equation, and an initial condition equation. Figure 2 As shown, the loss function of building a neural network model based on the photothermal response physical model in step S102 includes: S201, determining a laser scanning partial differential loss term based on an overall heat balance control equation, determining a boundary condition loss term based on a boundary condition equation, and determining an initial condition equation loss term based on an initial condition; S202, taking the weighted sum of the laser scanning partial differential loss term, the boundary condition loss term and the initial condition equation loss term as the loss function.

[0051] The embodiment of the present invention sets three loss terms, namely: laser scanning partial differential loss term, boundary condition loss term and initial condition equation loss term to form a loss function, which can further reduce training data, further reduce the dependence of the temperature prediction model embedded with physical information on training data, realize the rapid construction of the temperature prediction model embedded with physical information, and further improve the prediction efficiency of the temperature field.

[0052] In a specific embodiment of the present invention, the overall heat balance control equation is:

[0053] The boundary condition equation is: ,

[0054]

[0055]

[0056] The initial condition equation is:

[0057] In the formula, ρ is the density of the laser-induced coloring target; C p is the specific heat capacity of the target; is the temperature field vector; t For time; k is the heat transfer efficiency; x is the horizontal coordinate value; is the energy generated per unit volume of the laser source; is the total horizontal length of a single-line scanning track; T is temperature; For instantaneous moment; Q v The heat from the internal heat source; n is the temperature gradient; h is the convective heat transfer coefficient; is the emissivity; is the Stefan-Boltzmann constant; T0 is the initial temperature; At time 0, the horizontal coordinate x The temperature field at is the initial temperature field.

[0058] In a specific embodiment of the present invention, the loss function is:

[0059]

[0060]

[0061]

[0062]

[0063]

[0064]

[0065] In the formula, is the weight of the partial differential loss term of laser scanning; is the partial differential loss term of laser scanning; is the weight of the boundary condition loss term; is the boundary condition loss term; is the weight of the loss term in the initial condition equation; is the loss term of the initial condition equation; N pde is the number of sample points of the laser scanning partial differential loss term; N bc is the number of sample points of the boundary condition loss term; N init is the number of sample points of the loss term of the initial condition equation.

[0066] In a specific embodiment of the present invention, N pde , N init , N bc They are 20,000, 30,000 and 10,000 respectively. , , The values ​​are 2, 1, and 5 respectively.

[0067] To verify the effectiveness of the temperature prediction model with embedded physical information in predicting the temperature field, in some implementation cases of the present invention, after step S102, as Figure 3 As shown, the laser induced coloring temperature field prediction method also includes: S301, obtaining a verification set, where the verification set includes multiple groups of verification sample data carrying simulated temperature labels; S302, inputting the time-space coordinate points in the verification sample data into the temperature prediction model embedded with physical information to obtain the predicted temperature; S303 , verifying the accuracy of the temperature prediction model of the embedded physical information based on the temperature difference between the simulated temperature label and the predicted temperature.

[0068] When the temperature difference between the simulated temperature tag and the predicted temperature is less than the preset difference, the temperature prediction model with embedded physical information has a high accuracy, and step S103 can be executed.

[0069] The embodiment of the present invention verifies the validity of the temperature prediction model with embedded physical information by comparing the simulated temperature label with the predicted temperature, thereby improving the reliability and accuracy of the temperature field finally predicted.

[0070] In a specific embodiment of the present invention, the target process combination parameters are: laser waveform No. 31, laser frequency 400kHZ, laser power 30%, scanning speed 350mm / s, scanning spacing 0.002mm, and the temperature field predicted by the temperature prediction model based on embedded physical information and the simulated temperature label determined by the FEM finite element method based on COMSOL software are compared as shown in the figure. Figure 4 and Figure 5 As shown, Figure 4 The temperature field comparison with the increase of time frame is shown in Figure 2. Figure 5 Comparison of the temperature field as the laser scanning movement direction increases. Figure 4 and Figure 5 The curve composed of discrete points in is the real temperature determined based on finite element simulation, and the smooth curve is the temperature predicted by the temperature prediction model based on embedded physical information. Figure 4 From left to right, there are temperature data of three time frames: 0.010s, 0.015s, and 0.025s. Figure 5 From top to bottom, the temperature data of the three moving positions of the laser scanning movement value are 1.44mm, 2.8mm, and 4.2mm. Figure 4 and Figure 5 It can be seen that the simulated temperature label is basically consistent with the predicted temperature, which proves the effectiveness of the temperature prediction model with embedded physical information.

[0071] Since the temperature field distribution in the laser induced coloring process has significant spatial and temporal dependence, in order to further improve the training efficiency of the temperature prediction model embedded with physical information, in some embodiments of the present invention, such as Figure 6 As shown, the neural network model includes a spatiotemporal data feature extraction module, multiple parallel self-attention heads, and a temperature prediction module.

[0072] Among them, the spatiotemporal data feature extraction module is used to extract spatiotemporal features in the training set, multiple parallel self-attention heads are used to parallely calculate multiple attention scores, and then these results are concatenated together, so that the model can focus on different parts of the input sequence and capture long-range dependencies and global features between spatiotemporal features. The temperature prediction module performs temperature prediction on the outputs of multiple parallel self-attention heads.

[0073] The embodiment of the present invention can enhance the modeling ability of the neural network model for spatial and temporal dependencies by introducing a self-attention mechanism into the neural network model, thereby improving the training efficiency and prediction accuracy of the temperature prediction model with embedded physical information.

[0074] In a specific embodiment of the present invention, four self-attention heads are included.

[0075] In order to realize the prediction of the temperature field under any combination of process parameters, in some embodiments of the present invention, Figure 7 As shown, the laser induced coloring temperature field prediction method also includes: S701, obtaining a migration training set of a laser induced coloring process under an adjusted process parameter combination; S702, performing transfer learning on the temperature prediction model embedded with physical information based on the transfer training set to obtain a transfer temperature prediction model; S703 , predicting the temperature field under the adjusted process parameter combination based on the migration temperature prediction model.

[0076] The transfer learning in step S702 specifically includes: freezing most of the node parameters in the previously trained PINN model (temperature prediction model with embedded physical information), retraining the model, and obtaining a transfer temperature prediction model.

[0077] The embodiment of the present invention can realize the reuse of the temperature prediction model of the embedded physical information by performing transfer learning on the temperature prediction model of the embedded physical information obtained in step S103. Each time a new parameter combination is faced, it is only necessary to input the new training set into the temperature prediction model of the embedded physical information for fine-tuning training to obtain the migrated temperature prediction model. The fine-tuning training only takes a very short time. Compared with the existing data-driven model construction, which needs to re-run the training process every time, it can greatly save time and improve the efficiency of temperature prediction model construction.

[0078] In summary, the laser-induced coloring temperature field prediction method proposed in the embodiment of the present invention can realize the analysis of the laser-induced coloring quality from the temperature field, and realize the prediction mapping between the process parameter combination and the color uniformity quality of the laser-induced coloring. It greatly improves the application prospects of the laser-induced coloring process and provides an important reference for the optimization and improvement of the laser-induced coloring process.

[0079] In order to better implement the laser induced coloring temperature field prediction method in the embodiment of the present invention, based on the laser induced coloring temperature field prediction method, the embodiment of the present invention also provides a laser induced coloring temperature field prediction device, such as Figure 8 As shown, the laser induced coloring temperature field prediction device 800 includes: A physical model building unit 801, used to build a photothermal response physical model of laser-induced coloration; The model building and optimization unit 802 is used to build a neural network model with time and spatial position as input and temperature as output, and to build a loss function of the neural network model based on the photothermal response physical model, and to iteratively optimize the neural network model based on the loss function and the training set to obtain a temperature prediction model with embedded physical information; The temperature field prediction unit 803 is used to obtain the target time and target spatial position of the laser induced coloring process under the target process parameter combination, and determine the predicted temperature field under the target process parameter combination based on the target time, target spatial position and a temperature prediction model with embedded physical information.

[0080] The laser induced coloring temperature field prediction device 800 provided in the above embodiment can implement the technical solution described in the above laser induced coloring temperature field prediction method embodiment. The specific implementation principles of the above modules or units can refer to the corresponding contents in the above laser induced coloring temperature field prediction method embodiment, which will not be repeated here.

[0081] like Fig. 9 As shown, the present invention also provides a laser induced coloring temperature field prediction device 900. The laser induced coloring temperature field prediction device 900 includes a processor 901, a memory 902 and a display 903. Fig. 9 Only some components of the laser-induced coloring temperature field prediction device 900 are shown, but it should be understood that it is not required to implement all of the shown components, and more or fewer components may be implemented instead.

[0082] In some embodiments, the processor 901 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 902, such as the laser induced coloring temperature field prediction method of the present invention.

[0083] In some embodiments of the present invention, the processor 901 may be a single server or a server group. The server group may be centralized or distributed. In some embodiments, the processor 901 may be local or remote. In some embodiments, the processor 901 may be implemented in a cloud platform. In one embodiment, the cloud platform may include a private cloud, a public cloud, a hybrid cloud, a community cloud, a distributed cloud, an internal cloud, a multi-cloud, etc., or any combination thereof.

[0084] In some embodiments, the memory 902 may be an internal storage unit of the laser induced coloring temperature field prediction device 900, such as a hard disk or memory of the laser induced coloring temperature field prediction device 900. In other embodiments, the memory 902 may also be an external storage device of the laser induced coloring temperature field prediction device 900, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the laser induced coloring temperature field prediction device 900.

[0085] Furthermore, the memory 902 may include both an internal storage unit and an external storage device of the laser induced coloring temperature field prediction device 900. The memory 902 is used to store application software installed in the laser induced coloring temperature field prediction device 900 and various data.

[0086] In some embodiments, the display 903 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 903 is used to display information of the laser-induced coloring temperature field prediction device 900 and to display a visual user interface. The components 901-903 of the laser-induced coloring temperature field prediction device 900 communicate with each other through a system bus.

[0087] In some embodiments of the present invention, when the processor 901 executes the laser induced coloring temperature field prediction program in the memory 902, the following steps may be implemented: Construct a physical model of the photothermal response of laser-induced coloration; Construct a neural network model with time and spatial position as input and temperature as output, and construct a loss function of the neural network model based on the photothermal response physical model. Iteratively optimize the neural network model based on the loss function and training set to obtain a temperature prediction model with embedded physical information. The target time and target spatial position of the laser induced coloring process under the target process parameter combination are obtained, and the predicted temperature field under the target process parameter combination is determined by a temperature prediction model based on the target time, target spatial position and embedded physical information.

[0088] It should be understood that: when the processor 901 executes the laser induced coloring temperature field prediction program in the memory 902, in addition to the above functions, other functions can also be implemented. For details, please refer to the description of the corresponding method embodiment above.

[0089] Furthermore, the embodiment of the present invention does not specifically limit the type of the laser induced coloring temperature field prediction device 900 mentioned. The laser induced coloring temperature field prediction device 900 may be a portable laser induced coloring temperature field prediction device such as a mobile phone, a tablet computer, a personal digital assistant (PDA), a wearable device, a laptop computer (Laptop), etc. Exemplary embodiments of portable laser induced coloring temperature field prediction devices include but are not limited to portable laser induced coloring temperature field prediction devices equipped with IOS, Android, Microsoft or other operating systems. The above-mentioned portable laser induced coloring temperature field prediction device may also be other portable laser induced coloring temperature field prediction devices. It should also be understood that in some other embodiments of the present invention, the laser induced coloring temperature field prediction device 900 may not be a portable laser induced coloring temperature field prediction device, but a desktop computer with a touch-sensitive surface (such as a touch panel).

[0090] Accordingly, an embodiment of the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, it can implement the steps or functions of the laser-induced coloring temperature field prediction method provided in the above-mentioned method embodiments.

[0091] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware (such as a processor, a controller, etc.) through a computer program, and the computer program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0092] The above is a detailed introduction to the laser induced coloring temperature field prediction method, device, equipment and storage medium provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for technical personnel in this field, according to the idea of ​​the present invention, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A method for predicting temperature field of laser induced coloring, characterized in that: include: Construct a physical model of the photothermal response of laser-induced coloration; Constructing a neural network model with time and spatial position as input and temperature as output, and constructing a loss function of the neural network model based on the photothermal response physical model, iteratively optimizing the neural network model based on the loss function and the training set to obtain a temperature prediction model with embedded physical information; The target time and target spatial position of the laser induced coloring process under the target process parameter combination are obtained, and the predicted temperature field under the target process parameter combination is determined based on the target time, the target spatial position and the temperature prediction model of the embedded physical information.

2. The laser induced coloring temperature field prediction method according to claim 1, characterized in that: The photothermal response physical model includes an overall heat balance control equation, a boundary condition equation, and an initial condition equation. The loss function of the neural network model constructed based on the photothermal response physical model includes: Determine the laser scanning partial differential loss term based on the overall heat balance control equation, determine the boundary condition loss term based on the boundary condition equation, and determine the initial condition equation loss term based on the initial condition; The weighted sum of the laser scanning partial differential loss term, the boundary condition loss term and the initial condition equation loss term is used as the loss function.

3. The laser induced coloring temperature field prediction method according to claim 2, characterized in that: The overall heat balance control equation is: The boundary condition equation is: ， The initial condition equation is: In the formula, ρ is the density of the laser-induced coloring target; C p is the specific heat capacity of the target; is the temperature field vector; t For time; k is the heat transfer efficiency; x is the horizontal coordinate value; is the energy generated per unit volume of the laser source; is the total horizontal length of a single-line scanning track; T is temperature; For instantaneous moment; Q v The heat from the internal heat source; n is the temperature gradient; h is the convective heat transfer coefficient; is the emissivity; is the Stefan-Boltzmann constant; T 0 is the initial temperature; At time 0, the horizontal coordinate x The temperature field at is the initial temperature field.

4. The laser induced coloring temperature field prediction method according to claim 3, characterized in that: The loss function is: In the formula, is the weight of the partial differential loss term of laser scanning; is the partial differential loss term of laser scanning; is the weight of the boundary condition loss term; is the boundary condition loss term; is the weight of the loss term in the initial condition equation; is the loss term of the initial condition equation; N pde is the number of sample points of the laser scanning partial differential loss term; N bc is the number of sample points of the boundary condition loss term; N init is the number of sample points of the loss term of the initial condition equation.

5. The laser induced coloring temperature field prediction method according to claim 1, characterized in that: After obtaining the temperature prediction model with embedded physical information, it also includes: Obtain a validation set, which includes multiple groups of validation sample data with simulated temperature labels; Inputting the spatiotemporal coordinate points in the verification sample data into the temperature prediction model embedded with physical information to obtain a predicted temperature; The accuracy of the temperature prediction model of the embedded physical information is verified based on the temperature difference between the simulated temperature tag and the predicted temperature.

6. The laser induced coloring temperature field prediction method according to claim 1, characterized in that: The neural network model includes a spatiotemporal data feature extraction module, multiple parallel self-attention heads and a temperature prediction module.

7. The laser induced coloring temperature field prediction method according to claim 1, characterized in that: The method further comprises: Obtaining a migration training set of the laser induced coloring process under an adjusted process parameter combination; Performing transfer learning on the temperature prediction model of the embedded physical information based on the transfer training set to obtain a transfer temperature prediction model; The temperature field under the adjustment process parameter combination is predicted based on the migration temperature prediction model.

8. A laser induced coloring temperature field prediction device, characterized in that: include: A physical model building unit, used to build a photothermal response physical model of laser-induced coloration; A model building and optimization unit, used to build a neural network model with time and spatial position as input and temperature as output, and to build a loss function of the neural network model based on the photothermal response physical model, and to iteratively optimize the neural network model based on the loss function and the training set to obtain a temperature prediction model with embedded physical information; The temperature field prediction unit is used to obtain the target time and target spatial position of the laser induced coloring process under the target process parameter combination, and determine the predicted temperature field under the target process parameter combination based on the target time, the target spatial position and the temperature prediction model of the embedded physical information.

9. A laser induced coloring temperature prediction device, characterized in that: comprising a memory and a processor, wherein: The memory is used to store programs; The processor is coupled to the memory and is used to execute the program stored in the memory to implement the steps in the laser induced coloring temperature field prediction method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program or instruction, and when the program or instruction is executed by a processor, the steps of the laser-induced coloring temperature field prediction method described in any one of claims 1 to 7 are implemented.

Images