Titanium alloy processing interface thermal distribution prediction model construction method and prediction method
By constructing a thermal distribution prediction model for the titanium alloy processing interface and combining molecular dynamics and neural networks, the problem of microstructure influence in titanium alloy femtosecond laser processing was solved, a more uniform processing effect was achieved, and the industrial application of titanium alloy was promoted.
Patent Information
- Application Number
- CN202510565830.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-30
- Publication Date
- 2025-09-19
AI Technical Summary
Existing technologies fail to effectively consider the internal microstructure of the metal in femtosecond laser processing of titanium alloys, resulting in energy concentration and lattice defects, which affects processing reliability.
A prediction model for the thermal distribution of the titanium alloy processing interface was constructed. The molecular dynamics model of the titanium alloy was simulated using multiple sets of femtosecond laser process parameters. Combined with neural network training, the electron temperature distribution at the interface between the α phase and the β phase was predicted.
Effectively predict the heat transfer concentration of femtosecond laser energy at the two-phase interface of titanium alloy, guide parameter optimization, improve processing uniformity, promote the surface functionalization of large-size titanium alloys and promote industrial applications.
Smart Images
Figure CN120673923A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of femtosecond laser technology, and in particular to a method for constructing a prediction model for thermal distribution at a titanium alloy processing interface and a prediction method. Background Art
[0002] Titanium alloys, a typical example of lightweight metals, are widely used in aerospace, shipping, medical, and other fields, and their femtosecond laser precision processing has a broad market application. However, due to the uneven phase distribution within titanium alloys (the coexistence of α and β phases), the energy of the femtosecond laser is easily concentrated during its internal transmission, resulting in a large number of lattice defects, which seriously affect the reliability of the femtosecond laser processing process. Therefore, it is possible to control lattice defects by regulating femtosecond laser processing parameters such as laser energy and scanning speed.
[0003] Currently, there are few methods for predicting the temperature distribution at the machining interface based on femtosecond laser machining parameters, and most of them focus on macroscopic heat transfer, ignoring the influence of the metal's internal microstructure. Therefore, finding a method to predict the temperature distribution at the machining interface that considers the metal's internal microstructure is crucial for femtosecond laser precision machining of titanium alloys. Summary of the Invention
[0004] In view of this, an embodiment of the present application provides a method for constructing a prediction model for thermal distribution of a titanium alloy processing interface and a prediction method.
[0005] According to the first aspect of the present application, an embodiment of the present application provides a method for constructing a thermal distribution prediction model for a titanium alloy processing interface, comprising:
[0006] The titanium alloy molecular dynamics model was simulated using multiple sets of femtosecond laser processing parameters to obtain the lattice temperature distribution corresponding to each set of femtosecond laser processing parameters. The titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α-phase and β-phase, as well as the corresponding potential function, temperature boundary conditions, dual-temperature model, and related thermophysical material parameters.
[0007] Under the NVT ensemble, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters was loaded into the titanium alloy molecular dynamics model, and iterative calculations were performed to obtain the corresponding electron temperature distribution at the interface between the α-phase and β-phase.
[0008] Each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network and obtain a prediction model for the thermal-mechanical distribution of the titanium alloy processing interface.
[0009] Optionally, the steps of constructing the titanium alloy molecular dynamics model include:
[0010] Construct a titanium alloy polycrystalline model with α phase and β phase;
[0011] The titanium alloy polycrystal model is subjected to a heat treatment under the NPT ensemble until the total energy of the titanium alloy polycrystal model reaches a stable state;
[0012] The potential function, temperature boundary conditions, dual-temperature model and related thermophysical material parameters corresponding to the titanium alloy polycrystalline model after Chiyu treatment were established to obtain the titanium alloy molecular dynamics model.
[0013] Optionally, the formula corresponding to the dual-temperature model is:
[0014]
[0015] Where S(x,t) represents the energy of the femtosecond laser in the x direction of the processing interface at time t; C e is the electron heat capacity; C l is the lattice heat capacity; T e is the electron temperature, T l is the lattice temperature; g is the electron-lattice coupling coefficient, k e is the electronic thermal conductivity.
[0016] Optionally, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model, including:
[0017] The titanium alloy polycrystalline model in the titanium alloy molecular dynamics model is divided into multiple sub-models according to the spatial front-back order;
[0018] Based on the lattice temperature distribution corresponding to each set of femtosecond laser process parameters, the lattice temperature distribution corresponding to the multiple sub-models is determined.
[0019] Optionally, each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train a neural network to obtain a prediction model for the thermal distribution of the titanium alloy processing interface, including:
[0020] Each set of femtosecond laser process parameters and the corresponding electron temperature distribution are sequentially input into the fusion module of the neural network for feature fusion and extraction to obtain the corresponding first feature;
[0021] Pulse-encoding the first feature to obtain a second feature;
[0022] The second feature is expanded according to the time sequence of pulse emission during pulse coding, and is input into the pulse unit of the neural network for fitting and prediction to obtain a prediction result;
[0023] The prediction results are output through the output layer of the neural network, and the parameters of the neural network are corrected based on the prediction results. The step of inputting each set of femtosecond laser process parameters and the corresponding electron temperature distribution into the fusion module of the neural network is returned until the model training convergence condition is reached, and a thermal distribution prediction model of the titanium alloy processing interface is obtained.
[0024] Optionally, the pulse transmission time is calculated as follows during pulse coding:
[0025] t f (z t )=(T-1)(1-x t );
[0026] Among them, t f (z t ) is the pulse emission time, T is the maximum pulse emission duration, x t is the second feature at the current moment, x t ∈[0,1].
[0027] Optionally, the prediction results are output through the output layer of the neural network, including:
[0028] The prediction results are mapped through the mapping unit of the output layer of the neural network to obtain the probability information of the prediction results;
[0029] The output layer of the neural network is leaky and integrated to release neurons to output the probability information of the prediction results.
[0030] According to the second aspect of the present application, an embodiment of the present application provides a method for predicting thermal distribution of a titanium alloy processing interface, comprising:
[0031] The molecular dynamics model of titanium alloy is simulated using the femtosecond laser process parameters to be measured to obtain the lattice temperature distribution corresponding to the femtosecond laser processing parameters to be measured. The molecular dynamics model of titanium alloy includes a titanium alloy polycrystalline model with α phase and β phase, as well as the corresponding potential function, temperature boundary conditions, dual-temperature model, and related thermophysical material parameters.
[0032] Under the NVT ensemble, the lattice temperature distribution corresponding to the femtosecond laser process parameters to be measured is loaded into the titanium alloy molecular dynamics model, and iterative calculations are performed to obtain the measured electron temperature distribution at the interface between the α-phase and β-phase.
[0033] The thermal distribution prediction model of the titanium alloy processing interface is used to process the femtosecond laser process parameters to be measured and the electron temperature distribution to be measured, and the thermal distribution of the titanium alloy processing interface corresponding to the femtosecond laser process parameters to be measured is predicted; the thermal distribution prediction model of the titanium alloy processing interface is obtained by the construction method as in the first aspect or any embodiment of the first aspect.
[0034] According to a third aspect of the present application, an embodiment of the present application provides an electronic device, including:
[0035] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so as to cause the at least one processor to execute the method for constructing a model for predicting the thermal-mechanical distribution of a titanium alloy processing interface as in the first aspect or any embodiment of the first aspect, or the method for predicting the thermal-mechanical distribution of a titanium alloy processing interface as in the second aspect.
[0036] According to the fourth aspect of the present application, an embodiment of the present application provides a computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable a computer to execute the method for constructing a thermal distribution prediction model for a titanium alloy processing interface as in the first aspect or any embodiment of the first aspect, or the method for predicting the thermal distribution of a titanium alloy processing interface as in the second aspect.
[0037] The embodiment of the present application provides a method for constructing a prediction model for the thermal distribution of a titanium alloy processing interface and a prediction method. By using multiple sets of femtosecond laser process parameters to simulate the titanium alloy molecular dynamics model respectively, the lattice temperature distribution corresponding to each set of femtosecond laser processing parameters is obtained; the titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α phase and β phase, as well as corresponding potential functions, temperature boundary conditions, dual-temperature models, and related thermal physical material parameters; under the NVT ensemble, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model respectively, and iterative calculations are performed to obtain the corresponding electron temperature distribution at the interface of α phase and β phase; each set of femtosecond laser process parameters and the corresponding electron temperature are loaded into the titanium alloy molecular dynamics model respectively, and the corresponding electron temperature distribution at the interface of α phase and β phase is obtained. The degree distribution is used as training data to train the neural network and obtain the thermal distribution prediction model of the titanium alloy processing interface; in this way, the electron temperature distribution at the interface between the α phase and β phase of the titanium alloy under the action of femtosecond laser with different process parameter groups is obtained by combining the dual-temperature model with molecular dynamics, and then each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network and obtain the thermal distribution prediction model of the titanium alloy processing interface, thereby constructing a thermal distribution prediction framework for the two-phase interface of titanium alloy femtosecond laser processing that integrates microscale simulation and neural network, and introduces the influence of the two-phase structure of titanium alloy into the heat transfer prediction of the femtosecond laser processing interface, which will effectively guide the uniform processing of titanium alloy femtosecond laser.
[0038] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and specific advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] Figure 1 Schematic diagram of a process for constructing a thermal distribution prediction model for a titanium alloy processing interface according to an embodiment of the present application;
[0040] Figure 2 This is a schematic diagram of the process of training a neural network in an embodiment of the present application;
[0041] Figure 3 This is a flow chart of a method for predicting thermal distribution on a titanium alloy processing interface in an embodiment of the present application;
[0042] Figure 4 This is a schematic diagram of the hardware structure of an electronic device in an embodiment of the present application. DETAILED DESCRIPTION
[0043] To make the purpose, technical solutions, and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative efforts shall fall within the scope of protection of this application.
[0044] The present application embodiment provides a method for constructing a prediction model of thermal distribution of titanium alloy processing interface, such as Figure 1 Shown, including:
[0045] S101, using multiple sets of femtosecond laser process parameters to simulate the titanium alloy molecular dynamics model respectively, and obtaining the lattice temperature distribution corresponding to each set of femtosecond laser processing parameters; the titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α phase and β phase, as well as the corresponding potential function, temperature boundary conditions, dual-temperature model, and related thermophysical material parameters.
[0046] In this embodiment, the femtosecond laser process parameters may include laser power, spot diameter, repetition frequency, etc.
[0047] In some embodiments, the femtosecond laser power may be 10 W to 100 W, the spot diameter may be 10 μm to 50 μm, and the repetition frequency may be 1 kHz to 10 GHz.
[0048] In some embodiments, the number of groups of femtosecond laser process parameters may be 100 groups.
[0049] In this embodiment, the temperature boundary conditions may include periodic boundary conditions in the X direction and the Y direction, and a free boundary condition in the Z direction.
[0050] In some embodiments, the steps of constructing a molecular dynamics model of titanium alloy include:
[0051] A titanium alloy polycrystalline model with α phase and β phase was constructed; the titanium alloy polycrystalline model was subjected to heat treatment under the NPT ensemble until the total energy of the titanium alloy polycrystalline model reached a stable state; the potential function, temperature boundary conditions, dual-temperature model and related thermophysical material parameters corresponding to the titanium alloy polycrystalline model after heat treatment were established to obtain the titanium alloy molecular dynamics model.
[0052] In some embodiments, relevant thermophysical material parameters may include electron heat capacity, lattice heat capacity, electron-lattice coupling coefficient, and electron thermal conductivity.
[0053] In some embodiments, constructing a titanium alloy polycrystalline model having α phase and β phase includes establishing, expanding, and filling a titanium alloy polycrystalline model having α phase and β phase.
[0054] In some embodiments, the formula corresponding to the dual-temperature model is:
[0055]
[0056] Where S(x,t) represents the energy of the femtosecond laser in the x direction of the processing interface at time t; C e is the electron heat capacity; C l is the lattice heat capacity; T e is the electron temperature, T l is the lattice temperature; g is the electron-lattice coupling coefficient, k e is the electronic thermal conductivity.
[0057] In some embodiments, the absorption of femtosecond laser energy within the titanium alloy satisfies the Beer-Lambert law, and the laser intensity at depth Z is:
[0058] I=(1-R)I0e -αz ;
[0059] Where R is the reflectivity, I0 is the initial intensity of the incident laser, and α is the laser absorption coefficient.
[0060] S102, under the NVT ensemble, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model, and iterative calculation is performed to obtain the corresponding electron temperature distribution at the interface between the α phase and the β phase.
[0061] In this example, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters can be loaded into the titanium alloy molecular dynamics model under the NVT ensemble. Iterative calculations are then performed under the NVE ensemble to obtain the corresponding electron temperature distribution at the interface between the α-phase and β-phase.
[0062] In this embodiment, the interface between the α phase and the β phase is a processing interface, that is, a two-phase interface.
[0063] In some embodiments, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model, including:
[0064] The titanium alloy polycrystalline model in the titanium alloy molecular dynamics model is divided into multiple sub-models according to the spatial front-to-back order; based on the lattice temperature distribution corresponding to each set of femtosecond laser process parameters, the lattice temperature distribution corresponding to the multiple sub-models is determined.
[0065] S103, using each set of femtosecond laser process parameters and the corresponding electron temperature distribution as training data, training the neural network to obtain a thermal distribution prediction model for the titanium alloy processing interface.
[0066] In this embodiment, for each set of femtosecond laser process parameters, process parameter-lattice peak temperature-cooling rate-coincidence atom data can be established, and then each set of process parameter-lattice peak temperature-cooling rate-coincidence atom data, as well as the corresponding electron temperature distribution, are used as training data to train the neural network to obtain a thermal distribution prediction model for the titanium alloy processing interface.
[0067] In this embodiment, the neural network may be an LSTM-SNN neural network.
[0068] In some embodiments, each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train a neural network to obtain a thermal distribution prediction model for the titanium alloy processing interface, including:
[0069] Each set of femtosecond laser process parameters and the corresponding electron temperature distribution are input into the fusion module of the neural network in turn for feature fusion and extraction to obtain the corresponding first feature; the first feature is pulse-encoded to obtain the second feature; the second feature is expanded according to the time sequence of pulse emission during pulse encoding, and input into the pulse unit of the neural network for fitting and prediction to obtain the prediction result; the prediction result is output through the output layer of the neural network, and the parameters of the neural network are corrected based on the prediction result, and the step of inputting each set of femtosecond laser process parameters and the corresponding electron temperature distribution into the fusion module of the neural network is returned until the model training convergence condition is reached to obtain the thermal distribution prediction model of the titanium alloy processing interface.
[0070] When implementing it specifically, Figure 2As shown in the figure, after the first feature is adaptively pulse coded, the second feature can be obtained. When the second feature is expanded according to the time sequence of pulse emission during pulse coding, lightweight feature extraction can be performed on the second feature. For example, after lightweight feature extraction of the second feature, the lattice peak temperature, cooling rate, coincident atoms, etc. can be obtained respectively. Then, it can be input into the pulse unit of the neural network, such as the LSTM unit, for fitting and prediction to obtain the prediction result h x The output layer may include Softmax, which can map the prediction results to a probability space, and then use leaky integrated release neurons to output information.
[0071] In some embodiments, the pulse transmission time during pulse encoding is calculated as follows:
[0072] t f (z t )=(T-1)(1-x t );
[0073] Among them, t f (z t ) is the pulse emission time, T is the maximum pulse emission duration, x t is the second feature at the current moment, x t ∈[0,1].
[0074] In some embodiments, outputting the prediction result through the output layer of the neural network includes:
[0075] The prediction results are mapped through the mapping unit of the output layer of the neural network to obtain the probability information of the prediction results; the prediction result probability information is output through the leakage integrated release neurons of the output layer of the neural network.
[0076] The embodiment of the present application provides a method for constructing a prediction model for the thermal distribution of the titanium alloy processing interface. By using multiple sets of femtosecond laser process parameters to simulate the titanium alloy molecular dynamics model respectively, the lattice temperature distribution corresponding to each set of femtosecond laser processing parameters is obtained; the titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α phase and β phase, as well as corresponding potential functions, temperature boundary conditions, dual-temperature models, and related thermal physical material parameters; under the NVT ensemble, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model respectively, and iterative calculations are performed to obtain the corresponding electron temperature distribution at the interface of α phase and β phase; each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network to obtain the titanium alloy processing interface. Surface thermal distribution prediction model; in this way, the electronic temperature distribution at the interface between the α-phase and β-phase of titanium alloy under the action of femtosecond laser with different process parameter groups is obtained by combining the dual-temperature model with molecular dynamics, and then each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network to obtain the thermal distribution prediction model of the titanium alloy processing interface, thereby constructing a thermal distribution prediction framework for the two-phase interface of titanium alloy femtosecond laser processing that integrates micro-scale simulation and neural network, and introduces the influence of the two-phase structure of titanium alloy into the heat transfer prediction of the femtosecond laser processing interface, which can effectively predict the heat transfer concentration of femtosecond laser energy at the two-phase interface, guide subsequent parameter optimization to improve processing uniformity, and has great significance for the functionalization of large-size titanium alloy surfaces and promote their industrial application.
[0077] The following describes a specific implementation method for constructing a thermal distribution prediction model for a titanium alloy processing interface of the present application.
[0078] The present application provides a method for constructing a thermal distribution prediction model for a titanium alloy processing interface, the method comprising the following steps:
[0079] S1. Based on the Voronoi diagram method, Atomsk software was used to construct polycrystalline models of titanium alloys with α and β phases. The atomic ratio of Ti, Al, and V was 7:2:1, and the grain orientations were all rotated about the Z axis.
[0080] S2. Energy minimization is performed under the NPT ensemble with a relaxation temperature of 300 K and a relaxation pressure of zero until the total energy of the system reaches a stable state.
[0081] S3. Use the 2NN-MEAM potential function to determine the potential functions of Ti-Al, Ti-V, and Al-V binary alloys respectively. Set the X and Y directions of the model as periodic boundary conditions, and the Z direction as a free boundary condition.
[0082] S4. Solve the two-temperature model using Matlab's PDEPE function. The initial temperature is 300K, the femtosecond laser wavelength is 1064nm, and the boundary conditions are set to adiabatic. The absorption of femtosecond laser energy inside the titanium alloy satisfies the Beer-Lambert law. The laser intensity at depth Z is:
[0083] I=(1-R)I0e -αz ;
[0084] Where R is the reflectivity, I0 is the initial intensity of the incident laser, and α is the laser absorption coefficient.
[0085] S5. Collect different parameter groups, including femtosecond laser power (10-100 W), spot diameter (10-50 μm), repetition frequency (1 kHz-10 GHz), etc. The number of parameters is 100 groups.
[0086] S6. The lattice temperatures calculated with different parameter groups are loaded into the titanium alloy molecular dynamics model. Considering the laser influence depth, the titanium alloy polycrystalline model is spatially divided into 8 equal layers. The temperature is distributed in each layer with a specified gradient. The NVT ensemble is stabilized to a stable state.
[0087] The S7 and NVE ensemble calculations were used to obtain the electron temperature distribution along the depth direction of the two-phase interface, and 100 sets of process parameters—lattice peak temperature—cooling rate—coincident atom data were established.
[0088] S8. Using Vienna filtering, we divided the 100 sets of process parameter data—lattice peak temperature, cooling rate, and coincident atom data—into 10 small segments, each containing 10 data sets. We assigned each segment a label, which is the segment number, to facilitate subsequent model learning and evaluation. The dataset was divided into training, validation, and test sets, accounting for 70%, 10%, and 20% of the total, respectively.
[0089] S9. Input the electronic temperature distribution image of the two-phase interface and the corresponding feature point data (process parameters - lattice peak temperature - cooling rate - overlapping atomic data) into the fusion module of the neural network to fuse the information between channels and perform deep 3D convolution, perform feature fusion and extraction, and effectively fuse low-level features with high-level features. The convolution kernel size of the deep 3D convolution on each branch is 3×3×3.
[0090] S10. Build a single-layer LSTM-SNN and a fully connected layer. The number of hidden layers is set to 12 and the learning rate is set to 0.001.
[0091] S11, perform adaptive pulse coding to generate a 40×280 threshold matrix, which obeys a Gaussian distribution with a mean of 0 and a variance of 1. When the value of the current feature point data exceeds the local maximum value, the feature point data is encoded as 1, otherwise it is 0. The pulse emission time is:
[0092] t f (z t )=(T-1)(1-x t );
[0093] Among them, t f (z t ) is the pulse emission time, T is the maximum pulse emission duration, x t is the second feature at the current moment, x t ∈[0,1].
[0094] S12. After pulse encoding, the features are expanded by time step and input into the pulse unit, which processes the current input and the output of the previous hidden layer. The output layer uses Softmax to map the prediction results to a probability space, and then uses leaky integrate-and-release neurons to transmit information. The specific process of charging and discharging leaky integrate-and-release neurons can be expressed as:
[0095] H t =V t-1 +ΔV t ,
[0096]
[0097] Where H t is the instantaneous voltage of the neuron at time t, V t-1 is the voltage at the previous moment, ΔV t is the voltage increment, θ t is the pulse identifier, V threshold is the threshold.
[0098] S13. Compare the test set with the prediction results, and use evaluation indicators such as accuracy, precision, and recall to judge the accuracy and operating efficiency of the prediction framework.
[0099] The present application also provides a method for predicting thermal distribution of titanium alloy processing interface. Figure 3 Shown, including:
[0100] S301 , simulating a titanium alloy molecular dynamics model using the femtosecond laser processing parameters to be measured to obtain a lattice temperature distribution corresponding to the femtosecond laser processing parameters to be measured. The titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model having α and β phases, as well as corresponding potential functions, temperature boundary conditions, a two-temperature model, and related thermophysical material parameters. For a specific implementation, see the above embodiment.
[0101] S302, in the NVT ensemble, the lattice temperature distribution corresponding to the femtosecond laser process parameter to be measured is loaded into the titanium alloy molecular dynamics model, and iterative calculation is performed to obtain the electron temperature distribution to be measured at the interface between the α phase and the β phase. Specific implementation methods refer to the above embodiment.
[0102] S303, through the titanium alloy processing interface thermal distribution prediction model, the femtosecond laser process parameters to be measured and the electron temperature distribution to be measured are processed to predict the thermal distribution of the titanium alloy processing interface corresponding to the femtosecond laser process parameters to be measured; the titanium alloy processing interface thermal distribution prediction model is obtained by the construction method in any of the above embodiments.
[0103] The embodiment of the present application provides a method for predicting the thermal distribution of the titanium alloy processing interface, which simulates the titanium alloy molecular dynamics model by using multiple groups of femtosecond laser process parameters to obtain the lattice temperature distribution corresponding to each group of femtosecond laser processing parameters; the titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α phase and β phase, as well as corresponding potential functions, temperature boundary conditions, dual-temperature models, and related thermal physical material parameters; under the NVT ensemble, the lattice temperature distribution corresponding to each group of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model, and iterative calculation is performed to obtain the corresponding electron temperature distribution at the interface of α phase and β phase; each group of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network to obtain the thermal distribution of the titanium alloy processing interface. force distribution prediction model; in this way, the electronic temperature distribution at the interface between the α-phase and β-phase of the titanium alloy under the action of femtosecond laser with different process parameter groups is obtained by combining the dual-temperature model with molecular dynamics, and then each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network to obtain the thermal-mechanical distribution prediction model of the titanium alloy processing interface, thereby constructing a thermal-mechanical distribution prediction framework for the two-phase interface of titanium alloy femtosecond laser processing that integrates micro-scale simulation and neural network, and introduces the influence of the two-phase structure of titanium alloy into the heat transfer prediction of the femtosecond laser processing interface, which can effectively predict the heat transfer concentration of femtosecond laser energy at the two-phase interface, guide subsequent parameter optimization to improve processing uniformity, and has great significance for the functionalization of large-size titanium alloy surfaces and promote their industrial application.
[0104] According to an embodiment of the present application, the present application also provides an electronic device and a readable storage medium.
[0105] Figure 4A schematic block diagram of an example electronic device 800 that can be used to implement an embodiment of the present application is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.
[0106] like Figure 4 As shown, the electronic device 800 includes a computing unit 801, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 802 or a computer program loaded from a storage unit 808 into a random access memory (RAM) 803. In the RAM 803, various programs and data required for the operation of the electronic device 800 can also be stored. The computing unit 801, the ROM 802, and the RAM 803 are connected to each other via a bus 804. An input / output (I / O) interface 805 is also connected to the bus 804.
[0107] Multiple components in the electronic device 800 are connected to the I / O interface 805, including an input unit 806, such as a keyboard, a mouse, etc.; an output unit 807, such as various types of displays, speakers, etc.; a storage unit 808, such as a magnetic disk, an optical disk, etc.; and a communication unit 809, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 809 allows the electronic device 800 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.
[0108] The computing unit 801 can be various general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 801 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 801 performs the various methods and processes described above, such as a method for constructing a thermal distribution prediction model for a titanium alloy processing interface or a method for predicting the thermal distribution of a titanium alloy processing interface. For example, in some embodiments, a method for constructing a thermal distribution prediction model for a titanium alloy processing interface or a method for predicting the thermal distribution of a titanium alloy processing interface can be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 808. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 800 via the ROM 802 and / or the communication unit 809. When the computer program is loaded into RAM 803 and executed by the computing unit 801, one or more steps of the method for constructing a thermal-mechanical distribution prediction model for a titanium alloy processing interface or the method for predicting thermal-mechanical distribution of a titanium alloy processing interface described above may be executed. Alternatively, in other embodiments, the computing unit 801 may be configured to execute the method for constructing a thermal-mechanical distribution prediction model for a titanium alloy processing interface or the method for predicting thermal-mechanical distribution of a titanium alloy processing interface by any other appropriate means (e.g., by means of firmware).
[0109] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.
[0110] The program code for implementing the methods of the present application can be written in any combination of one or more programming languages. Such program code can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that when the program code is executed by the processor or controller, the functions / operations specified in the flow charts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.
[0111] In the context of the present application, a machine-readable medium can be a tangible medium that can contain or store a program for use by an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
[0112] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).
[0113] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.
[0114] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises through computer programs running on the respective computers and having a client-server relationship with each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.
[0115] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this application can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions disclosed in this application can be achieved. This is not a limitation herein.
[0116] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.
[0117] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.
Claims
1. A method for constructing a thermal distribution prediction model for a titanium alloy processing interface, characterized in that: include: The molecular dynamics model of titanium alloy was simulated using multiple sets of femtosecond laser processing parameters to obtain the lattice temperature distribution corresponding to each set of femtosecond laser processing parameters. The titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α phase and β phase, as well as corresponding potential functions, temperature boundary conditions, a dual-temperature model, and related thermophysical material parameters; Under the NVT ensemble, the lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model, and iterative calculations are performed to obtain the corresponding electron temperature distribution at the interface between the α-phase and β-phase. Each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network and obtain a prediction model for the thermal-mechanical distribution of the titanium alloy processing interface.
2. The method according to claim 1, characterized in that The steps for constructing the titanium alloy molecular dynamics model include: Construct a titanium alloy polycrystalline model with α phase and β phase; performing a heat treatment on the titanium alloy polycrystal model under the NPT ensemble until the total energy of the titanium alloy polycrystal model reaches a stable state; The potential function, temperature boundary conditions, dual-temperature model and related thermophysical material parameters corresponding to the titanium alloy polycrystalline model after the Chiyu treatment are established to obtain the titanium alloy molecular dynamics model.
3. The method according to claim 1, characterized in that The formula corresponding to the dual-temperature model is: Where S(x,t) represents the energy of the femtosecond laser in the x direction of the processing interface at time t; C e is the electron heat capacity; C l is the lattice heat capacity; T e is the electron temperature, T l is the lattice temperature; g is the electron-lattice coupling coefficient, k e is the electronic thermal conductivity.
4. The method according to claim 1, wherein The lattice temperature distribution corresponding to each set of femtosecond laser process parameters is loaded into the titanium alloy molecular dynamics model, including: Dividing the titanium alloy polycrystal model in the titanium alloy molecular dynamics model into a plurality of sub-models according to a spatial front-to-back order; Based on the lattice temperature distribution corresponding to each set of femtosecond laser process parameters, the lattice temperature distribution corresponding to the plurality of sub-models is determined.
5. The method according to claim 1, wherein Each set of femtosecond laser process parameters and the corresponding electron temperature distribution are used as training data to train the neural network and obtain a prediction model for the thermal distribution of the titanium alloy processing interface, including: Each set of femtosecond laser process parameters and the corresponding electron temperature distribution are sequentially input into the fusion module of the neural network for feature fusion and extraction to obtain the corresponding first feature; Pulse-encoding the first feature to obtain a second feature; Expanding the second feature according to the time sequence of pulse emission during pulse coding, and inputting it into the pulse unit of the neural network to perform fitting and prediction to obtain a prediction result; The prediction results are output through the output layer of the neural network, and the parameters of the neural network are corrected based on the prediction results. The step of inputting each set of femtosecond laser process parameters and the corresponding electron temperature distribution into the fusion module of the neural network is returned until the model training convergence condition is reached, thereby obtaining a thermal distribution prediction model of the titanium alloy processing interface.
6. The method according to claim 5, characterized in that The calculation formula for pulse transmission time in pulse coding is as follows: t f (z t )=(T-1)(1-x t ); Among them, t f (z t ) is the pulse emission time, T is the maximum pulse emission duration, x t is the second feature at the current moment, x t ∈[0,1].
7. The method according to claim 5, characterized in that Outputting the prediction result through the output layer of the neural network includes: Mapping the prediction results through a mapping unit of an output layer of a neural network to obtain prediction result probability information; The prediction result probability information is output by leaky integrated release neurons in the output layer of the neural network.
8. A method for predicting thermal distribution of titanium alloy processing interface, characterized in that: include: The molecular dynamics model of titanium alloy is simulated using the femtosecond laser processing parameters to be measured, and the lattice temperature distribution corresponding to the femtosecond laser processing parameters to be measured is obtained; The titanium alloy molecular dynamics model includes a titanium alloy polycrystalline model with α phase and β phase, as well as corresponding potential functions, temperature boundary conditions, a dual-temperature model, and related thermophysical material parameters; Under the NVT ensemble, the lattice temperature distribution corresponding to the femtosecond laser process parameter to be measured is loaded into the titanium alloy molecular dynamics model, and an iterative calculation is performed to obtain the electron temperature distribution to be measured at the interface between the α phase and the β phase; The thermal distribution prediction model of the titanium alloy processing interface is used to process the femtosecond laser process parameters to be measured and the electron temperature distribution to be measured, and the thermal distribution of the titanium alloy processing interface corresponding to the femtosecond laser process parameters to be measured is predicted; the thermal distribution prediction model of the titanium alloy processing interface is obtained by the method for constructing a thermal distribution prediction model of the titanium alloy processing interface as described in any one of claims 1 to 7.
9. An electronic device, characterized in that: include: at least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the method for constructing a thermal-mechanical distribution prediction model of a titanium alloy processing interface as described in any one of claims 1 to 7 or the method for predicting thermal-mechanical distribution of a titanium alloy processing interface as described in claim 8.
10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, which are used to enable a computer to execute the method for constructing a thermal distribution prediction model for a titanium alloy processing interface as described in any one of claims 1 to 7 or the method for predicting thermal distribution of a titanium alloy processing interface as described in claim 8.