Laser shock peening stress mapping model construction method and laser shock corrosion prevention method

Abstract

The embodiment of the application provides a laser shock peening stress mapping model construction method and a laser shock corrosion prevention method, and is applied to the technical field of laser processing and metal material surface modification. The model construction method comprises the following steps: performing laser shock experiment on a target material and extracting adaptive simulation parameters, then combining a plurality of laser process conditions to carry out simulation, fitting to obtain characteristic quantities such as surface residual compressive stress and compressive stress influence depth, and establishing a mapping model between the characteristic quantities and laser parameters through nonlinear regression. The method can improve the residual stress field prediction accuracy and process design efficiency, reduce the test cost, enhance the parameter migration, and provide a reliable basis for the laser shock corrosion prevention and service life improvement of the grounding grid. The technical effect of reducing the corrosion prevention cost is achieved.

Description

Technical Field

[0001] This application relates to the fields of laser processing and surface modification of metallic materials, and in particular to a method for constructing a laser shock strengthening stress mapping model and a laser shock corrosion prevention method. Background Technology

[0002] Grounding grids are core infrastructure for the safe operation of power systems, widely used in substations, transmission lines, and industrial facilities. Their primary function is to safely conduct lightning currents and fault currents underground, ensuring the safety of equipment and personnel. Grounding grids need to be buried in soil for extended periods; however, grounding grid materials are prone to uniform corrosion, pitting corrosion, and stress corrosion cracking in highly corrosive soils. This leads to increased ground resistance, decreased conductivity, and may even cause material fracture, seriously threatening the stability and safety of the power system. Therefore, how to improve the corrosion resistance of grounding grid materials through surface treatment technology has become a critical issue that urgently needs to be addressed in the field of power engineering.

[0003] In existing technologies, when applying corrosion protection to the grounding grid, a high-power short-pulse laser interacts with the metal surface to generate a plasma shock wave, thereby forming a deep residual compressive stress field inside the material to achieve the purpose of corrosion protection.

[0004] Because existing laser shock peening corrosion protection technology involves extensive physical experiments and trial and error, it suffers from high corrosion protection costs. Summary of the Invention

[0005] This application provides a method for constructing a laser shock stabilization stress mapping model and a laser shock corrosion prevention method, which aim to reduce corrosion prevention costs.

[0006] In a first aspect, embodiments of this application provide a method for constructing a laser shock strengthening stress mapping model, including:

[0007] Laser shock experiments were conducted on the target material based on a pre-designed experimental scheme to obtain experimental data.

[0008] Data analysis is performed based on experimental data to obtain simulation parameters corresponding to the target material;

[0009] Multiple laser shock simulations were performed on the target material based on simulation parameters and preset laser parameters, resulting in multiple simulation results. Each simulation result includes a set of simulated values ​​of preset laser parameters and a set of residual stress distribution data points.

[0010] For each set of residual stress distribution data points, a fitting calculation is performed to obtain a set of fitting parameter values; the fitting parameters include the surface residual compressive stress and the depth of compressive stress influence.

[0011] Nonlinear regression processing is performed on the simulated values ​​of each set of preset laser parameters and their corresponding fitted parameter values ​​to obtain the mapping relationship between each fitted parameter and the preset laser parameters, and a laser shock strengthening stress mapping model is generated based on the mapping relationship.

[0012] In one possible implementation, a laser shock experiment is conducted on the target material based on a preset experimental scheme to obtain experimental data, including:

[0013] Quasi-static tensile tests were conducted on the target material based on a pre-designed experimental scheme to obtain quasi-static stress-strain data.

[0014] Based on the pre-set experimental scheme, a split Hopkinson bar dynamic compression test was conducted on the target material to obtain dynamic stress-strain data at multiple strain rate levels.

[0015] High-temperature tensile tests were conducted on the target material based on a pre-designed experimental scheme to obtain high-temperature stress-strain data at multiple temperature levels.

[0016] Based on a pre-designed experimental scheme, the shock wave pressure during the laser shock process of the target material was measured to obtain shock wave pressure time history data.

[0017] In one possible implementation, data analysis is performed based on experimental data to obtain simulation parameters corresponding to the target material, including:

[0018] The quasi-static yield strength, strain hardening coefficient and hardening index in the simulation parameters are obtained by fitting the quasi-static stress-strain data.

[0019] The strain rate sensitivity coefficient in the simulation parameters is obtained by fitting dynamic stress-strain data.

[0020] The thermal softening index in the simulation parameters is obtained by fitting high-temperature stress-strain data.

[0021] Based on the shock wave pressure time history data, the energy conversion efficiency in the preset confined expansion model is calibrated, and the load-pressure curve in the simulation parameters is generated by combining the preset confined expansion model.

[0022] In one possible implementation, multiple laser shock simulations are performed on the target material based on simulation parameters and preset laser parameters to obtain multiple simulation results, including:

[0023] For each type of laser parameter in the preset laser parameters, determine the corresponding value range for the laser parameter;

[0024] For each laser parameter, values ​​are taken within its corresponding range to obtain multiple sets of simulated laser parameter values;

[0025] Laser shock simulation is performed on the target material based on multiple sets of simulated laser parameters, and residual stress distribution data points corresponding to each set of simulated laser parameters are obtained, and simulation results corresponding to each set of simulated laser parameters are generated.

[0026] In one possible implementation, laser shock simulation is performed on the target material based on multiple sets of simulated laser parameter values ​​and simulation parameters to obtain residual stress distribution data points corresponding to each set of simulated laser parameter values, including:

[0027] The load-pressure curve is used as the dynamic load, and the remaining parameters in the simulation parameters are used as material properties. Explicit dynamic analysis is performed based on the dynamic load and material properties to simulate the propagation of the dynamic load in the target material and the response of the target material, thus obtaining the final state of the dynamic response. The final state of the dynamic response is a snapshot of the instantaneous state of the target material at the moment the dynamic load ends.

[0028] Using the final state of the dynamic response as the initial condition, an implicit static analysis is performed on the target material after the dynamic load is removed to solve the static equilibrium residual stress field and obtain the residual stress distribution data points.

[0029] Secondly, embodiments of this application provide a laser shock corrosion protection method, which is implemented based on the laser shock strengthening stress mapping model of any one of the first aspects described above. The method includes:

[0030] Obtain the corrosion protection requirements for the target material, and generate the target residual stress field index corresponding to the target material based on the corrosion protection requirements;

[0031] An inverse optimization problem is constructed based on a pre-built laser shock strengthening stress mapping model and a target residual stress field index. The objective function of the inverse optimization problem is the laser processing energy consumption function, and the constraint condition is that at least one of the fitting parameters of the laser shock strengthening stress mapping model satisfies the target residual stress field index.

[0032] Solving the inverse optimization problem yields a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function.

[0033] Laser shock peening is performed on the target material according to the optimal laser parameters.

[0034] In one possible implementation, an inverse optimization problem is solved to obtain a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function, including:

[0035] Based on the range of laser parameters, multiple initial iteration points are generated;

[0036] For each initial iteration point, a sequential quadratic programming iteration is performed based on the inverse optimization problem to obtain a set of locally optimal laser parameter values ​​corresponding to each initial iteration point;

[0037] Compare the objective function values ​​corresponding to each group of locally optimal laser parameter values, and determine the group of locally optimal laser parameter values ​​with the lowest objective function value as the group of optimal laser parameter values ​​with the smallest laser processing energy consumption function value.

[0038] In one possible implementation, for each initial iteration point, a sequential quadratic programming iterative solution is performed based on the inverse optimization problem to obtain a set of locally optimal laser parameter values ​​corresponding to each initial iteration point, including:

[0039] In each iteration, the objective function is approximated twice at the current iteration point, and the constraints are approximated linearly at the current iteration point, thus obtaining the quadratic programming subproblem corresponding to the iteration.

[0040] Solve the quadratic programming subproblem to obtain the search direction;

[0041] The step size is determined by performing a line search based on the search direction, and the current iteration point is updated.

[0042] When the current iteration point meets the preset convergence condition, the current iteration point is determined as a set of locally optimal laser parameter values;

[0043] If the current iteration point does not meet the preset convergence condition, proceed to the next iteration round until the current iteration point meets the preset convergence condition.

[0044] In one possible implementation, the preset convergence conditions include:

[0045] The change in the objective function value between two consecutive iterations is less than a first preset threshold; and / or,

[0046] The constraint violation at the current iteration point is less than the second preset threshold.

[0047] In one possible implementation, the constraints of the inverse optimization problem also include the range of values ​​for the laser parameters; when solving the inverse optimization problem, the range of values ​​is used as the upper and lower boundary constraints of the laser parameters.

[0048] Thirdly, embodiments of this application provide a laser shock strengthening stress mapping model construction device, comprising:

[0049] The acquisition module is used to conduct laser shock experiments on target materials based on a preset experimental scheme and obtain experimental data.

[0050] The first processing module is used to perform data analysis based on experimental data to obtain the simulation parameters corresponding to the target material.

[0051] The second processing module is used to perform multiple laser shock simulations on the target material based on simulation parameters and preset laser parameters to obtain multiple simulation results; each simulation result includes a set of preset laser parameter simulation values ​​and a set of residual stress distribution data points;

[0052] The third processing module is used to perform fitting calculations on each set of residual stress distribution data points to obtain a set of fitting parameter values; the fitting parameters include surface residual compressive stress and compressive stress influence depth.

[0053] The fourth processing module is used to perform nonlinear regression processing based on the simulated values ​​of each set of preset laser parameters and their corresponding fitted parameter values, to obtain the mapping relationship between each fitted parameter and the preset laser parameters, and to generate a laser shock strengthening stress mapping model based on the mapping relationship.

[0054] In one possible implementation, the acquisition module is further configured to:

[0055] Quasi-static tensile tests were conducted on the target material based on a pre-designed experimental scheme to obtain quasi-static stress-strain data.

[0056] Based on the pre-set experimental scheme, a split Hopkinson bar dynamic compression test was conducted on the target material to obtain dynamic stress-strain data at multiple strain rate levels.

[0057] High-temperature tensile tests were conducted on the target material based on a pre-designed experimental scheme to obtain high-temperature stress-strain data at multiple temperature levels.

[0058] Based on a pre-designed experimental scheme, the shock wave pressure during the laser shock process of the target material was measured to obtain shock wave pressure time history data.

[0059] In one possible implementation, the first processing module is further configured to:

[0060] The quasi-static yield strength, strain hardening coefficient and hardening index in the simulation parameters are obtained by fitting the quasi-static stress-strain data.

[0061] The strain rate sensitivity coefficient in the simulation parameters is obtained by fitting dynamic stress-strain data.

[0062] The thermal softening index in the simulation parameters is obtained by fitting high-temperature stress-strain data.

[0063] Based on the shock wave pressure time history data, the energy conversion efficiency in the preset confined expansion model is calibrated, and the load-pressure curve in the simulation parameters is generated by combining the preset confined expansion model.

[0064] In one possible implementation, the second processing module is further configured to:

[0065] For each type of laser parameter in the preset laser parameters, determine the corresponding value range for the laser parameter;

[0066] For each laser parameter, values ​​are taken within its corresponding range to obtain multiple sets of simulated laser parameter values;

[0067] Laser shock simulation is performed on the target material based on multiple sets of simulated laser parameters, and residual stress distribution data points corresponding to each set of simulated laser parameters are obtained, and simulation results corresponding to each set of simulated laser parameters are generated.

[0068] In one possible implementation, the second processing module is further configured to:

[0069] The load-pressure curve is used as the dynamic load, and the remaining parameters in the simulation parameters are used as material properties. Explicit dynamic analysis is performed based on the dynamic load and material properties to simulate the propagation of the dynamic load in the target material and the response of the target material, thus obtaining the final state of the dynamic response. The final state of the dynamic response is a snapshot of the instantaneous state of the target material at the moment the dynamic load ends.

[0070] Using the final state of the dynamic response as the initial condition, an implicit static analysis is performed on the target material after the dynamic load is removed to solve the static equilibrium residual stress field and obtain the residual stress distribution data points.

[0071] Fourthly, embodiments of this application provide a laser shock osmosis corrosion protection device, comprising:

[0072] The acquisition module is used to acquire the anti-corrosion requirements for the target material and generate the target residual stress field index corresponding to the target material based on the anti-corrosion requirements.

[0073] The first processing module is used to construct an inverse optimization problem based on a pre-built laser shock strengthening stress mapping model and a target residual stress field index. The objective function of the inverse optimization problem is the laser processing energy consumption function, and the constraint condition is that at least one of the fitting parameters of the laser shock strengthening stress mapping model satisfies the target residual stress field index.

[0074] The second processing module is used to solve the inverse optimization problem to obtain a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function.

[0075] The third processing module is used to perform laser shock strengthening treatment on the target material according to the optimal laser parameters.

[0076] In one possible implementation, the second processing module is further configured to:

[0077] Based on the range of laser parameters, multiple initial iteration points are generated;

[0078] For each initial iteration point, a sequential quadratic programming iteration is performed based on the inverse optimization problem to obtain a set of locally optimal laser parameter values ​​corresponding to each initial iteration point;

[0079] Compare the objective function values ​​corresponding to each group of locally optimal laser parameter values, and determine the group of locally optimal laser parameter values ​​with the lowest objective function value as the group of optimal laser parameter values ​​with the smallest laser processing energy consumption function value.

[0080] In one possible implementation, the second processing module is further configured to:

[0081] In each iteration, the objective function is approximated twice at the current iteration point, and the constraints are approximated linearly at the current iteration point, thus obtaining the quadratic programming subproblem corresponding to the iteration.

[0082] Solve the quadratic programming subproblem to obtain the search direction;

[0083] The step size is determined by performing a line search based on the search direction, and the current iteration point is updated.

[0084] When the current iteration point meets the preset convergence condition, the current iteration point is determined as a set of locally optimal laser parameter values;

[0085] If the current iteration point does not meet the preset convergence condition, proceed to the next iteration round until the current iteration point meets the preset convergence condition.

[0086] In one possible implementation, the preset convergence conditions include:

[0087] The change in the objective function value between two consecutive iterations is less than a first preset threshold; and / or,

[0088] The constraint violation at the current iteration point is less than the second preset threshold.

[0089] In one possible implementation, the constraints of the inverse optimization problem also include the range of values ​​for the laser parameters; when solving the inverse optimization problem, the range of values ​​is used as the upper and lower boundary constraints of the laser parameters.

[0090] Fifthly, embodiments of this application provide an electronic device, including: a memory and a processor;

[0091] The memory stores instructions that the computer executes;

[0092] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect above and various possible implementations of the first aspect.

[0093] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement various possible implementations of the first and second aspects described above.

[0094] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements various possible implementations of the first and second aspects described above.

[0095] This application provides a method for constructing a laser shock peening stress mapping model and a laser shock corrosion protection method. The model construction method determines corresponding simulation parameters based on experimental data of the target material and performs multiple laser shock simulations using preset laser parameters to obtain simulated values ​​of the preset laser parameters and residual stress distribution data points. Fitting parameters such as surface residual compressive stress and compressive stress influence depth are extracted from the residual stress distribution data points, and a mapping relationship between each fitted parameter and the preset laser parameters is further established. This enables quantitative characterization and rapid prediction of the residual stress characteristics of laser shock peening, thereby improving the accuracy and efficiency of laser shock peening process parameter design for the target material. Then, a target residual stress field index is generated based on the corrosion protection requirements of the target material. An inverse optimization problem is constructed and solved based on the laser shock peening stress mapping model to obtain the optimal laser parameters, and laser shock peening treatment is implemented accordingly. This avoids multiple forward trial-and-error processes in laser shock corrosion protection, achieving the technical effect of reducing corrosion protection costs. Attached Figure Description

[0096] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0097] Figure 1 A flowchart illustrating the laser shock strengthening stress mapping model construction method provided in this application. Figure 1 ;

[0098] Figure 2 This is a schematic diagram of the laser shock wave pressure-time history curve corresponding to shock wave pressure-time history data for a target material.

[0099] Figure 3 A flowchart illustrating the laser shock strengthening stress mapping model construction method provided in this application. Figure 2 ;

[0100] Figure 4 Flowchart of the laser shock osmosis corrosion protection method provided in this application Figure 1 ;

[0101] Figure 5Flowchart of the laser shock osmosis corrosion protection method provided in this application Figure 2 ;

[0102] Figure 6 The flowchart of the laser shock osmosis corrosion protection method for Q235 steel provided in this application is as follows. Figure 1 ;

[0103] Figure 7 The flowchart of the laser shock osmosis corrosion protection method for Q235 steel provided in this application is as follows. Figure 2 ;

[0104] Figure 8 A comparative schematic diagram showing the residual stress depth distribution after LSP treatment of Q235 steel, provided in this application.

[0105] Figure 9 A comparative diagram of the empirical solution and the SQP optimization solution for Q235 steel provided in this application;

[0106] Figure 10 A schematic diagram of the structure of the laser shock strengthening stress mapping model construction device provided in this application;

[0107] Figure 11 A schematic diagram of the laser shock osmosis protection device provided in this application;

[0108] Figure 12 A schematic diagram of the structure of the electronic device provided in this application.

[0109] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0110] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0111] First, let me explain the terms used in this application:

[0112] The Johnson-Cook model refers to an empirical viscoplastic dynamic constitutive model, which is widely used in simulations of impact dynamics, explosions, penetration, laser shock, and high-speed cutting.

[0113] Fabbro confined laser-induced plasma shock wave model: refers to a one-dimensional analytical model in the field of laser shock strengthening, used to describe the evolution of target shock wave pressure driven by plasma generated by laser irradiation under confined layer conditions and plasma expansion.

[0114] Savitzky-Golay filter: refers to a digital filter that uses local polynomial least squares fitting in the time domain. The corresponding filtering method is to use a sliding window mechanism to filter discrete signals.

[0115] The Levenberg-Marquardt algorithm is a numerical iterative optimization algorithm used for nonlinear least squares problems.

[0116] The Courant–Friedrichs–Lewy (Courant) stability condition refers to the numerical stability criterion that must be satisfied when solving partial differential equations using numerical methods such as the finite difference method, finite element method, and finite volume method.

[0117] Mechanical Testing System (MTS): This refers to a multi-functional mechanical testing device for materials with electro-hydraulic servo drive as its core. It can carry out quasi-static, medium-to-high strain rate tensile, compression, bending, fatigue, creep, and high-temperature mechanical property tests.

[0118] Taylor Bar Impact Test: This refers to a high-strain-rate dynamic mechanical testing method in which a cylindrical specimen is impacted at high speed against a rigidly fixed target plate. Large plastic deformation occurs at the impact end. The deformation profile and length change are measured by recovering the specimen, and the dynamic yield strength and flow stress of the material are inferred from the theory.

[0119] Polyvinylidene fluoride (PVDF) is a type of piezoelectric polymer material. PVDF films, after polarization treatment, exhibit significant positive and negative piezoelectric effects. PVDF shock wave pressure measurement involves applying a plasma shock wave to the surface of a PVDF piezoelectric sensor film. The shock load causes the film to be compressed and generate an electrical charge. A charge amplifier collects the charge signal, which, after calibration, is converted into a time-history pressure curve, enabling dynamic measurement of the shock wave pressure.

[0120] In existing technologies, when using laser shock peening to enhance corrosion resistance of grounding grid materials, the main approach is to create an impact effect on the material surface through laser action, introducing residual compressive stress within the material to improve its corrosion and crack resistance. The overall corrosion protection process typically requires first determining the laser parameters, then combining experiments or simulations to observe the distribution of residual stress, and repeatedly adjusting the process parameters accordingly.

[0121] However, existing laser shock peening corrosion protection technology requires repeated trial and error to test the specific corrosion protection effect on the target material under different laser settings when used for grounding grid corrosion protection. This forward trial-and-error approach requires a large number of physical experiments, resulting in high corrosion protection costs in existing technologies.

[0122] To address the aforementioned technical problems, this application proposes the following technical concept: using data obtained from physical experiments to determine the simulation parameters required for laser shock simulation of the target material; using these simulation parameters to perform various laser parameter simulations on the target material and obtain corresponding simulation results; and using the simulation results for further fitting calculations and nonlinear regression processing to obtain a laser shock strengthening stress mapping model that reflects the mapping relationship between the fitted parameters and the laser parameters. This model can be used to generate corresponding laser parameter configurations based on the parameter requirements of different corrosion protection needs. Compared with existing technologies, this application can effectively reduce the number of physical experiments, thereby achieving the technical effect of reducing corrosion protection costs.

[0123] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0124] Figure 1 A flowchart illustrating the laser shock strengthening stress mapping model construction method provided in this application. Figure 1 ,like Figure 1 As shown, the method includes:

[0125] S101. Based on the preset experimental scheme, a laser shock experiment is conducted on the target material to obtain experimental data.

[0126] In this step, the target material serves as the object of processing in laser shock experiments, simulations, and stress mapping modeling. Specifically, a Q235 steel sample used for grounding grids can be selected. To ensure a consistent data foundation for subsequent simulation parameter extraction, plate-shaped or block-shaped samples of uniform size are first processed according to sample preparation specifications. The sample surfaces are then mechanically polished, degreased, and cleaned to ensure that the surface roughness and contamination status of the laser-affected area are within a controllable range. A pre-defined experimental scheme is used to pre-define the implementation conditions and related test conditions of the laser shock experiment. This includes at least the laser pulse energy, spot diameter, pulse width, overlap rate, constraint layer setting method, absorption layer setting method, number of impact locations, test temperature range, and number of samples. The laser shock experiment is used to obtain experimental data on the target material under laser shock conditions. In practice, an absorption layer and a constraint layer can be set on the sample surface, followed by impacting the target area with a pulsed laser to form confined plasma and input impact pressure onto the material surface. The experimental data serves as the basic input for subsequent data analysis and simulation parameter extraction, including not only the response data after laser shock but also a dataset used to establish the material response and load boundaries.

[0127] In this step, experimental data may include raw load signals, displacement signals, strain signals, temperature conditions, and laser parameter records. For each set of samples, a unified data identifier can be established after data acquisition, writing the sample number, material batch, heat treatment state, surface treatment state, and corresponding experimental conditions into the database to ensure that various types of data can be associated according to the same sample source during subsequent analysis. If a set of data has obvious noise, sensor saturation, or data anomalies caused by sample processing defects, the anomaly is marked while retaining the original records, and the same test conditions are supplemented again to ensure the repeatability of the data used in the subsequent fitting process.

[0128] Alternatively, one possible way to obtain experimental data through laser shock experiments is as follows:

[0129] S1011. Based on the preset experimental scheme, a quasi-static tensile test is conducted on the target material to obtain quasi-static stress-strain data.

[0130] In this step, quasi-static tensile testing is used to obtain the basic mechanical properties of the target material at low strain rates, including elastic modulus, yield strength, tensile strength, and plastic flow stress. Quasi-static stress-strain data refers to the trajectory of the correspondence between axial stress and axial strain of the target material directly recorded under constant low-speed loading conditions, reflecting the basic plastic flow behavior of the material in thermodynamic equilibrium.

[0131] Optionally, the quasi-static tensile test can be implemented as follows: The pretreated specimen is installed between the upper and lower clamps of an electronic universal testing machine, and the extensometer is clamped in the gauge length section of the specimen. The ambient temperature and strain rate are set according to a preset scheme. The testing machine is started to apply an axial tensile load at a constant low speed. The force value output by the load cell and the displacement value output by the extensometer are collected in real time and transmitted to the host computer via a data acquisition card. The data is then converted into an engineering stress-strain curve and further converted into true stress-true strain data.

[0132] For example, Q235 steel specimens for grounding grids were selected as the target material. These specimens were wire-cut into standard tensile specimens with a rectangular cross-section, a gauge length of 50 mm, a width of 12.5 mm, and a thickness of 6 mm. The parallel sections were sequentially polished or sandblasted with 80#, 400#, and 800# sandpaper to control the surface roughness between 0.8 and 1.6 μm. Then, the specimens were ultrasonically cleaned in acetone and anhydrous ethanol for 15 minutes each to remove oil contaminants. An absorption layer and a constraint layer were then prepared, resulting in five duplicate specimens to reduce dispersion. The experiment was conducted on a computer-controlled electronic universal testing machine, with the ambient temperature maintained stably at 298 K ± 2 K using a constant temperature chamber. The beam displacement rate was set to achieve a strain rate of 1 × 10⁻³ s⁻¹ for the gauge section of the specimen. A full-range clamp-on extensometer was used to measure the real-time deformation, and the data acquisition system synchronously recorded the load and deformation at a sampling frequency of 100 Hz. After the test, the load-displacement data were converted into engineering stress-strain curves, and then corrected into true stress-strain curves based on the principle of constant volume, thus obtaining quasi-static stress-strain data.

[0133] S1012. Based on the preset experimental scheme, a split Hopkinson bar dynamic compression test was conducted on the target material to obtain dynamic stress-strain data at multiple strain rate levels.

[0134] In this step, the split Hopkinson bar dynamic compression test can employ a testing system consisting of an incident bar, a transmission bar, and a projectile. Stress wave signals are acquired under multiple strain rate conditions, and dynamic stress-strain curves are calculated. Dynamic stress-strain data refers to the dynamic stress and strain response trajectory of the material at high strain rates, obtained by directly converting the collected incident wave, reflected wave, and transmitted wave voltage signals using one-dimensional stress wave theory formulas.

[0135] Optionally, the dynamic compression test of the split Hopkinson bar can be implemented as follows: The target material is processed into a small cylindrical compression specimen, which is clamped between the incident bar and the transmission bar. The air gun device is activated, and the projectile is driven by adjusting the air pressure to impact the end of the incident bar at different velocities, generating a compressive stress wave in the incident bar. When the stress wave propagates to the specimen interface, part of it is reflected back to the incident bar, and part is transmitted to the transmission bar. Resistance strain gauges attached to the incident bar and the transmission bar collect the voltage signals of the incident wave, reflected wave, and transmitted wave, which are amplified by a hyperdynamic strain gauge and recorded by a high-speed digital oscilloscope. Based on the one-dimensional stress wave theory and the homogeneity assumption, the three-wave signals are converted into the dynamic stress-strain curve of the specimen, obtaining the dynamic yield strength and flow stress at multiple strain rate levels.

[0136] For example, based on a pre-defined experimental scheme, the dynamic compression test of the split Hopkinson bar was determined to use a steel incident bar and transmission bar with a diameter of 14.5 mm, an elastic modulus of 210 GPa, and a wave velocity of 5190 m / s. The target material was Q235 steel. The pretreated sample was machined into a cylinder with a diameter of 8 mm and a thickness of 4 mm, with the parallelism of both end faces less than 0.02 mm, and coated with Vaseline for lubrication and friction reduction. The projectile impact velocities were set to 6 m / s, 12 m / s, and 18 m / s, respectively, corresponding to three dynamic compression conditions with effective strain rates of approximately 1200 s⁻¹, 2500 s⁻¹, and 4800 s⁻¹. Each condition was repeated 3 times. Semiconductor strain gauges (resistance 120 Ω, sensitivity coefficient 2.1) attached to the incident bar and transmission bar were connected to a super-dynamic strain gauge (frequency response 1 MHz), and the waveform was acquired by an oscilloscope at a sampling rate of 10 MHz. Subsequently, the waveform distortion segment was removed by the three-wave processing program, and the average stress and strain values ​​during the overlapping time period of the reflected wave and the transmitted wave were taken. Dynamic true stress-true strain curves at various strain rates were plotted to obtain the dynamic yield strength and flow stress at multiple strain rate levels.

[0137] S1013. Based on the preset experimental scheme, conduct high-temperature tensile tests on the target material to obtain high-temperature stress-strain data at multiple temperature levels.

[0138] In this step, the high-temperature tensile test is conducted in a high-temperature furnace with a tensile fixture. High-temperature stress-strain data are obtained at different temperature points. High-temperature stress-strain data refers to the stress-strain trajectory of the material at a specific temperature point under quasi-static strain rate tension in a high-temperature furnace, directly recorded to reflect the influence of thermal activation on the material's flow stress.

[0139] Optionally, the high-temperature tensile test can be implemented as follows: The tensile specimen is mounted on a tensile testing machine equipped with a high-temperature environmental furnace, with the gauge length of the specimen placed in the uniform temperature zone of the furnace. After closing the furnace door, the specimen is heated to the target temperature according to a preset heating curve and held at that temperature for a sufficient time to ensure uniform internal temperature. Subsequently, a tensile load is applied at the same strain rate as in the quasi-static test. Since the extensometer is used at high temperatures, a high-temperature ceramic rod extensometer or a non-contact video extensometer is required for deformation measurement. Temperature, load, and displacement are collected throughout the test and converted into high-temperature stress-strain data at various temperature levels.

[0140] For example, Q235 plate tensile specimens of the same batch and size as those used in step S1011 were used. The high-temperature tensile test was conducted on an electronic universal testing machine equipped with a three-stage resistance heating furnace. Six temperature levels were set: 100℃, 200℃, 300℃, 400℃, 500℃, and 600℃, corresponding to 373K, 473K, 573K, 673K, 773K, and 873K, with a temperature control accuracy of ±3K. After specimen installation, the temperature was increased to the target temperature at a rate of 10K / min and held for 15 minutes to ensure temperature uniformity. After the temperature stabilized, the tensile test was initiated at the same beam displacement rate as in S1011. Due to the limitations of contact extensometers in high-temperature environments, laser video extensometers were used for non-contact real-time tracking of gauge length deformation. The data acquisition frequency was also set to 100Hz. Three duplicate specimens were tested at each temperature point, resulting in a total of 18 sets of data. After the experiment was completed, the load-displacement data were deducted to remove the influence of system flexibility caused by the thermal expansion of the high-temperature furnace, and the true stress-strain curves at each temperature were obtained to obtain high-temperature stress-strain data at multiple temperature levels.

[0141] S1014. Based on the preset experimental scheme, the shock wave pressure of the target material during the laser shock process is measured to obtain the shock wave pressure time history data.

[0142] In this step, the actual measurement of shock wave pressure can be achieved by deploying a pressure sensor or piezoelectric pressure sensing element during the laser shock process, recording the pressure change signal over time in real time, and forming a pressure time history curve. The selected model can meet the requirements of high-frequency shock signal acquisition. In practical applications, other models of this component can also be selected, and this application does not limit this. The shock wave pressure time history data obtained in the experiment refers to the continuous waveform record of the transient shock wave pressure acting on the target surface under the actual laser shock conditions, captured by the high-frequency piezoelectric sensor, as it evolves with real physical time; it is a purely time-domain signal.

[0143] Optionally, one possible implementation of shock wave pressure measurement is as follows: A test target plate made of the same material as the target material is placed below the laser shock beam path. A high-frequency response pressure sensor (such as a PVDF piezoelectric film sensor or a quartz piezoelectric meter) is embedded below the surface of the target plate or attached to the back of the target plate. The laser is activated to perform a single impact according to preset parameters (power density, spot diameter, pulse width). The sensor converts the sensed transient pressure into a charge / voltage signal, which is amplified by a high-frequency charge amplifier and the complete pressure waveform is recorded by a high-speed oscilloscope (sampling rate ≥1GS / s). The voltage signal is converted into pressure amplitude through calibration coefficients, and characteristic parameters such as peak pressure, full width at half maximum (FWHM) pulse width, and rise time are extracted and converted into a load-time curve that can be imported into finite element software.

[0144] For example, a Q235 steel plate of the same thickness as in step S1011 is used as the test substrate, and its surface is polished. A blind hole with a diameter of 2 mm and a depth of 1 mm is opened at the center of the back of the substrate, and a PVDF piezoelectric thin film sensor (model: PCB138A10, response frequency >1MHz, range 0~10GPa) is embedded therein. The sensor surface is flush with the back of the substrate and fixed with epoxy resin insulation. The laser impact is perpendicularly incident according to a preset laser impact scheme (e.g., power density 2.5GW / cm², spot diameter 3 mm, pulse width 20ns). The laser used is an Nd:YAG nanosecond pulse laser (wavelength 1064nm, energy stability ≤5%). During the impact, the charge signal output by the PVDF sensor is converted by a charge amplifier (frequency response 0.1Hz~1MHz) and then connected to a digital oscilloscope (bandwidth 1GHz, sampling rate 2.5GS / s) for waveform recording. The impact is repeated 5 times under each combination of process parameters, and the average waveform is taken as the characteristic pressure-time history of that condition. The acquired voltage-time waveforms were converted into pressure-time curves using calibration curves (pre-acquired using the Hopkinson bar dynamic calibration method) to obtain shock wave pressure time history data.

[0145] It should be noted that this step characterizes the response features of the target material under quasi-static, dynamic, high-temperature, and laser shock conditions, enabling subsequent simulation parameters to be calibrated and fitted based on measured data. The pressure time history data can also be used to constrain the input form of the laser shock load, thereby improving the consistency between simulation conditions and actual working conditions. Using the above experimental scheme, a complete data foundation covering the static, rate effects, thermal effects, and shock load characteristics of the material can be formed, providing reliable input for establishing a laser shock strengthening stress mapping model.

[0146] S102. Based on the experimental data, perform data analysis to obtain the simulation parameters corresponding to the target material.

[0147] In this application, after obtaining the experimental data, a unified data analysis is required on the raw test results from different sources to form simulation parameters that can be directly used for laser shock simulation. These simulation parameters characterize the material properties and / or load characteristics of the target material in the laser shock simulation. The simulation parameters include the JC parameters corresponding to the Johnson-Cook model and the load-pressure curves corresponding to the Fabbro plasma shock wave model.

[0148] Optionally, data analysis can be implemented by: preprocessing experimental data; performing inversion, fitting, or calibration according to the parameter requirements of the preset material and load models, and finally outputting a set of simulation parameter files corresponding to the target material. Here, data analysis is not simply a summary of experimental results, but rather the transformation of raw test signals into material parameters and boundary condition parameters that can be used by finite element simulation or other numerical calculation models. Pre-data analysis includes, but is not limited to: performing zero-point correction, filtering, calibration, time axis alignment, parameter conversion, and curve fitting on experimental data, and extracting parameters for simulation calculations based on the response characteristics of the target material under laser shock conditions.

[0149] For example, if a certain experimental data point has a large dispersion, it can be corrected by averaging multiple sets of parallel experiments, removing outliers that exceed the statistical threshold, and using least squares or robust fitting methods to obtain stable simulation parameters.

[0150] Alternatively, one possible way to obtain simulation parameters through data analysis is as follows:

[0151] S1021. Based on the quasi-static stress-strain data, the quasi-static yield strength, strain hardening coefficient, and hardening index in the simulation parameters are obtained through fitting.

[0152] In this step, the quasi-static yield strength refers to the initial yield stress corresponding to a 0.2% shift in plastic strain under reference strain rate and temperature conditions; the strain hardening coefficient is a coefficient reflecting the increase in the material's resistance to plastic deformation; and the hardening exponent is a dimensionless exponent used to characterize the steepness of the hardening curve. The quasi-static yield strength, strain hardening coefficient, and hardening exponent are all core parameters in the (Johnson-Cook, JC) model, and the expression for the JC model is shown in Equation 1:

[0153]

[0154] in, Equivalent flow stress (MPa). Equivalent plastic strain (dimensionless). strain rate (s) -1 ), The reference strain rate (taken as 1 s) -1 ), T is a dimensionless temperature. r For reference temperature, T m The melting point of the target material. A jc The yield strength (MPa) under reference conditions is the quasi-static yield strength in this step; B jc n is the strain hardening coefficient (MPa). jc C is the hardening index. jc Let m be the strain rate sensitivity coefficient. jc This is the thermal softening index.

[0155] Alternatively, one possible way to obtain the above parameters by fitting quasi-static stress-strain data is as follows:

[0156] a1. Based on the true stress-true strain curves corresponding to the quasi-static stress-strain data, the uniform plastic deformation segment from the yield point to the maximum load point is selected as the effective fitting interval.

[0157] a2. Using the 0.2% offset method, the stress value at the intersection of the curve and the line offset by 0.2% strain along the elastic segment is determined as A. jc The initial value.

[0158] a3. Select no fewer than 20 data points at equal intervals within the plastic range (the strain range of these data points is 0.005~0.15), and calculate the corresponding values ​​for each point. and .

[0159] a4, with The x-axis is... Perform a linear least squares fit on the ordinate and solve the slope and intercept of the linear equation to obtain B. jc With n jc， and the current goodness of fit R 2 Among them, B jc n is the exponential value of the intercept of the linear equation. jc This represents the slope of the linear equation. If the goodness-of-fit R² is lower than a preset value (which can be set to 0.99), then A is fine-tuned. jc The value (with a fine-tuning range that can be set to ±5 MPa) regresses until R... 2 It has reached its maximum value.

[0160] S1022. Based on the dynamic stress-strain data, the strain rate sensitivity coefficient in the simulation parameters is obtained by fitting.

[0161] In this step, the strain sensitivity coefficient refers to a dimensionless coefficient used to characterize the rate at which flow stress increases with the logarithmic growth of the strain rate. In the JC model, this term is expressed as... The thermal softening index exists in the form of a coefficient of performance (CQS) to reflect the degree of additional strengthening of a material as it transitions from a quasi-static state to an impact dynamic state. The thermal softening index is a dimensionless index used to characterize the rate at which flow stress decays with increasing dimensionless temperature. Both the strain rate sensitivity coefficient and the thermal softening index are core parameters of the JC model.

[0162] Alternatively, one possible way to obtain the strain rate sensitivity coefficient based on fitting dynamic stress-strain data is as follows:

[0163] b1. Under room temperature conditions (T=T) r Under the same plastic strain value, the dynamic flow stress corresponding to different strain rates is extracted, and the reference flow stress calculated by S1021 at that plastic strain rate is obtained. .

[0164] b2. Flow stress data under different strain rates were analyzed. Linear fitting yields the strain rate sensitivity coefficient.

[0165] In this step, the specific method for obtaining the strain rate sensitivity coefficient is as follows: calculate the ratio σ / σ0 of the dynamic flow stress and the reference flow stress at different strain rates, and then... The x-axis is (σ / σ) 0-1 A scatter plot was plotted with the vertical axis as the ordinate, and a least-squares linear regression was performed through the origin. The slopes obtained at multiple plastic strain points were averaged to obtain the strain rate sensitivity coefficient.

[0166] S1023. Based on the high-temperature stress-strain data, the thermal softening index in the simulation parameters is obtained through fitting.

[0167] Alternatively, one possible way to obtain the thermal softening index by fitting high-temperature stress-strain data is as follows:

[0168] c1, at the reference strain rate Under the same conditions, the high-temperature flow stress σ at different test temperatures T was extracted by selecting the same plastic strain value. T And the quasi-static flow stress σ at room temperature under the same strain rate static .

[0169] c2. Compare high-temperature flow stress and quasi-static flow stress, through... vs The thermal softening index was obtained by fitting the data.

[0170] Specifically, this step involves calculating the dimensionless temperature T based on the set temperature levels and the melting point of the material. * Then, the flow stress decay at each temperature is calculated, expressed as ln(T) * ) as the x-axis, ln(1-σ T / σ static Using the ordinate as the vertical axis, a least-squares linear regression is performed, and the slope of the resulting regression line is the thermal softening index.

[0171] S1024. Based on the shock wave pressure time history data, calibrate the energy conversion efficiency in the preset confined expansion model, and generate the load pressure curve in the simulation parameters by combining the preset confined expansion model.

[0172] In this step, the pre-defined confined expansion model refers to the analytical model of plasma shock wave pressure under confinement layer based on Fabbro theory. This model contains two core formulas, as shown in Formula 2 and Formula 3:

[0173]

[0174]

[0175] Equation 2 refers to the peak pressure equation, which maps the laser power density I0 and the material / confinement impedance Z to the peak pressure of the shock wave through the energy conversion efficiency α. Equation 3 refers to the pressure time history equation, which describes the complete time evolution of the shock wave from zero to its peak value, followed by exponential decay. peak The peak pressure of the shock wave is referred to as P(t), where α is the conversion efficiency of laser energy to plasma internal energy, and Z is the reduced impact impedance of the target material's confinement layer and target material. P(t) refers to the pressure at time t, and τ refers to the moment when the shock wave reaches its peak. The effective duration of the pressure pulse is approximately 2–3τ. For example, when τ = 15 ns, the effective duration of the shock wave is approximately 30–45 ns. The energy conversion efficiency α characterizes the proportion of laser energy converted into plasma internal energy. This parameter cannot be directly and accurately calculated theoretically; its value is highly dependent on the optical absorption characteristics of the interface between the confinement layer, absorption layer, and target material, as well as the plasma expansion dynamics. The energy conversion efficiency α is determined by back-calculating the specific parameter value using the measured peak pressure. The load-pressure curve refers to obtaining the peak pressure P of the shock wave by substituting the calibrated energy conversion efficiency α back into the peak pressure formula. peak This is combined with the pressure time history equation to generate a discretized P(t) numerical sequence that can be imported into finite element software. For example, Figure 2 This is a schematic diagram of the laser shock wave pressure-time history curve corresponding to shock wave pressure-time history data for a target material, such as... Figure 2As shown, the horizontal axis represents time and the vertical axis represents shock wave pressure. The overall pressure time history equation P(t) corresponds to the theoretical peak shock wave pressure P. peak The pressure is 4.18 GPa, and the corresponding shock wave reaches its peak value τ at 15 ns.

[0176] Alternatively, one possible way to calibrate the energy conversion efficiency and generate load-pressure curves is as follows:

[0177] d1. Based on the shock wave pressure time history data, the measured peak pressure under this working condition is extracted using a peak detection algorithm. Simultaneously, the time taken for the pressure-time waveform to rise from zero pressure to the peak value is measured as the measured rise time, and the full width at half maximum (FWHM) pulse width corresponding to the pressure decaying to half the peak value is recorded. These characteristic quantities serve as the target reference values ​​for subsequent calibration.

[0178] d2. Based on the physical properties of the constraint layer and the target material (i.e., the target material), the impact impedance Z is pre-calculated as a constant.

[0179] In this step, the impact impedance is calculated as shown in Formula 4:

[0180]

[0181] Where Z1 is the impact impedance of the confinement layer, Z1 = ρ1 × c1, where ρ1 refers to the density of the confinement layer and c1 refers to the sound velocity in the confinement layer. Z2 is the impact impedance of the target material, Z2 = ρ2 × c2, where ρ2 refers to the density of the target material and c2 refers to the sound velocity in the target material.

[0182] For example, the constraint layer has a water density of 1.0 g / cm³. 3 The speed of sound is 1.65 × 10⁻⁶. 5 cm / s; The target material is Q325 steel with a density of 7.85 g / cm³. 3 The speed of sound is 5.04 × 10⁻⁶. 5 cm / s. Therefore, the calculated Z1 = 0.165 × 10⁻⁶ cm / s. 6 g·cm -2 ·s -1 Z2 = 3.96 × 10 6 g·cm -2 ·s -1 The final impact impedance Z = 0.317 × 10⁻⁶ 6 g·cm -2 ·s -1 .

[0183] d3. Substitute the extracted measured peak pressure into Formula 2 to obtain the measured calibrated value of the energy conversion efficiency corresponding to this impact.

[0184] d4. For impact conditions under different laser power densities, calculate the measured calibrated values ​​of their respective energy conversion efficiencies, and then take the arithmetic average to obtain the energy conversion efficiency.

[0185] d5. Take the measured rise time corresponding to the shock wave pressure time history data as the moment when the shock wave reaches its peak value, and complete the calibration of the time base parameters.

[0186] In this step, if the original pressure-time waveform has oscillation interference, the rising segment is smoothed and filtered, and the point where the first derivative is zero is taken as the peak moment.

[0187] d6. Substitute the energy conversion efficiency determined in step d4 into Formula 2 above to obtain the peak pressure P of the shock wave. peak Substituting the time τ at which the shock wave reaches its peak value, determined in step d5, into Equation 3, the complete theoretical pressure time history function is discretized and reconstructed in the time domain, as shown in Equation 5:

[0188]

[0189] Among them, t end This refers to the total duration of the shock wave pressure time history curve, that is, the entire time window from the start of the laser shock to the point where the shock wave pressure completely decays to a negligible level. It is determined based on the actual recorded length of the pressure-time waveform obtained in step S1014 above. The time step between each discrete time point is set according to the stability limit of the subsequent finite element simulation, and can range from 0.5 to 1 ns. A discrete sequence is obtained using a discretized function, which serves as the load-pressure curve in the simulation parameters.

[0190] It should be noted that after obtaining the simulation parameters, a data input table for the simulation solver can be generated, including material property parameters and load-related parameters. If the subsequent simulation software supports user material subroutines, these parameters can also be written into the constitutive card or subroutine interface file. Based on the above analysis, this step realizes the conversion from experimental data to simulation parameters, enabling the constitutive behavior and impact load of the target material under laser shock conditions to be parameterized, thereby providing a unified, callable, and experimentally correlated input basis for subsequent numerical simulations under multiple sets of laser parameters. It should be understood that the above example is only for demonstration and is not limiting.

[0191] S103. Based on the simulation parameters and preset laser parameters, multiple laser shock simulations were performed on the target material to obtain multiple simulation results.

[0192] In this step, each simulation result includes a set of preset laser parameter simulation values ​​and a set of residual stress distribution data points. The preset laser parameters are used as input variables to control the laser shock simulation and establish the mapping relationship, and may specifically include laser power density, spot diameter, pulse width, overlap rate, and number of impacts.

[0193] Optionally, in this step, multiple simulation results can be generated by inputting multiple sets of preset laser parameter simulation values ​​and simulation parameters into the laser shock simulation model, performing multiple calculations on the target material, and obtaining multiple simulation results. Laser shock simulation is used to simulate the response process of the target material under laser shock in a computational environment. Specifically, a sample geometric model, mesh model, material model, and pressure boundary condition model can be established in a finite element platform, and the residual stress field can be solved during the loading stage and the fully relaxed stage after unloading.

[0194] In this step, the preset laser parameter simulation values ​​are used to characterize the specific parameter combination used in the current simulation, such as a set of laser pulse energy, spot diameter, and pulse width values. The residual stress distribution data points are used to characterize the distribution of residual stress inside the material along the depth direction after laser impact. Specifically, Mises equivalent stress, principal stress, or residual stress values ​​in a certain direction can be extracted at fixed depth intervals on the central axis of the spot, the near-surface normal path, or a predetermined cross-section.

[0195] Optionally, during simulation, the target material can be modeled as a two-dimensional axisymmetric model or a three-dimensional solid model. When meshing this model, a fine mesh is used for the laser-affected area and near-surface layer, while a relatively coarse mesh is used for areas far from the impact zone, balancing stress gradient analysis capability and solution efficiency. Regarding setting the model's boundary conditions, rigid body displacement of the specimen's bottom surface can be constrained, and symmetrical or free boundaries can be set for the sides based on actual symmetry. For setting the model's load input, the load-related parameters obtained in S102 can be applied to the impacted surface area, and the residual stress results of the material after unloading and stabilization can be extracted after the simulation is completed.

[0196] Optionally, to ensure consistency with the subsequent fitting parameter definitions, a sequence of discrete residual stress points along the thickness direction can be extracted, and the depth coordinates and compressive stress values ​​corresponding to each data point can be recorded. If local numerical oscillations or mesh distortions occur in multiple simulation results, the simulation can be recalculated by reducing the time step, adjusting the element type, increasing the mesh density, or using a smooth contact algorithm to ensure the continuity and fitability of the residual stress distribution data points.

[0197] It should be noted that the specific implementation process of this simulation step is as follows: Figure 3 Further explanation will be provided in the embodiments shown, and will not be repeated here.

[0198] S104. Perform fitting calculations for each set of residual stress distribution data points to obtain a set of fitting parameter values.

[0199] In this step, the residual stress distribution data points are discrete depth stress samples. When this data is directly used for comparison between different simulation results, the data dimensionality is high and it's not convenient for unified regression processing. Therefore, it's necessary to perform fitting calculations on each set of residual stress distribution data points to extract fitting parameters that can stably characterize the residual stress field. The fitting parameters, as a set of parameters representing the residual stress distribution characteristics, include surface residual compressive stress and the depth of influence of compressive stress. Surface residual compressive stress characterizes the compressive stress level on the surface of the target material after laser shock, specifically defined as the fitted value of the residual compressive stress at the material surface or at the first effective sampling layer below the surface. The depth of influence of compressive stress characterizes the depth to which the compressive stress introduced by the laser shock extends within the material, specifically defined as the depth position corresponding to when the residual stress changes from compressive stress to zero or to tensile stress, or as the depth position when the absolute value of the compressive stress decays to a threshold. However, the judgment rules remain consistent within the same modeling task.

[0200] Alternatively, one possible implementation of the fitting calculation is as follows:

[0201] e1. Sort the residual stress distribution data points of each group according to the depth coordinates and detect outliers.

[0202] e2. For abnormal fluctuation points caused by local numerical noise, adjacent point smoothing, moving average, Savitzky-Golay filtering or spline-based smoothing can be used to maintain the overall continuity of the stress distribution curve.

[0203] e3. Based on the distribution law of the residual stress of the material along the depth direction, which is usually characterized by the peak value of the surface compressive stress, the gradual decay of the internal stress, and the possibility of crossing zero, a fitting function is selected for parameterization.

[0204] In this step, the fitting function can be an exponential decay function, a piecewise polynomial function, a Gaussian superposition function, or a spline interpolation function. If a piecewise fitting method is used, the near-surface compressive stress decay segment can be fitted first, followed by the deep transition segment, and the overall curve smoothness can be ensured through the function continuity condition.

[0205] For example, the fitting function is shown in Equation 6:

[0206]

[0207] in, This represents the surface residual compressive stress (MPa, negative values ​​indicate compressive stress). The depth of compressive stress influence (μm, defined as follows) The depth at which the stress first crosses zero becomes tensile stress), p and q are fitting parameters controlling the shape of the stress distribution (p controls the peak position, q controls the decay rate). The Levenberg-Marquardt algorithm is used to fit the four parameters. .in, This reflects the compressive stress level in the near-surface region. The larger the value, the stronger the anodic dissolution inhibition effect; This reflects the thickness of the compressive stress protective layer. The larger the value, the better the suppression effect on deep stress corrosion cracks. p and q together control the shape of stress distribution—the stress distribution of typical laser shock peening (LSP) exhibits the characteristics of "slight increase near the surface → reaching peak value → gradual decay → crossing zero → shallow tensile stress equilibrium layer".

[0208] e4. After fitting, extract the stress value corresponding to the surface from the fitted curve as the surface residual compressive stress, and calculate the depth of the intersection point between the curve and the zero stress baseline as the compressive stress influence depth. For cases where the stress value never crosses zero, the influence depth can be determined according to a pre-set compressive stress threshold, and the same judgment criterion can be uniformly applied in the samples.

[0209] Alternatively, in one possible implementation, the residual stress distribution data points can be represented as the depth variable z0 and the corresponding residual stress σ. s The discrete correspondence, the essence of fitting calculation is to construct σ. r =f(z0) is a continuous function, and representative quantities are extracted from the function. Here, f(z0) reflects the overall trend of residual stress variation with depth; the surface residual compressive stress corresponds to the function value when z approaches zero; and the depth of influence of compressive stress corresponds to the z0 value when f(z0) satisfies the zero-crossing condition or the threshold condition. By uniformly compressing each set of discrete data points into two core fitting parameters, the output dimension of the subsequent regression model can be significantly reduced, while retaining the two indicators most important to process design: residual compressive stress intensity and depth of influence. For each set of preset laser parameter simulation values, a unique set of fitting parameter values ​​can be obtained, thus forming a sample pair between parameter combinations and residual stress characteristics.

[0210] It should be noted that after obtaining the fitting parameters corresponding to each set of residual stress distribution data points, the fitting parameters corresponding to all simulation results can be summarized into a data table. The fields include the preset simulated values ​​of laser parameters, surface residual compressive stress, and compressive stress influence depth. If the goodness of fit is lower than the set threshold, for example, if the coefficient of determination is lower than the predetermined value or there is a systematic deviation in the residual distribution, the sampling interval, smoothing strategy, or fitting function type of the original residual stress distribution data points can be checked back, and the fitting can be re-executed until stable characteristic parameters are obtained.

[0211] S105. Based on the simulated values ​​of each group of preset laser parameters and their corresponding fitted parameter values, perform nonlinear regression processing to obtain the mapping relationship between each fitted parameter and the preset laser parameters, and generate a laser shock strengthening stress mapping model based on the mapping relationship.

[0212] In this application, after obtaining multiple sets of simulated values ​​of preset laser parameters and their corresponding surface residual compressive stress and compressive stress influence depth, nonlinear regression processing can be performed on the sample set to establish a quantitative correspondence between laser parameters and residual stress characteristics. Specifically, nonlinear regression processing is used to establish a mathematical mapping between input and output variables; the input is the simulated values ​​of each set of preset laser parameters, and the output is the fitted parameter values. Since there is a significant nonlinear coupling relationship between process variables such as laser pulse energy, spot diameter, pulse width, and number of impacts and surface residual compressive stress and compressive stress influence depth, linear fitting is insufficient to accurately characterize the stress response caused by parameter changes. Therefore, a nonlinear mapping model is constructed in this step. The mapping relationship can be established separately for surface residual compressive stress and compressive stress influence depth, or a joint regression model can be established that simultaneously outputs two fitted parameters.

[0213] Optionally, one possible implementation of nonlinear regression processing is to normalize or dimensionlessly process the input parameters to eliminate the influence of different dimensions and value ranges on regression stability; then, the sample data is divided into training and validation sets, and an appropriate nonlinear regression form is selected. The nonlinear regression form can be multinomial nonlinear regression, exponential regression, power function regression, response surface regression, radial basis function regression, or a machine learning-based nonlinear regression model.

[0214] For example, if an explicit function expression is used, the function of the surface residual compressive stress with respect to the laser parameters and the function of the compressive stress influence depth with respect to the laser parameters can be constructed, and the function coefficients can be obtained using least squares, regularized least squares or other parameter estimation methods.

[0215] If an implicit learning model is used, a predictor from input to output can be obtained through training, and the model parameters, structure, and prediction rules together form a mapping relationship.

[0216] In this step, during the nonlinear regression process, mean squared error, mean absolute error, and coefficient of determination can be used as evaluation indicators to perform cross-validation on the model. If necessary, the fitting accuracy can be improved by adding interaction terms, adjusting the number of nonlinear basis functions, or resampling the parameter space.

[0217] For example, one possible implementation for generating mapping relationships is as follows:

[0218] f1. Normalize the simulated values ​​of preset laser parameters and their corresponding fitted parameter values ​​for each group to obtain a standardized dataset.

[0219] f2. Perform multivariate power function regression on the fitted parameters in the standardized dataset, combined with the corresponding preset laser parameter simulation values, as shown in Equations 7 and 8:

[0220]

[0221]

[0222] Where I0 is the laser power density (GW / cm²), D is the spot diameter (mm), τ is the pulse width (ns), η is the overlap ratio (%), and N is the laser power density (GW / cm²). s σ represents the number of impacts. s This refers to the residual compressive stress on the surface; z0 refers to the depth of pressure influence. i and b i These refer to the fitting coefficients, where a1 and b1 are proportionality coefficients used to reflect the overall dimensional scaling, which depends on the inherent properties of the material and the combination conditions of the constraint layers; a2~a5 and b2~b5 are elastic coefficients, representing the percentage change of the dependent variable when the independent variable changes by 1%. For example, a2≈0.45 means that for every 1% increase in laser power density, the surface participation pressure increases by about 0.45%.

[0223] f3. For each set of data in the training set of the standardized dataset, calculate the transformed variable, which corresponds to Equation 7: Y1=ln(σ s ), X 1_1 =ln(I0),X 1_2 =ln(D),X 1_3 =ln(η),X 1_4 =ln(N s ), and the corresponding formula 8: Y2=ln(z0), X 2_1 =ln(I0),X 2_2 =ln(D),X 2_3 =ln(τ),X 2_4 =ln(N s ).

[0224] f4. Construct design matrices X1 and X2 and corresponding response vectors Y1 and Y2 respectively, and solve for the coefficient vector β1=[ln(a1),a2,a3,a4,a5] using the least squares method. T And β2=[ln(b1),b2,b3,b4,b5] T That is, β1=(X1) T X1) -1 X1 TY1 and β2=(X2) T X2) -1 X2 T Y2, thus obtaining the fitting coefficients for each group.

[0225] f5. For each fitted parameter, calculate its goodness of fit R. 2 When the goodness of fit is greater than the preset threshold, the mapping relationship is determined to have good prediction accuracy. The fitting coefficients are then substituted into the corresponding multivariate power function to obtain the mapping relationship between the preset laser parameters and the fitting parameters.

[0226] For example, σ s The goodness of fit for z is 0.963, and the goodness of fit for z0 is 0.951, both of which are greater than the preset threshold of 0.95, indicating that the power function form can capture sufficient data patterns.

[0227] f6. The validater in the standardized dataset can also be used to validate the fitted power function, calculate the mean absolute error and mean squared error, verify the generalization ability of the model, and prevent overfitting.

[0228] f7. Package the two sets of fitting coefficients, the range of input preset laser parameter simulation values, and the goodness of fit to construct a laser shock strengthening stress mapping model corresponding to the mapping relationship.

[0229] The laser shock strengthening stress mapping model construction method provided in this application includes: conducting laser shock experiments on a target material based on a preset experimental scheme to obtain experimental data; performing data analysis based on the experimental data to obtain simulation parameters corresponding to the target material; performing multiple laser shock simulations on the target material based on the simulation parameters and preset laser parameters to obtain multiple simulation results; performing fitting calculations on each set of residual stress distribution data points to obtain a set of fitting parameter values; performing nonlinear regression processing based on the simulated values ​​of each set of preset laser parameters and their corresponding fitting parameter values ​​to obtain the mapping relationship between each fitting parameter and the preset laser parameters, and generating a laser shock strengthening stress mapping model based on the mapping relationship. In this application, by combining experimental parameters with simulation modeling, the actual mechanical response of the target material, the characteristics of laser shock load, and the residual stress distribution under different laser parameters are uniformly incorporated into the mapping modeling process, so that the surface residual compressive stress and the depth of influence of compressive stress can be directly predicted by the model. This provides an executable parameter design basis for the laser shock strengthening process of materials for grounding grids, avoids a large number of forward experiments, and thus achieves the technical effect of reducing the cost of materials.

[0230] Figure 3 A flowchart illustrating the laser shock strengthening stress mapping model construction method provided in this application. Figure 2 ,like Figure 3 As shown, the method includes:

[0231] S301. For each type of laser parameter in the preset laser parameters, determine the value range corresponding to the laser parameter.

[0232] In this step, determining the range of laser parameters refers to setting corresponding variation ranges for different types of parameters in the preset laser parameters, and defining upper and lower limits according to their physical meaning. The laser power density in the preset laser parameters is limited by the laser's output capability and the absorption layer damage threshold; too low a density will not generate an effective shock wave, while too high a density may lead to ablation of the material surface. The spot diameter is limited by the aberrations and diffraction limits of the optical focusing system and must match the size of the impact region. The pulse width is determined by the laser's mode-locking characteristics and is typically on the order of nanoseconds. The overlap rate is limited by the mechanical scanning accuracy and spot overlap planning, generally ranging from 0% to 80%. The number of impacts is limited by the laser's processing efficiency and the cumulative thermal / mechanical damage accumulation.

[0233] For example, the value ranges of various parameters determined based on the performance of a certain laser and the stability requirements of the water film confinement are as follows: laser power density (2~8GW / cm²). 2 ), spot diameter (1.5~4.5mm), pulse width (5~25ns), overlap rate (30~70%), number of impacts (1~5).

[0234] S302. For each laser parameter, take values ​​within its corresponding range to obtain multiple sets of simulated laser parameter values.

[0235] In this step, parameters such as laser power density, spot diameter, pulse width, overlap rate, and number of impacts are first defined within their respective ranges based on the feasibility of the process and the output capacity of the equipment. Then, values ​​are taken within each range according to a preset step size or a preset sampling method, forming multiple sets of simulated laser parameter values. Each set of parameters can use single-variable incremental sampling or multi-parameter combination sampling to cover laser impact conditions under different operating conditions. In practical applications, other models of this component can also be selected; this application does not limit this selection.

[0236] Alternatively, sampling can be performed using a central composite method to obtain multiple sets of simulated laser parameter values. One possible implementation of sampling is as follows:

[0237] g1. For each parameter, take 5 levels, which correspond to the minimum value, 1 / 4 quantile, center value, 3 / 4 quantile, and maximum value of the value range. The codes for these 5 levels are (-2, -1, 0, 1, 2).

[0238] For example, the laser power density ranges from 2 to 8 GW / cm². 2The obtained five parameter levels were 2, 3.5, 5, 6.5, and 8 GW / cm². 2 .

[0239] g2. Based on the ±1 level combination of each parameter, factorial points are constructed to obtain 32 sets of simulated laser parameter values.

[0240] g3. Each time, only the ±2 level combination of one parameter is taken, and the center value is taken for the rest to construct the axial point and obtain 10 sets of simulated laser parameter values.

[0241] g4. Set all parameters to 0 level, repeatedly evaluate random errors, construct center points, and obtain 3-5 sets of simulated laser parameter values.

[0242] g5. Based on the simulated laser parameters corresponding to the factorial point, axial point, and center point, 43 to 45 sets of simulated laser parameters were obtained.

[0243] S303. Based on multiple sets of simulated laser parameters and simulation parameters, laser shock simulation is performed on the target material to obtain residual stress distribution data points corresponding to each set of simulated laser parameters, and simulation results corresponding to each set of simulated laser parameters are generated.

[0244] In this step, laser shock simulation refers to inputting each set of simulated laser parameters and the previously determined simulation parameters into the finite element model, performing a complete laser shock numerical simulation, extracting a discrete array of residual profit data points distributed along the depth after the shock ends, and obtaining the simulation results corresponding to each set of simulated laser parameters.

[0245] Alternatively, one possible way to obtain residual stress distribution data points through laser shock simulation is as follows:

[0246] S3031. Using the load-pressure curve as the dynamic load and the remaining parameters in the simulation parameters as material properties, perform explicit dynamic analysis based on the dynamic load and material properties to simulate the propagation of the dynamic load in the target material and the response of the target material, and obtain the final state of the dynamic response.

[0247] The purpose of this step is to simulate the transient physical processes during the shock wave loading phase. During the dynamic analysis, the complete response of the target material is recorded during shock wave compression, plastic flow, stress wave reflection, and transmission. The instantaneous state at the end of the calculation is called the dynamic response final state. The dynamic response final state is a snapshot of the target material's instantaneous state at the end of the dynamic load.

[0248] Alternatively, one possible implementation of explicit dynamic analysis is as follows:

[0249] h1. Apply load to the pre-configured simulation model: The load-pressure curve is used as a uniform surface load and applied to the light spot area on the upper surface of the model. The light spot area refers to a D×D square region, where D refers to the light spot diameter. Each pressure pulse is triggered sequentially based on the preset scanning path.

[0250] In this step, the pre-configured simulation model refers to a pre-built three-dimensional finite element model. For example, the model parameters can be set as follows: model dimensions 5mm × 5mm × 2mm (length × width × thickness), with the laser impact region located at the center of the upper surface. The mesh generation adopts a gradient strategy: a uniform fine mesh is used in the core impact region, with a mesh size Δ... e ≤D / 20, where D refers to the spot diameter (e.g., when D=3mm, Δ e =150μm); the transition zone mesh gradually changes from 150μm to 500μm; the far zone mesh is 500μm. The total number of elements is approximately 300,000 to 500,000, the element type is eight-node reduced integral linear hexahedral elements, and the model is enhanced hourglass control enabled.

[0251] h2. Applying simulation parameters to simulate material response: The relevant parameters of the JC model are embedded into the simulation model. In explicit iteration, each incremental step needs to be implemented according to the equation corresponding to the JC model, as shown in Equation 1 above. The equation yields the flow stress, and the adiabatic temperature rise also needs to be considered. The calculation method for the adiabatic temperature rise is shown in Equation 9.

[0252]

[0253] Where σ refers to flow stress; η T This refers to the plastic work conversion rate, which can be as low as 0.99, meaning that 90-99% of the plastic work is converted into heat; ρ refers to the density of the target material, for example, if the target material is Q25 steel, its density is 7850 kg / m³. 3 ;c p ε refers to the specific heat of the target material. For example, if the target material is Q25 steel, its specific heat can be taken as 470 J / (kg·K); p It refers to plastic strain; ΔT refers to adiabatic temperature rise.

[0254] h3. Determine the time step for explicit analysis: The time step for explicit analysis is automatically constrained by the Courant stability condition, i.e., Δt ≤ ΔL / c d Where △L is the minimum grid size, c d=√(E(1-v) / ρ(1+v)(1-2v)) refers to the propagation speed of the shock wave in the target material (e.g., the wave speed of Q325 steel is 5100m / s). Here, E refers to the elastic modulus, characterizing the stiffness of the target material against tensile / compressive elastic deformation. The larger the value, the "harder" the material, and the smaller the elastic strain produced under the same stress. For example, the standard value of the elastic modulus of Q235 steel is 210GPa; v refers to Poisson's ratio, characterizing the ratio of transverse strain to longitudinal strain when the target material is subjected to axial tension or compression, which can be taken as 0.3 here. This condition ensures that the stress wave will not cross a unit within one time step, typically with a step size in the nanosecond range (ranging from 1 to 5ns). For a single impact, the total simulation time can be set to 500ns (approximately t). end (2 to 3 times that of the original text) to ensure complete attenuation of the shock wave; where t end This refers to the total duration of the shock wave pressure time history curve.

[0255] h4. Set the boundaries and mesh for explicit analysis: The bottom surface of the simulation model is fixed with constraints, and infinite element boundaries are set on the four sides to absorb outward-propagating stress waves and prevent false reflections. The core impact zone uses a fine mesh (diameter can be set to 150μm), gradually tapering outwards to 500μm. The element type is eight-node reduced integral, and hourglass control is enabled.

[0256] For example, under the condition of a spot diameter of 3 mm and a pulse width of τ = 15 ns, the load pressure curve (peak value 4.18 GPa, rise time 16 ns) is applied to a 3 × 3 mm area at the center of the surface. Explicit analysis automatically uses a step size Δt ≈ 2.5 ns, with a total of 200 steps, taking approximately 1.5 hours. After 500 ns of calculation, a snapshot of the displacement field, stress field, and temperature field at that moment is output to obtain the final state of the dynamic response.

[0257] S3032. Using the final state of the dynamic response as the initial condition, perform implicit static analysis on the target material after removing the dynamic load, solve the static equilibrium residual stress field, and obtain the residual stress distribution data points.

[0258] The implicit static analysis in this step is used to simulate the stress redistribution and static equilibrium process after unloading. Specifically, the implicit static analysis uses the Newton-Raphson iterative algorithm to solve the static equilibrium equation [K]{u}={F}. The final state of the dynamic response from the explicit dynamic analysis is imported as the initial state, and the dynamic oscillation components are iteratively eliminated until the model reaches a stable convergent state that satisfies the balance between internal and external forces. This finally converged stress field is the stable residual stress field in an engineering sense.

[0259] Alternatively, one possible implementation of implicit static analysis is as follows:

[0260] i1. Transfer the data from the explicit dynamic analysis: the nodal displacements, stress tensors, and equivalent plastic strains corresponding to the final state of the dynamic response are mapped to the initial conditions of the implicit step.

[0261] i2. Remove all dynamic color boxes in the current simulation model environment, retain only the boundary constraints, solve the static equilibrium residual stress field, and obtain the residual stress distribution data points.

[0262] For example, continuing the example in step S3031 above, the implicit analysis step uses the explicit final state as the initial field, and converges after about 15 iterations (taking about 0.5 to 1 hour). Stress data along the central axis of the model are extracted, revealing that σ = -420 MPa (compressive stress) at the surface (z=0), and decreases exponentially with increasing depth, reaching zero at z≈0.25 mm. σ under this set of parameters is obtained through exponential fitting. s =420MPa, z0=0.22mm. These two values, along with the input preset laser parameter simulation values ​​(5,3,15,50,2), are stored in the database as one of the data points for the nonlinear regression in step S105. Finally, after multiple sets of preset laser parameter simulation values ​​are processed in this way, a complete dataset is formed for fitting the mapping relationship between laser parameters and fitted parameters.

[0263] It should be noted that in practical implementation, explicit dynamic analysis and implicit static analysis can be executed in series within the same simulation platform. The displacement, velocity, and stress fields output in the explicit analysis stage are directly transferred to the implicit analysis stage to ensure the continuity of the state before and after unloading. The input length, peak pressure, and rise time of the load-pressure curve can be matched with the corresponding simulated laser parameters, thus ensuring that each set of simulated laser parameters yields a unique residual stress distribution data point. This method combines the transient propagation calculation of the impact process with the equilibrium solution after unloading, enabling the dynamic action results to be stably mapped into discrete residual stress data that can be used for subsequent fitting, and improving the consistency between the simulation results and the actual laser impact response.

[0264] Figure 4 Flowchart of the laser shock osmosis corrosion protection method provided in this application Figure 1 ,like Figure 4 As shown, the method includes:

[0265] S401. Obtain the corrosion protection requirements for the target material, and generate the target residual stress field index corresponding to the target material based on the corrosion protection requirements.

[0266] In this step, the target residual stress field index corresponding to the target material can be generated by combining the corrosion conditions and structural characteristics of the target material in its service environment to determine the allowable corrosion rate, crack-sensitive areas, and key reinforcement areas that need to be covered. The aforementioned corrosion protection requirements are then converted into calculable residual stress field constraint parameters. The target residual stress field index includes at least the surface residual compressive stress value and the compressive stress influence depth. The surface residual compressive stress value characterizes the material surface's ability to resist the intrusion of corrosive media and the propagation of microcracks, while the compressive stress influence depth characterizes the effective range of compressive stress along the thickness direction. The target residual stress field index can be determined jointly by design service requirements, component dimensions, local structural conditions, and environmental corrosion levels. It can also be converted using empirical databases, process specifications, or pre-established protection thresholds, transforming the corrosion protection requirements into quantitative boundaries required for subsequent optimization, thereby ensuring a correspondence between the target residual stress field index and the actual protection target.

[0267] S402. Construct an inverse optimization problem based on a pre-built laser shock strengthening stress mapping model and target residual stress field index.

[0268] In this step, the objective function of the inverse optimization problem is the laser processing energy consumption function, and the constraint condition is that at least one of the fitting parameters of the laser shock strengthening stress mapping model satisfies the target residual stress field index.

[0269] Optionally, when constructing the inverse optimization problem, the laser shock strengthening stress mapping model is used to characterize the mapping relationship between preset laser parameters and residual stress response. The fitting parameters may include the surface residual compressive stress value, the compressive stress influence depth, or a response vector composed of both. The model input corresponds to laser parameters such as laser power density, spot diameter, pulse width, overlap rate, and number of shocks. After introducing the target residual stress field index into the constraint conditions, the original forward parameter matching can be transformed into an inverse solution process guided by the target residual stress field index. The laser processing energy consumption function can be composed of a weighted average of single pulse energy, total number of pulses, scanning path length, and processing power per unit area, used to characterize the total energy input required to complete the target strengthening. By setting the laser processing energy consumption function as the optimization objective and requiring at least one fitting parameter in the model output to reach the target residual stress field index, it is possible to perform calculable screening of laser parameter combinations while meeting the corrosion protection strengthening requirements. This allows the solution result to be simultaneously constrained by the target residual stress field index and the laser processing energy consumption function, thus forming an inverse optimization problem with a clear engineering boundary. Since the laser shock strengthening stress mapping model has solidified the residual stress response relationship obtained from experiments and simulations into a callable correlation, the optimization process no longer relies on round-by-round experimental adjustments, but directly uses the laser shock strengthening stress mapping model for solution and convergence.

[0270] Optionally, the constraints of the inverse optimization problem also include the range of values ​​for the laser parameters; when solving the inverse optimization problem, the range of values ​​is used as the upper and lower boundary constraints of the laser parameters.

[0271] In this step, introducing this boundary constraint when solving the inverse optimization problem ensures that the objective function minimization process always searches along the manufacturable and processable parameter space. Since the relationship between laser parameters and residual stress response is typically nonlinear, without boundary constraints, the optimization iteration may produce results beyond the applicable range of the laser shock peening stress mapping model due to local extrema or numerical extrapolation, leading to distortion of the obtained optimal laser parameter values ​​in actual execution. By using the value range as upper and lower boundary constraints, the optimization algorithm updates variables only within the effective interval during iteration. This maintains the consistency between the laser shock peening stress mapping model and actual process conditions, avoids interference from invalid solutions in determining the optimal solution, and thus improves the feasibility and stability of the inverse optimization results.

[0272] For example, the energy consumption function for laser processing is shown in Equation 10:

[0273]

[0274] Here, E refers to the total laser output energy, which is the laser processing energy consumption corresponding to the optimized target. pulse For single-pulse energy; Ntotal L is the total number of pulses. scan and W scan Here, A represents the length and width of the rectangular region to be processed, i.e., the length and width of the scanning area. D refers to the spot diameter, and A... spot This refers to the area of ​​the laser spot, calculated from the spot diameter; η refers to the overlap ratio, N s This refers to the number of impacts, τ refers to the pulse width, and I0 refers to the laser power density. This can be simplified to Formula 11:

[0275]

[0276] For an explanation of the parameters, please refer to Formula 10 above.

[0277] The corresponding constraints are shown in Equations 12 to 15:

[0278]

[0279]

[0280]

[0281]

[0282] Where, σ s This refers to the residual compressive stress on the surface; z0 refers to the depth of pressure influence, as described above. Figure 1 In the illustrated embodiment, the explicit power function formulas 7 and 8, calibrated in step S105, are used for calculation, eliminating the need to call the finite element method. σ s_target , z 0_target This refers to the target residual stress field index obtained in step S401. For explanations of the remaining parameters, refer to Formulas 1-10 above. Formulas 12 and 13 refer to stress constraints, while Formulas 14 and 15 refer to the upper limit boundary constraints of the process parameters themselves, which are determined by equipment capacity and physical limits.

[0283] S403. Solve the problem based on the inverse optimization problem to obtain a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function.

[0284] In this step, the inverse optimization problem can be solved iteratively using numerical optimization methods, optimizing variables such as laser pulse energy, spot diameter, pulse width, overlap rate, and pulse count. During the solution process, the laser shock peening stress mapping model is first invoked based on the current candidate parameters to obtain the corresponding residual stress response. Then, it is determined whether this residual stress response meets the target residual stress field index. Only parameter combinations that meet the constraints are included in the laser processing energy consumption function comparison set, and the laser processing energy consumption function value is used as the sorting criterion. Since the laser processing energy consumption function is directly related to the total laser action, the optimization results will favor reducing repeated shocks and excessive energy input while achieving the target residual stress field index, thus obtaining a parameter set with more concentrated energy utilization. This solution method ensures that the output parameters not only correspond to the achievable target residual stress field index but also have a lower processing burden and stronger parameter transferability in engineering implementation, adapting to the process design requirements of different component specifications and different corrosion environments.

[0285] Alternatively, one possible approach to solving the inverse optimization problem to obtain the optimal laser parameter values ​​is as follows:

[0286] S4031. Based on the range of laser parameters, generate multiple initial iteration points.

[0287] In this step, the initial iteration points are generated by using uniform sampling, random sampling, or a multi-starting-point strategy within the range of laser parameter values. The purpose of generating multiple initial iteration points is to avoid getting trapped in local optima due to the influence of initial values ​​in a single solution.

[0288] For example, within the bounding boxes corresponding to Equations 14 and 15, 20 initial points are generated using Latin hypercube sampling or pure random sampling. Each initial point serves as the starting vector for the inverse solution.

[0289] S4032. For each initial iteration point, perform sequential quadratic programming iterative solution based on the inverse optimization problem to obtain a set of locally optimal laser parameter values ​​corresponding to each initial iteration point.

[0290] In this step, at each iteration, the Lagrangian function of the original problem is expanded using a second-order Taylor expansion at the initial iteration point corresponding to the current iteration, and the nonlinear inequality constraints are expanded using a first-order Taylor expansion. This transforms the difficult-to-solve original problem into a standard quadratic programming subproblem. Then, this quadratic programming subproblem is solved to obtain a set of locally optimal laser parameter values ​​corresponding to the initial iteration point.

[0291] It should be noted that the quadratic programming iterative solution in this step is as follows: Figure 5 Further explanation will be provided in the embodiments shown, and will not be repeated here.

[0292] S4033. Compare the objective function values ​​corresponding to each group of local optimal laser parameter values, and determine the group of local optimal laser parameter values ​​with the lowest objective function value as the group of optimal laser parameter values ​​with the smallest energy consumption function value for non-laser processing.

[0293] In this step, each initial iteration point corresponds to a set of locally optimal laser parameter values. Substitute each set of locally optimal laser parameter values ​​into the laser processing energy consumption function corresponding to Formula 10 above, and determine the set of locally optimal laser parameter values ​​with the lowest energy consumption value as the final optimal laser parameter values.

[0294] S404. Perform laser shock strengthening treatment on the target material according to the optimal laser parameters.

[0295] In this step, the laser shock peening treatment can be implemented by: writing the obtained optimal laser parameters into the laser shock equipment control program, and performing zone-by-zone shock peening on the target material surface according to a predetermined scanning trajectory. This causes the laser pulse to form a high strain rate shock wave on the material surface, thereby introducing a compressive stress distribution corresponding to the target residual stress field index within the target area. Since the aforementioned parameters have been constrained by the laser shock peening stress mapping model and screened by minimizing the laser processing energy consumption function, during actual processing, it is possible to ensure that the target material surface obtains a surface residual compressive stress value and compressive stress influence depth that meet the anti-corrosion requirements, while reducing unnecessary energy input and repeated experimental corrections, so that the treated component maintains a relatively stable surface stress state.

[0296] Optionally, laser shock peening also requires processing path planning, which can be implemented as follows:

[0297] For flat steel requiring processing on all four sides, each side is laser-scanned using a zigzag reciprocating scanning path. The center-to-center distance between adjacent laser spots can be set as: Δx = D·(1-η). For example, when D = 3mm and η = 50%, Δx = 1.5mm. The scanning speed can be set as: v scan =f rep ⋅△x, where f rep Let f be the laser repetition frequency (Hz). rep =10Hz, then v scan =15mm / s. When the number of scans N... s >1 means that after completing a full-coverage scan, the absorption layer must be re-attached before the next scan can be performed to ensure that the constraint effect of each impact is consistent.

[0298] Optionally, process monitoring and safety monitoring are also required during the processing. Specific implementation methods include:

[0299] Energy stability: Monitored by a real-time energy meter, the pulse energy fluctuation is required to be <±3%; if it exceeds this range, an alarm will be triggered and the line will be shut down.

[0300] Position accuracy: The position deviation of the online monitoring spot must be controlled within ±0.1mm.

[0301] Water film and absorber layer: The flow meter monitors the continuity of the water film; the absorber layer is visually inspected after each impact, and any damage is immediately repaired.

[0302] Safety precautions: The operating area is equipped with a 1064nm laser protective barrier, and operators must wear special protective glasses with an OD ≥ 7; a laser safety interlock device is provided, which immediately cuts off the laser when the door is opened. Wastewater and aluminum foil residue are collected and filtered according to hazardous waste treatment procedures.

[0303] Optionally, the processed target material also needs to undergo quality testing to determine whether it meets the acceptance criteria. The specific implementation method can be as follows:

[0304] Surface residual stress detection: using sin 2 The ψ method was used, with an X-ray source of CrKα and a wavelength of 2.29 Å. The diffraction plane was Fe(211), and the 2θ angle of the diffraction line was approximately 156°. At least nine tilt angles ranging from 0° to 45° were measured.

[0305] Measurement Points and Acceptance: At least 5 points should be measured on each side, including the four corners and the center. The acceptance criterion is: the absolute value of the measured residual compressive stress on the surface |σ s |≥0.9×σ s_target That is, reaching more than 90% of the target value is considered qualified.

[0306] Layer-by-layer stress distribution verification:

[0307] For representative specimens, electropolishing was used to remove the layers one by one (electrolyte: HClO4:C2H5OH=1:9, voltage 30V), with each layer being approximately 10μm thick.

[0308] After each layer is removed, the stress at that depth is measured immediately by surface residual stress detection, and a "depth-stress" curve is plotted. This curve is then compared with the exponential decay prediction curve of the mapping model or formula (5) to verify the actual achievement of z0.

[0309] Surface quality inspection:

[0310] Use a stereomicroscope (20×) to inspect the surface to ensure there are no ablation pits, microcracks, or carbonized residues in the absorption layer.

[0311] Roughness inspection: The surface roughness Ra increment after laser shock treatment is required to be <0.2μm.

[0312] Spot check of corrosion resistance:

[0313] Randomly select ≥10% of batches for accelerated corrosion testing (e.g., immersion in simulated soil solution for 14 days). Acceptance criteria: The corrosion rate of the treated sample is reduced by ≥70% compared to the untreated base material for the batch to be deemed to meet the corrosion protection design life requirements.

[0314] The laser shock peening corrosion protection method provided in this application includes generating target residual stress field indices based on the corrosion protection requirements of the target material, constructing and solving an inverse optimization problem based on a laser shock peening stress mapping model, obtaining optimal laser parameters, and implementing laser shock peening treatment accordingly. By transforming corrosion protection requirements into constrained target residual stress field indices and using a laser shock peening stress mapping model and a laser processing energy consumption function for joint optimization, the laser parameter design process can be transformed from empirical experimentation into a quantitative solution process, thereby improving the repeatability of process parameter determination and the degree of matching with target protection requirements.

[0315] Figure 5 Flowchart of the laser shock osmosis corrosion protection method provided in this application Figure 2 ,like Figure 5 As shown, the method includes:

[0316] S501. In each iteration, the objective function is approximated twice at the current iteration point, and the constraints are approximated linearly at the current iteration point to obtain the quadratic programming subproblem corresponding to the iteration.

[0317] In this step, after setting an initial iteration point, a local quadratic approximation of the objective function is constructed based on the function values, first derivatives, and second derivatives of the current laser parameters. Simultaneously, the nonlinear constraints are transformed into linear constraints through a first-order Taylor expansion at the current iteration point, thus forming a quadratic programming subproblem corresponding to the current iteration point. This process transforms the originally difficult-to-solve inverse optimization problem into a locally solvable standard optimization model, enabling the search process to simultaneously consider the decreasing trend of the laser processing energy consumption function and the satisfaction of the target residual stress field index.

[0318] Alternatively, assume the current iteration point is x. k The resulting quadratic programming subproblem is shown in Formula 16:

[0319]

[0320] The corresponding constraints are shown in Formula 17:

[0321]

[0322] Where, x kThis refers to the current laser parameter combination vector at the k-th iteration; d refers to the search direction vector to be solved, indicating how the parameters should change starting from the current point. k ) refers to the original objective function, that is, the laser processing energy consumption function E corresponding to Formula 10 above, at the current point. ▽f(x k H refers to the first-order gradient vector of the objective function at the current point, which determines the direction of the fastest descent of the objective function. k This refers to the Hessian matrix of the Lagrange function, a 5×5 symmetric matrix that contains the objective function and the second-order curvature information of the constraints at the current point. i (x k () refers to the value of the i-th nonlinear inequality constraint function at the current point, for example, constraint g1=σ s_target -σ s (I0,D,η,N s )≤0. ▽g i (x k ) refers to the first-order gradient vector of the i-th constraint function at the current point, σ s_target This refers to the target residual stress field index obtained in step S401.

[0323] S502. Solve the quadratic programming subproblem to obtain the search direction.

[0324] In this step, after solving the quadratic programming subproblem, the resulting search direction reflects which direction the current parameter point changes in to more quickly reduce the objective function while maintaining constraint feasibility. A line search is then performed along this direction to determine the step size that satisfies the descent condition without excessively crossing the feasible region boundary, and the current iteration point is updated accordingly. The line search can employ backtracking line search, exact line search, or a step size determination method that satisfies preset descent conditions. Backtracking line search avoids the objective function from rising by successively reducing the step size; exact line search directly determines the step size in the search direction that leads to a local optimum for the objective function; and methods that satisfy preset descent conditions ensure that each update brings sufficient improvement to the objective function, thereby improving iteration stability.

[0325] In this step, the solution method is to input the constructed quadratic programming subproblem into the pre-called quadratic programming solver for calculation, and obtain an optimal increment vector, i.e., the search direction. The significance of this direction is that by moving the parameters along this direction, the objective function will achieve the maximum decrease while satisfying the current linearization constraints.

[0326] S503. Perform a line search based on the search direction to determine the step size and update the current iteration point.

[0327] In this step, after obtaining the direction, it is also necessary to determine how far to travel along that direction, i.e., to determine the step size. The step size determines the actual new parameter point. Let's assume the step size is α. k Then the update of the new parameter point is x. k+1 =x k +α k d k ∗ Among them, d k ∗ This refers to the search direction; it's not a single numerical value but a five-dimensional vector used to guide the adjustment of the five laser parameters in the next round. The step size ranges from 0 to 1. The step size can be determined by first taking the maximum value of 1, calculating the objective function value at the new parameter point, and checking if the sufficient descent condition is met. If the condition is met, the step size is accepted; otherwise, it's multiplied by a shrinkage factor to obtain a new step size. This process is repeated until a step size that satisfies the sufficient descent condition is found. The sufficient descent condition is shown in Formula 18.

[0328]

[0329] Here, c1 is a very small constant used to discount the "theoretical reduction" as an acceptance threshold; f is the objective function, which is the laser processing energy consumption function E corresponding to Formula 10 above. k This refers to the current laser parameter combination vector at the k-th iteration; d k ∗ This refers to the search direction; α k This refers to the step size.

[0330] S504. When the current iteration point meets the preset convergence condition, the current iteration point is determined as a set of locally optimal laser parameter values.

[0331] In this step, when the updated current iteration point meets the preset convergence condition, it usually indicates that the change in the objective function, the parameter increment, or the constraint violation has fallen below a set threshold. At this point, the iteration point is determined as a set of locally optimal laser parameter values. If the convergence condition is not met, the quadratic programming subproblem is repeatedly constructed and solved and updated based on the new current iteration point until convergence. Through this iterative mechanism combining "local quadratic approximation, direction search, and step size control," the laser parameters can converge to different locally optimal solutions at multiple initial points, thus providing reliable candidate results for subsequently selecting a set of optimal laser parameter values ​​that have the lowest laser processing energy consumption function and meet the target residual stress field index.

[0332] Optionally, the preset convergence conditions include: the change in the objective function value between two adjacent iterations is less than a first preset threshold; and / or, the constraint violation at the current iteration point is less than a second preset threshold.

[0333] In this step, after making a quadratic approximation of the objective function at the current iteration point and a linear approximation of the constraints, the search direction obtained from the solution will drive the current iteration point to move towards a better solution. When the change in the objective function value between two adjacent iterations is less than the first preset threshold, it indicates that the laser processing energy consumption function has basically stopped decreasing significantly, and the current solution is close to a local stable point. At the same time, when the constraint violation at the current iteration point is less than the second preset threshold, it indicates that the laser parameter combination corresponding to that point has basically satisfied the constraint relationship corresponding to the target residual stress field index, thus enabling the obtained solution to simultaneously possess target stability and feasibility.

[0334] S505. If the current iteration point does not meet the preset convergence condition, proceed to the next iteration round until the current iteration point meets the preset convergence condition.

[0335] In this step, by setting the first and second preset thresholds to values ​​that match the target residual stress field index, the range of laser parameter values, and the accuracy of solving the inverse optimization problem, the convergence determination can avoid premature termination that would cause the solution to deviate from the constraint requirements, as well as excessive iteration that would increase computational overhead. This convergence determination method can work synergistically with the local optimization process of sequential quadratic programming, ensuring that the current iteration point is determined as the locally optimal laser parameter value when the objective function stabilizes and the constraint deviation is controlled. This improves the solution stability and parameter matching accuracy of the laser shock strengthening stress mapping model when used for process inverse optimization.

[0336] Figure 6 The flowchart of the laser shock osmosis corrosion protection method for Q235 steel provided in this application is as follows. Figure 1 This embodiment explains the selection of materials, laser system, and determination of preset laser parameters for laser shock irradiation. Figure 6 As shown, the method includes:

[0337] A1. Select the raw materials for the grounding grid.

[0338] In this step, the raw materials can be Q235B or Q235C grade hot-rolled flat steel (-60×6mm, -50×5mm) or round steel (Φ12mm, Φ16mm). The chemical composition requirements for the materials are: C content 0.14~0.22%, Si content ≤0.35%, Mn content 0.30~0.65%, P content ≤0.045%, and S content ≤0.045%.

[0339] A2. Pre-treatment of raw materials.

[0340] In this step, the pretreatment can be performed by sequentially grinding or sandblasting with 80# → 400# → 800# sandpaper (Al2O3 abrasive, 60#~80#, 0.3~0.5MPa) to control the surface roughness to 0.8~1.6μm. Then, the surface is ultrasonically cleaned in acetone and anhydrous ethanol for 15 minutes each (frequency 40kHz), and dried with compressed air.

[0341] A3. Prepare an absorbent layer for the pretreated raw materials.

[0342] The specific method for this step is to either apply aluminum foil tape (100±10μm thick) or spray black acrylic paint (80~120μm thick, air-drying for 2 hours) to the clean and dry workpiece surface. The function of the absorption layer is to absorb laser energy and protect the workpiece surface from direct laser ablation. Aluminum foil tape is suitable for flat surfaces, while black paint is suitable for complex curved surfaces. The absorption layer must be evenly applied, without bubbles or wrinkles.

[0343] A4. Prepare a constraint layer for the pretreated raw materials.

[0344] The specific method for this step is as follows: a flowing water film is used as a constraint layer, with a thickness of 1–3 mm. A uniform and stable water film layer is formed on the workpiece surface using an adjustable flow rate water spray device. The function of the water film is to restrict the direction of plasma expansion, allowing the shock wave energy to concentrate and propagate into the material. The water film flow rate is controlled at 1–3 L / min to ensure continuous and uninterrupted water film coverage.

[0345] A5. Selection of LSP laser system for the experiment.

[0346] The LSP laser system selected in this step can be an Nd:YAG solid-state pulsed laser or a fiber laser. The core technical specifications of the laser system include: wavelength of 1064 nm (fundamental frequency) or 532 nm (second harmonic), adjustable single-pulse energy of 1–10 J, pulse width of 5–25 ns (FWHM), repetition frequency of 1–10 Hz, and beam quality M² < 3. Spot shaping employs diffractive optical elements (DOE) or a lens focusing system to obtain a uniform square or circular spot with a diameter of 1–5 mm. Spot energy uniformity is >85% (flat-top distribution).

[0347] A6. Determine the motion control system corresponding to the LSP laser system.

[0348] The motion control system in this step uses a five-axis CNC worktable or a six-DOF robot to mount the workpiece, and controls the scanning trajectory of the laser spot on the workpiece surface through a CNC program. The scanning path adopts a "Z" shape or a spiral shape, and the overlap rate η is controlled by the spot spacing Δx = D(1-η), where D refers to the spot diameter and Δx refers to the distance between two spots. The positioning accuracy can be set to ±0.05mm, and the repeatability can be set to ±0.02mm.

[0349] A7. Determine the process parameter system corresponding to the LSP laser system, i.e., the preset laser parameters used in this application.

[0350] The process parameters in this step include: laser power density I0 (GW / cm²), spot diameter D (mm), pulse width τ (ns), overlap rate η (%), and number of impacts N. s (Times). This also includes fixed laser parameters, such as: wavelength 1064nm, absorption layer aluminum foil 100μm, constraint layer water film 2mm, and square spot shape. The laser power density can be calculated using the laser's single-pulse energy and the corresponding spot diameter and pulse width, calculated as: I0=E pulse / (A spot ·τ).

[0351] For example, when the single pulse energy E pulse When the light intensity is 3J, the light spot is a square spot with a diameter D = 3mm and a spot area A. spot Given a pulse width τ of 0.09 cm² and a pulse width τ of 15 ns, the calculated laser power density I0 is 2.22 GW / cm². 2 .

[0352] Figure 7 The flowchart of the laser shock osmosis corrosion protection method for Q235 steel provided in this application is as follows. Figure 2 In this embodiment, the above Figure 6 Based on the illustrated embodiments, explanations are provided through physical experiments, data simulations, and reverse optimization, such as... Figure 7 As shown, the method includes:

[0353] B1. Calibrate the JC parameters through physical experiments.

[0354] The first step in this process is to prepare the test material: Q235B hot-rolled flat steel (-60×6mm), from the same batch. Chemical composition: C 0.16%, Si 0.22%, Mn 0.48%, P 0.032%, S 0.028%.

[0355] Five sets of quasi-static tensile tests were conducted to obtain the quasi-static yield strength, strain hardening coefficient, and hardening index from the simulation parameters. The experimental equipment was an electronic universal testing machine with a strain rate of 1×10⁻³ s⁻¹ and an ambient temperature maintained at a constant 298 K. The simulation parameters obtained by fitting the experimental data are: quasi-static yield strength... strain hardening coefficient Hardening index and the goodness of fit .

[0356] The strain rate sensitivity coefficient in the simulation parameters was obtained through 12 sets of SHPB dynamic compression tests. The testing equipment was an ALT1000 SHPB, using Φ14.5mm martensitic steel rods. The strain rate levels were divided into 1000 / 5000 / 10000 / 50000. ×3 groups; the high-speed data acquisition rate was set to 2MHz, and Wheatstone full-bridge strain gauges were used. Stress balance was verified using the three-wave method based on the experimentally acquired data. The strain rate sensitivity coefficient obtained from the fitting was... The corresponding goodness of fit .

[0357] The thermal softening index in the simulation parameters was obtained through 18 sets of high-temperature tensile tests. The testing equipment was an MTS+ high-temperature furnace, and three sets of tests were conducted at 100 / 200 / 300 / 400 / 500 / 600℃ respectively. The heating rate was set to 10℃ / min, the holding time was 15min, and the strain rate was 1×10⁻⁶. -3 s -1 The thermal softening index obtained by fitting The corresponding goodness of fit .

[0358] Based on the calculated five parameters, a Taylor bar impact test was simulated, and the deviation between the simulated and measured length values ​​was less than 4%, confirming the accuracy of the modified parameters. Thus, the JC parameter was obtained: A jc =235, B jc =520, n jc =0.48, C jc =0.042, m jc =0.85.

[0359] B2. By calibrating the Fabbro model through laser shock testing, the load-pressure curve corresponding to Q235 steel is obtained.

[0360] In this step, a PVDF shock wave pressure measurement test is first performed. The measurement conditions are as follows: a 25μm thick PVDF piezoelectric film sensor is used, with a time resolution of <5ns. The sensor is embedded 0.5mm below the surface of the Q235 steel specimen (via a precision groove and epoxy adhesive). The signal is amplified and then acquired by an oscilloscope (2.5GHz bandwidth, 40GS / s sampling rate). The measurement method is as follows: Measured three times at each of the seven levels (GW / cm²). Compare with the above. Figure 1 The calculated value of Formula 2 in the illustrated embodiment is compared with the measured value of PVDF to inversely deduce the energy conversion efficiency. .result: (coefficient of variation) ),exist The approximate constant value within the parameter range verifies the applicability of the Fabbro model within this parameter range. The calculated energy conversion efficiency is combined with Equation 2 above to construct the Fabbro model. Then, a set of data is used for verification under the conditions of I0 = 5 GW / cm² and τ = 15 ns. Under these conditions, the peak pressure measured by the PVDF is 2.65 GPa, while the calculated value from Equation 2 is 2.74 GPa, resulting in a calculation error of 3.4%. The pressure rise time and attenuation pattern are in good agreement with Equation 3.

[0361] B3. Based on the JC parameters and load-pressure curves, construct and verify a simulation model to obtain a suitable simulation model.

[0362] In this step, a simulation model is constructed based on the JC parameters and the load-pressure curve; a set of specific laser parameters is input: , , , , The corresponding constraint is a water film constraint, with a 100μm aluminum foil absorption layer. The simulation model has approximately 380,000 elements. The calculation time was 2.8 hours with an explicit step of 500ns and an implicit step of 10 iterations in 0.3 hours. The obtained fitted parameter values ​​are: surface residual compressive stress. Depth of the impact of pressure Peak position control parameters decay rate control parameters goodness of fit Stress was measured layer by layer by X-ray diffraction: 12 layers were electropolished (approximately 10 μm per layer). CrKα, Fe(211) plane, 9 ψ angles. Results are as follows: Figure 8 As shown: Surface measurement (Model 385MPa, error 6.0%), affecting actual depth measurement (Model depth 78μm, error 4.9%). The overall MAPE of the stress values ​​at each depth point and the model curve is 7.2%, verifying the high accuracy of the simulation model.

[0363] B4. Simulation tests were conducted on Q235 steel using a simulation model, resulting in 45 sets of simulation data. The simulation data were then fitted to obtain the mapping relationship between preset laser parameters, pressure influence depth, and surface participation pressure.

[0364] B5. Based on the mapping relationship, perform inverse optimization of the corrosive environment using Sequential Quadratic Programming (SQP) to obtain the first preset combination of laser parameter values ​​for processing.

[0365] Specifically, this step involves determining the target compressive stress influence depth z based on the medium corrosion environment. 0_target ≥50μm and residual compressive stress σ on the target surface s_target ≥300MPa. Set 20 random initial points corresponding to the laser parameters. Constraint range: , , , , A single SQP iteration takes approximately 12 iterations and converges in 0.3 seconds. The optimal objective function value corresponding to 20 runs is: , , , , Then, based on empirical methods, a set of laser parameter solutions is determined as follows: , , , , Next. For example. Figure 9 The results show that the surface residual compressive stress corresponding to the empirical scheme is 385 MPa, while the surface residual compressive stress corresponding to the SQP scheme is 312 MPa. The compressive stress influence depth corresponding to the empirical scheme is 77 μm, while that of the SQP scheme is 55 μm. Furthermore, the processing energy consumption of the two schemes is as follows: the laser processing energy consumption of the SQP scheme is 72 J, while that of the empirical scheme is 100 J. It should be noted that the processing energy consumption of the two schemes is not significantly different. Figure 9 As shown in the figure. Therefore, determining the SQP scheme based on the depth of compressive stress influence and residual compressive stress on the surface yields higher accuracy. Furthermore, through a 14-day accelerated corrosion experiment, the measured corrosion rate was... Compared to the corrosion rate of untreated steel It decreased by 79.7%.

[0366] B6. Based on the mapping relationship, perform SQP inverse optimization in a highly corrosive environment to obtain a second preset combination of laser parameter values ​​for processing.

[0367] In this step, the target compressive stress influence depth z is determined based on the highly corrosive environment. 0_target ≥80μm and residual compressive stress σ on the target surface s_target ≥400MPa. Set 20 random initial points corresponding to the laser parameters. Constraint range: , , , , A single SQP iteration takes approximately 12 iterations and converges in 0.3 seconds. The optimal objective function value corresponding to 20 runs is: , , , , The SQP method yielded a surface residual compressive stress of 428 MPa and a compressive stress influence depth of 95 μm, meeting the target requirements. Furthermore, through a 14-day accelerated corrosion experiment, the measured corrosion rate was... Compared to the corrosion rate of untreated steel It decreased by 85.2%.

[0368] B7 and Q235B hot-rolled flat steel were replaced with 16mm round steel of the same material. SQP reverse optimization was performed on the material to obtain the target laser parameter value combination for the material.

[0369] In this step, the material is Φ16mm-Q235B round steel. The curved surface of the round steel causes uneven light spots, so a curvature correction coefficient is used. Correct the equivalent spot area, and then re-optimize SQP based on the corrected equivalent spot area: , , , , Actual measurement , And the corrosion rate over 14 days. Compared to the corrosion rate of untreated steel It decreased by 81.4%.

[0370] Figure 10 A schematic diagram of the laser shock strengthening stress mapping model construction device provided in this application is shown below. Figure 10 As shown, the laser shock strengthening stress mapping model construction device provided in this embodiment includes:

[0371] The acquisition module 1001 is used to conduct laser shock experiments on target materials based on a preset experimental scheme and obtain experimental data.

[0372] The first processing module 1002 is used to perform data analysis based on experimental data to obtain the simulation parameters corresponding to the target material;

[0373] The second processing module 1003 is used to perform multiple laser shock simulations on the target material based on simulation parameters and preset laser parameters to obtain multiple simulation results; wherein, each simulation result includes a set of preset laser parameter simulation values ​​and a set of residual stress distribution data points;

[0374] The third processing module 1004 is used to perform fitting calculations on each set of residual stress distribution data points to obtain a set of fitting parameter values; the fitting parameters include surface residual compressive stress and compressive stress influence depth.

[0375] The fourth processing module 1005 is used to perform nonlinear regression processing based on the simulated values ​​of each set of preset laser parameters and their corresponding fitted parameter values, to obtain the mapping relationship between each fitted parameter and the preset laser parameters, and to generate a laser shock strengthening stress mapping model based on the mapping relationship.

[0376] Alternatively, in one possible implementation, the acquisition module 1001 is further configured to:

[0377] Quasi-static tensile tests were conducted on the target material based on a pre-designed experimental scheme to obtain quasi-static stress-strain data.

[0378] Based on the pre-set experimental scheme, a split Hopkinson bar dynamic compression test was conducted on the target material to obtain dynamic stress-strain data at multiple strain rate levels.

[0379] High-temperature tensile tests were conducted on the target material based on a pre-designed experimental scheme to obtain high-temperature stress-strain data at multiple temperature levels.

[0380] Based on a pre-designed experimental scheme, the shock wave pressure during the laser shock process of the target material was measured to obtain shock wave pressure time history data.

[0381] Optionally, in one possible implementation, the first processing module 1002 is further configured to:

[0382] The quasi-static yield strength, strain hardening coefficient and hardening index in the simulation parameters are obtained by fitting the quasi-static stress-strain data.

[0383] The strain rate sensitivity coefficient in the simulation parameters is obtained by fitting dynamic stress-strain data.

[0384] The thermal softening index in the simulation parameters is obtained by fitting high-temperature stress-strain data.

[0385] Based on the shock wave pressure time history data, the energy conversion efficiency in the preset confined expansion model is calibrated, and the load-pressure curve in the simulation parameters is generated by combining the preset confined expansion model.

[0386] Optionally, in one possible implementation, the second processing module 1003 is further configured to:

[0387] For each type of laser parameter in the preset laser parameters, determine the corresponding value range for the laser parameter;

[0388] For each laser parameter, values ​​are taken within its corresponding range to obtain multiple sets of simulated laser parameter values;

[0389] Laser shock simulation is performed on the target material based on multiple sets of simulated laser parameters, and residual stress distribution data points corresponding to each set of simulated laser parameters are obtained, and simulation results corresponding to each set of simulated laser parameters are generated.

[0390] Optionally, in one possible implementation, the second processing module 1003 is further configured to:

[0391] The load-pressure curve is used as the dynamic load, and the remaining parameters in the simulation parameters are used as material properties. Explicit dynamic analysis is performed based on the dynamic load and material properties to simulate the propagation of the dynamic load in the target material and the response of the target material, thus obtaining the final state of the dynamic response. The final state of the dynamic response is a snapshot of the instantaneous state of the target material at the moment the dynamic load ends.

[0392] Using the final state of the dynamic response as the initial condition, an implicit static analysis is performed on the target material after the dynamic load is removed to solve the static equilibrium residual stress field and obtain the residual stress distribution data points.

[0393] Figure 11 The schematic diagram of the laser shock osmosis corrosion protection device provided in this application is as follows: Figure 11 As shown, the laser shock osmosis corrosion protection device provided in this embodiment includes:

[0394] The acquisition module 1101 is used to acquire the anti-corrosion requirements for the target material and generate the target residual stress field index corresponding to the target material based on the anti-corrosion requirements.

[0395] The first processing module 1102 is used to construct an inverse optimization problem based on a pre-built laser shock strengthening stress mapping model and a target residual stress field index. The objective function of the inverse optimization problem is the laser processing energy consumption function, and the constraint condition is that at least one of the fitting parameters corresponding to the laser shock strengthening stress mapping model satisfies the target residual stress field index.

[0396] The second processing module 1103 is used to solve the inverse optimization problem to obtain a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function.

[0397] The third processing module 1104 is used to perform laser shock strengthening treatment on the target material according to the optimal laser parameters.

[0398] Optionally, in one possible implementation, the second processing module 1103 is further configured to:

[0399] Based on the range of laser parameters, multiple initial iteration points are generated.

[0400] For each initial iteration point, a set of locally optimal laser parameter values ​​is obtained by iteratively solving a sequential quadratic programming problem based on the inverse optimization problem.

[0401] Compare the objective function values ​​corresponding to each group of locally optimal laser parameter values, and determine the group of locally optimal laser parameter values ​​with the lowest objective function value as the group of optimal laser parameter values ​​with the smallest laser processing energy consumption function value.

[0402] Optionally, in one possible implementation, the second processing module 1103 is further configured to:

[0403] In each iteration, the objective function is approximated twice at the current iteration point, and the constraints are approximated linearly at the current iteration point, resulting in the quadratic programming subproblem corresponding to the iteration.

[0404] Solve the quadratic programming subproblem to obtain the search direction.

[0405] The step size is determined by performing a line search based on the search direction, and the current iteration point is updated.

[0406] When the current iteration point meets the preset convergence condition, the current iteration point is determined as a set of locally optimal laser parameter values.

[0407] If the current iteration point does not meet the preset convergence condition, proceed to the next iteration round until the current iteration point meets the preset convergence condition.

[0408] Optionally, in one possible implementation, the preset convergence conditions include:

[0409] The change in the objective function value between two consecutive iterations is less than a first preset threshold; and / or,

[0410] The constraint violation at the current iteration point is less than the second preset threshold.

[0411] Optionally, in one possible implementation, the constraints of the inverse optimization problem also include the range of values ​​for the laser parameters; when solving the inverse optimization problem, the range of values ​​is used as the upper and lower boundary constraints of the laser parameters.

[0412] The apparatus provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0413] Figure 12 A schematic diagram of the structure of the electronic device provided in this application. Figure 12 As shown, the electronic device provided in this embodiment includes at least one processor 1201 and a memory 1202. Optionally, the device further includes a communication component 1203. The processor 1201, memory 1202, and communication component 1203 are connected via a bus 1204.

[0414] In a specific implementation, at least one processor 1201 executes computer execution instructions stored in memory 1202, causing at least one processor 1201 to execute the above-described (invention title) method.

[0415] The specific implementation process of processor 1201 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0416] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0417] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0418] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0419] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0420] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0421] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0422] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0423] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0424] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0425] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0426] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0427] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0428] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for constructing a laser shock peening stress mapping model, characterized in that, include: Laser shock experiments were conducted on the target material based on a pre-designed experimental scheme to obtain experimental data. Based on the experimental data, data analysis is performed to obtain the simulation parameters corresponding to the target material; Based on the simulation parameters and preset laser parameters, multiple laser shock simulations were performed on the target material to obtain multiple simulation results; wherein, each simulation result includes a set of preset laser parameter simulation values ​​and a set of residual stress distribution data points; For each set of residual stress distribution data points, a fitting calculation is performed to obtain a set of fitting parameter values; the fitting parameters include the surface residual compressive stress and the depth of compressive stress influence. Nonlinear regression processing is performed based on the simulated values ​​of each set of preset laser parameters and their corresponding fitted parameter values ​​to obtain the mapping relationship between each fitted parameter and the preset laser parameters, and a laser shock strengthening stress mapping model is generated based on the mapping relationship.

2. The method according to claim 1, characterized in that, The laser shock experiment on the target material based on the preset experimental scheme yields experimental data, including: Based on the preset experimental scheme, a quasi-static tensile test was conducted on the target material to obtain quasi-static stress-strain data. Based on the preset experimental scheme, a split Hopkinson bar dynamic compression test was conducted on the target material to obtain dynamic stress-strain data at multiple strain rate levels. Based on the preset experimental scheme, a high-temperature tensile test was conducted on the target material to obtain high-temperature stress-strain data at multiple temperature levels. Based on the preset experimental scheme, the shock wave pressure during the laser shock process of the target material was measured to obtain shock wave pressure time history data.

3. The method according to claim 2, characterized in that, The data analysis based on the experimental data to obtain the simulation parameters corresponding to the target material includes: Based on the quasi-static stress-strain data, the quasi-static yield strength, strain hardening coefficient, and hardening index in the simulation parameters are obtained by fitting the data. The strain rate sensitivity coefficient in the simulation parameters is obtained by fitting the dynamic stress-strain data. The thermal softening index in the simulation parameters is obtained by fitting the high-temperature stress-strain data. Based on the shock wave pressure time history data, the energy conversion efficiency in the preset confined expansion model is calibrated, and the load-pressure curve in the simulation parameters is generated in combination with the preset confined expansion model.

4. The method according to claim 3, characterized in that, The target material is subjected to multiple laser shock simulations based on the simulation parameters and preset laser parameters, resulting in multiple simulation results, including: For each type of laser parameter in the preset laser parameters, determine the corresponding value range of the laser parameter; For each laser parameter, values ​​are taken within its corresponding range to obtain multiple sets of simulated laser parameter values; Based on the simulated values ​​of the multiple sets of laser parameters and the simulation parameters, laser shock simulation is performed on the target material to obtain residual stress distribution data points corresponding to each set of simulated laser parameters, and simulation results corresponding to each set of simulated laser parameters are generated.

5. The method according to claim 4, characterized in that, The laser shock simulation of the target material based on the multiple sets of simulated laser parameters and the simulation parameters is used to obtain residual stress distribution data points corresponding to each set of simulated laser parameters, including: The load-pressure curve is used as the dynamic load, and the remaining parameters in the simulation parameters are used as material properties. Explicit dynamic analysis is performed based on the dynamic load and the material properties to simulate the propagation of the dynamic load in the target material and the response of the target material, thereby obtaining the final state of the dynamic response. The final state of the dynamic response is a snapshot of the instantaneous state of the target material at the moment the dynamic load ends. Using the final state of the dynamic response as the initial condition, an implicit static analysis is performed on the target material after the dynamic load is removed to solve the static equilibrium residual stress field and obtain the residual stress distribution data points.

6. A laser shock osmosis corrosion protection method, characterized in that, The method is implemented based on the laser shock strengthening stress mapping model according to any one of claims 1-5, and the method includes: Obtain the corrosion protection requirements for the target material, and generate the target residual stress field index corresponding to the target material based on the corrosion protection requirements; An inverse optimization problem is constructed based on a pre-built laser shock strengthening stress mapping model and the target residual stress field index. The objective function of the inverse optimization problem is the laser processing energy consumption function, and the constraint condition is that at least one of the fitting parameters corresponding to the laser shock strengthening stress mapping model satisfies the target residual stress field index. Solving the inverse optimization problem yields a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function. The target material is subjected to laser shock peening treatment according to the optimal laser parameters.

7. The method according to claim 6, characterized in that, The process of solving the inverse optimization problem yields a set of optimal laser parameter values ​​that satisfy the target residual stress field index and minimize the laser processing energy consumption function, including: Based on the range of values ​​for the laser parameters, multiple initial iteration points are generated; For each initial iteration point, a sequential quadratic programming iteration is performed based on the inverse optimization problem to obtain a set of locally optimal laser parameter values ​​corresponding to each initial iteration point; Compare the objective function values ​​corresponding to each group of locally optimal laser parameter values, and determine the group of locally optimal laser parameter values ​​with the lowest objective function value as the group of optimal laser parameter values ​​with the smallest laser processing energy consumption function value.

8. The method according to claim 7, characterized in that, For each initial iteration point, a sequential quadratic programming iteration is performed based on the inverse optimization problem to obtain a set of locally optimal laser parameter values ​​corresponding to each initial iteration point, including: In each iteration, the objective function is approximated twice at the current iteration point, and the constraints are approximated linearly at the current iteration point to obtain the quadratic programming subproblem corresponding to the iteration. Solve the quadratic programming subproblem to obtain the search direction; Based on the search direction, a line search is performed to determine the step size, and the current iteration point is updated; When the current iteration point satisfies the preset convergence condition, the current iteration point is determined as the set of locally optimal laser parameter values; If the current iteration point does not meet the preset convergence condition, proceed to the next iteration round until the current iteration point meets the preset convergence condition.

9. The method according to claim 8, characterized in that, The preset convergence conditions include: The change in the objective function value between two consecutive iterations is less than a first preset threshold; and / or, The constraint violation at the current iteration point is less than the second preset threshold.

10. The method according to claim 6, characterized in that, The constraints of the inverse optimization problem also include the range of values ​​for the laser parameters; when solving the inverse optimization problem, the range of values ​​is used as the upper and lower boundary constraints for the laser parameters.

Images