Automatic SHPB data processing method based on genetic algorithm

Abstract

The invention discloses an automatic SHPB data processing method based on a genetic algorithm, and belongs to the technical field of SHPB data extraction and processing. The system comprises a data acquisition module, a waveform processing module and a parameter determination module, the method is realized based on the system, the wave alignment method provided by the invention is better in consistency, different people have different wave alignment results on the same group of waveform curves, but the value is converged to the minimum value based on the genetic algorithm, and the method is more accurate. The consistency of data processing is ensured; besides, compared with a traditional single curve fitting method, the method has the advantages that the requirement for multiple times of fitting is avoided, all undetermined parameters of the constitutive model can be obtained at a time, the data processing time of a single group of SHPB is shortened to be within 2 minutes from 30 minutes, and the error is theoretically minimum.

Description

Technical Field

[0001] The present invention relates to the technical field of SHPB data extraction and processing, and in particular to an automatic SHPB data processing method based on a genetic algorithm. Background Art

[0002] The Split Hopkinson Pressure Bar (SHPB) test is an experimental method widely used in testing the dynamic mechanical properties of materials. The SHPB test measures the stress-strain behavior of the material at different strain rates by applying an impact load with a high strain rate. A typical SHPB test device includes an incident rod, a transmission rod and a sample to be tested. Strain gauges are arranged on the incident rod and the transmission rod to measure the strain signal. The SHPB experimental process is as follows: 1. Loading process: The shock wave is loaded onto the incident rod through an air gun or other high-speed loading device. 2. Strain measurement: The strain gauge records the strain time history signals of the incident wave, reflected wave and transmitted wave. 3. Data acquisition: These strain signals are recorded and stored by the data acquisition system to form a time-strain curve. 4. Waveform processing: The time-strain curve is symmetrical, and the time-strain curve is converted into a stress-strain curve using the two-wave method or the three-wave method. 5. Determine the constitutive parameters: Obtain the parameters of the given constitutive model based on the stress-strain curve. The simulation process of SHPB also has a similar process, such as Figure 2 As shown in the figure, the bullet impact is simulated by applying a waveform load to the specimen at the incident rod end. The historical variable output unit is set on the incident rod and the transmission rod to output the strain value on the unit to obtain the incident wave, reflected wave and transmitted wave signals. After the SHPB simulation, the same waveform processing and constitutive parameter determination steps as the SHPB test are required to obtain the equivalent stress-strain relationship of the specimen in the simulation.

[0003] Since the incident wave, transmitted wave and reflected wave have different starting points on the time-strain curve, it is necessary to move the waveform so that the incident wave, transmitted wave and reflected wave start at the same time starting point, that is, the wave. By adjusting the parameters in the constitutive model, the curve of the constitutive model and the stress-strain curve obtained from the experiment are made as consistent as possible, that is, the constitutive parameters are determined. In the SHPB (split Hopkinson pressure bar) experiment or simulation, the determination of the wave and constitutive parameters is a key link in data processing and directly determines the accuracy of the final result. However, this process is usually time-consuming, and the rationality of the parameters is crucial to the results. Therefore, ensuring that the determination of the wave and constitutive parameters is accurate and reasonable is a core step to improve the reliability of experimental data and processing efficiency.

[0004] Currently, the determination of waves and parameters mainly relies on manual processing, which has the following obvious deficiencies: Large subjective error: The manual process of waves requires experimenters to manually adjust the waveform. Without a strict and unified standard, it is easy to introduce subjective errors. This human factor leads to poor consistency and repeatability of the results, affecting the reliability of test data. Low efficiency: The manual wave alignment process is cumbersome and time-consuming. Especially when dealing with a large amount of test data, the efficiency is extremely low. This inefficient working method not only increases the workload of experimenters but also limits the processing speed and analysis efficiency of experimental data.

[0005] The determination of waves and parameters plays a crucial role in the data processing process. They directly affect the accuracy and reliability of test results. Therefore, improving the automation and precision of wave and parameter determination is crucial for improving the processing efficiency and consistency of test data.

[0006] Chinese Patent CN108375501A discloses a data processing method based on split Hopkinson bar test technology. This method uses the method of manually selecting points one by one for wave alignment. Although this method can accurately process each data point, its efficiency is low and it takes a long time to process a large amount of data. Compared with this patent, the data processing efficiency of the method of this patent is greatly improved. The processing time of each group of SHPB test data is reduced from 30 minutes to within 2 minutes, and the error value of the wave alignment effect reaches the minimum theoretically.

[0007] Chinese Patent CN11770893A discloses a method for extracting stress waves in SHPB tests based on Python. This method uses the Python language to process oscillation data and extracts a smooth time-strain curve. However, this method does not solve the problem of wave alignment and cannot achieve the automated processing of SHPB test data. Manual wave alignment is still required. This patent only solves the problem of waveform smoothing and does not solve the problems of wave alignment and parameter determination.

[0008] In summary, there are still obvious deficiencies in the current methods for wave alignment and parameter determination in SHPB. There is an urgent need for a new method that can automate and efficiently process wave alignment and parameter determination to improve the processing efficiency and consistency of test data and reduce the influence of subjective errors. Summary of the Invention

[0009] The purpose of the present invention is to provide an automated SHPB data processing method based on genetic algorithms to solve the problems mentioned in the background technology. The present invention performs wave alignment on SHPB data and determines constitutive parameters based on genetic algorithms, with high automation, accurate determination of constitutive parameters, and good consistency.

[0010] To achieve the above-mentioned invention purpose, the present invention provides the following technical solutions:

[0011] A genetic algorithm-based automatic SHPB data processing system, comprising a data acquisition module, a waveform processing module, and a parameter determination module, wherein:

[0012] The data acquisition module acquires waveform data through an SHPB test platform or a simulation platform. The waveform data of the test platform is obtained through sensors or measuring devices, and the waveform data of the simulation platform is obtained by setting the unit strain history variable;

[0013] The waveform processing module includes a preprocessing unit and an automatic wave alignment unit, which are used to preprocess the waveform data and perform wave alignment through a genetic algorithm;

[0014] The parameter determination module is used to encode the constitutive parameter determination problem as an optimization problem, find a set of parameters such that the fitting curve of the constitutive parameters is close to the stress-strain curve obtained from experiments or simulations; according to the stress-strain curves obtained from multiple experiments or simulations, use the genetic algorithm to solve the constitutive parameters and determine the optimal solution.

[0015] A genetic algorithm-based automatic SHPB data processing method, comprising the following steps:

[0016] S1. Use the data acquisition module to acquire waveform data, which is obtained through an SHPB test platform or a simulation platform. Among them, the waveform data of the test platform is obtained through sensors or measuring devices, and the waveform data of the simulation platform is obtained by setting the unit strain history variable;

[0017] S2. Use the waveform processing module to preprocess the waveform data obtained in S1 and perform wave alignment through a genetic algorithm;

[0018] S3. Use the parameter determination module to encode the constitutive parameter determination problem as an optimization problem, find a set of parameters such that the fitting curve of the constitutive parameters is close to the stress-strain curve obtained from experiments or simulations; according to the stress-strain curves obtained from multiple experiments or simulations, use the genetic algorithm to solve the constitutive parameters and determine the optimal solution.

[0019] Preferably, the preprocessing of the waveform data in S2 specifically refers to automatically smoothing, interpolating, and cropping the input waveform using the preprocessing unit to obtain separate incident waves, reflected waves, and transmitted waves for subsequent wave alignment processing.

[0020] Preferably, the wave alignment through the genetic algorithm in S2 specifically includes the following steps:

[0021] Use the automatic wave alignment unit to take the movement amounts of the transmitted wave and the reflected wave as optimization parameters, and transform the problem into an optimization problem: find a set of movement amounts of the transmitted wave and the reflected wave such that the sum of the incident wave and the reflected wave satisfies the balance relationship with the transmitted wave;

[0022] Solve the above optimization problem using a genetic algorithm:

[0023] Encode the independent variables, and then generate an initial population as the starting point of the algorithm;

[0024] Calculate the fitness function to evaluate the effect of the initial population corresponding to the independent variables;

[0025] Perform selection, crossover, and mutation operations on the population to optimize the independent variables through the genetic algorithm;

[0026] During the optimization process, the fitness function continuously decreases, and the independent variables corresponding to the population gradually converge to the optimal solution;

[0027] After the wave alignment is completed, use the three-wave method for processing to obtain the stress-strain curve;

[0028] The specific process of the genetic algorithm for wave alignment is as follows: During the process of using the genetic algorithm for wave alignment, the movement amounts of the reflected wave and the transmitted wave are used as variables, and when "incident wave + reflected wave = transmitted wave", it indicates that the wave alignment is successful.

[0029] Preferably, during the process of using the genetic algorithm to solve the constitutive parameters, the parameters of the constitutive model are used as independent variables. When a set of parameters is found such that the curve of the constitutive model fits the stress-strain curve obtained from experiments or numerical simulations, the determination of the constitutive parameters is successful.

[0030] The present invention further protects a computer device, which includes a processor and a memory. At least one instruction, at least one program, a code set, or an instruction set is stored in the memory, and the instruction, program, code set, or instruction set is loaded and executed by the processor to implement the above-mentioned SHPB data automatic processing method based on the genetic algorithm.

[0031] The present invention further protects a computer-readable storage medium, in which at least one instruction, at least one program, a code set, or an instruction set is stored, and the instruction, program, code set, or instruction set is loaded and executed by the processor to implement the above-mentioned SHPB data automatic processing method based on the genetic algorithm.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0033] (1) Compared with the existing SHPB data processing technology, the present invention adopts a wave alignment method based on the genetic algorithm. Compared with the traditional method of manually finding the wave head and wave tail, it can better meet the requirement of incident wave + reflected wave = transmitted wave, and the efficiency is significantly improved.

[0034] (2) Compared with the traditional method of manually finding the wave head and wave tail, the wave alignment method based on genetic algorithm proposed by the present invention has better consistency. On the same set of waveform curves, different people may obtain different wave alignment results, but the genetic algorithm converges to the minimum value numerically, ensuring the consistency of data processing.

[0035] (3) The present invention adopts a method for overall determination of constitutive parameters based on genetic algorithm, considering multiple stress-strain curves together. Compared with the traditional single-curve fitting method, it avoids the need for multiple fittings. All undetermined parameters of the constitutive model can be obtained at one time, shortening the processing time of single-group SHPB data from 30 minutes to within 2 minutes, and achieving the minimum error theoretically.

[0036] (4) The present invention has good scalability. The SHPB test or simulation data processing system based on genetic algorithm can be easily extended to SHTB test or simulation data processing and the determination of parameters of different constitutive models, and the core part can be implemented without modification. Brief Description of the Drawings

[0037] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings involved in the embodiments are briefly introduced below. Obviously, the drawings in the following description are only schematic illustrations of some embodiments of the present invention. For those skilled in the art, other forms of drawings can be constructed based on these drawings without creative efforts.

[0038] Figure 1 It is a module and connection relationship diagram of an SHPB data automatic processing system based on genetic algorithm proposed in Embodiment 1 of the present invention;

[0039] Figure 2 It is a schematic diagram of the SHPB test platform proposed in the background technology and Embodiment 1 of the present invention;

[0040] Figure 3 It is a schematic diagram of the SHPB simulation platform proposed in Embodiment 1 of the present invention;

[0041] Figure 4 It is a schematic diagram of waveform smoothing, interpolation, and clipping proposed in Embodiment 1 of the present invention;

[0042] Figure 5 It is the principle and flowchart of the genetic algorithm proposed in Embodiment 1 of the present invention;

[0043] Figure 6 It is a schematic diagram of waveform movement proposed in Embodiment 1 of the present invention;

[0044] Figure 7 It is a graph showing the change of the error value of the genetic algorithm with the number of genetic generations proposed in Embodiment 1 of the present invention;

[0045] Figures 8 - 10 For the waveform movement during the automatic wave alignment proposed in Embodiment 1 of the present invention ( Figure 7 Points A, B, and C in

[0046] Figure 11 Figure of multiple groups of simulated curves after the automatic waveform processing proposed in Embodiment 1 of the present invention;

[0047] Figure 12 Schematic diagram of the process of automatically determining parameters proposed in Embodiment 1 of the present invention;

[0048] Figure 13 Comparison diagram of the fitting value and the simulated value after the automatic parameter determination proposed in Embodiment 1 of the present invention;

[0049] Figure 14 Schematic diagram of the original data of the SHPB test proposed in Embodiment 2 of the present invention;

[0050] Figure 15 Schematic diagram of the wave alignment result proposed in Embodiment 2 of the present invention;

[0051] Figure 16 Schematic diagram of the determination result of the constitutive parameters proposed in Embodiment 2 of the present invention. Detailed implementation manners

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0053] Embodiment 1:

[0054] The present invention proposes an SHPB data automatic processing system based on a genetic algorithm, in which the data acquisition module, waveform processing module, and parameter determination module of the automatic processing system (the connection relationships of each module are as Figure 1 shown), and further proposes an SHPB data automatic processing method based on the proposed system, which specifically includes the following contents:

[0055] Data acquisition module: Through the SHPB test platform (such as Figure 2 shown) or the simulation platform (such as Figure 3 shown), obtain the waveform as shown in Figure 4 . In the experiment, this data usually comes from sensors or measuring devices, such as strain gauges. In numerical simulation, this data can be obtained by setting the unit strain history variable.

[0056] Waveform processing module: Preprocess the waveform and perform wave alignment through the genetic algorithm, specifically including:

[0057] First, preprocess the waveform. For example, Figure 4 as shown, this module automatically smooths, interpolates, and clips the input waveform to obtain separate incident waves, reflected waves, and transmitted waves, facilitating subsequent wave processing.

[0058] For the automatic wave part, use the movement amounts of the transmitted wave and the reflected wave as optimization parameters. For example, Figure 6 as shown, transform the problem into an optimization problem: find a set of movement amounts of the transmitted wave (x t ) and the reflected wave (x r ) such that the sum of the incident wave (I) and the reflected wave (R) is as equal as possible to the transmitted wave (T), that is, satisfy the balance relationship.

[0059] Use the genetic algorithm to solve the above optimization problem. The principle of the genetic algorithm is as Figure 5 shown.

[0060] First, encode the independent variables, then generate an initial population as the starting point of the algorithm; calculate the fitness function to evaluate the effects of the independent variables corresponding to the initial population; then perform selection, crossover, and mutation operations on the population, and continuously optimize the independent variables through the genetic algorithm; during this process, the fitness function continuously decreases, and the independent variables corresponding to the population gradually converge to the optimal solution; after processing the wave, use the three-wave method for processing and use the volume-invariant property to convert to obtain the true stress-strain curve.

[0061] The above process specifically includes the following contents:

[0062] 1. Generation of the initial population

[0063] Generate a set of initial solutions (x t0 , x r0 ). The initial population can be randomly generated or generated based on certain heuristic rules.

[0064] 2. Definition of the fitness function

[0065] Define the fitness function to evaluate the quality of each solution. Define the fitness function as:

[0066]

[0067] where I i , T i , R i are the i-th sampling points on the incident wave, transmitted wave, and reflected wave after movement, respectively, with a total of n sampling points; the number of sampling points should ensure that there are at least 100 sampling points within the waveform pulse width.

[0068] 3. Selection, crossover, and mutation

[0069] Use the basic operations of the genetic algorithm to evolve the population:

[0070] Selection: Select better individuals according to the fitness value to enter the next generation.

[0071] Crossover: Perform crossover operations on the selected individuals to generate new individuals.

[0072] Mutation: Randomly mutate the individuals generated by crossover to increase the diversity of solutions.

[0073] Generate new parameters (x t , x r ) through the above operations.

[0074] 4. Convergence judgment

[0075] Iteratively execute the selection, crossover, and mutation operations until the convergence condition is met (the fitness value reaches a certain threshold, such as 1%, or the number of iterations reaches the upper limit, such as 100 times). The convergence iteration process is as Figure 7 shown. As the number of genetic generations increases, the error value gradually decreases. In Figure 7 , the wave pair situations of three points A, B, and C are intercepted, as Figures 8 - 10 shown. The error is relatively large after point A moves. As the number of generations increases, a better point (B) is gradually found. At this time, the incident wave + reflected wave is basically equal to the transmitted wave. After further genetic iteration, it enters the convergence to reach the optimal solution (C), and at this time the error value reaches the theoretically minimum value.

[0076] 5. Three-wave method processing

[0077] Use the three-wave method to process the stress wave data:

[0078] The three-wave method has the best credibility in processing data based on the incident wave, reflected wave, and transmitted wave and maximally avoids human errors in data processing. The formula is as follows:

[0079]

[0080] where the subscripts i, r, and t represent the incident wave, reflected wave, and transmitted wave respectively; b and s represent the rod and the specimen; E and A are the Young's modulus and area. The stress-strain curves obtained through multiple tests or simulation processes are as Figure 11 shown in the curve in.

[0081] 6. Convert the true stress-strain curve

[0082] What is obtained after the three-wave method processing is the engineering stress-strain curve, and it is still necessary to further process it to convert the engineering stress-strain curve into the true stress-strain curve. According to the volume invariance, the conversion formula is:

[0083] σ = σs (1 - ε s )

[0084] ε = -ln(1 - ε s )

[0085] The specific process of the genetic algorithm for wave is as follows: In the process of using the genetic algorithm for wave, taking the movement amounts x t 、x r of the reflected wave and the transmitted wave as variables, when the incident wave + the reflected wave = the transmitted wave, the wave is successfully paired.

[0086] Parameter determination module: Encodes the constitutive parameter determination problem as an optimization problem. Taking the determination of Johnson - Cook constitutive parameters as an example, this problem is transformed into: finding a set of parameters (A, B, n, c) such that the fitting curve of the Johnson - Cook constitutive and the stress - strain curve obtained from experiments or simulations is as close as possible. Based on the stress - strain curves obtained from multiple experiments or simulations, the genetic algorithm can be used to solve and determine the optimal solution.

[0087] The specific process of using the genetic algorithm to determine constitutive parameters is as follows: In the process of using the genetic algorithm to determine parameters, taking the parameters of the constitutive model as independent variables, such as A (initial yield parameter), B (strain hardening parameter), c (strain rate parameter), and n (strain hardening exponent) in the Johnston - Cook constitutive. When a set of parameters is found such that the curve of the constitutive model fits the stress - strain curve obtained from experiments or numerical simulations, the parameter determination is successful.

[0088] 1. Generation of the initial population

[0089] Generate a set of initial solutions (A0, B0, c0, n0). The initial population can be randomly generated or generated based on certain heuristic rules.

[0090] 2. Definition of the fitness function

[0091] Define the fitness function to evaluate the quality of each solution. Here, the constitutive parameter determination is defined according to the constitutive model. For example, the fitness function of the Johnson - Cook constitutive is defined as:

[0092]

[0093] Among them, σ JCIt is the curve of the Johnson-Cook constitutive model corresponding to a certain set of parameters; σ represents the stress-strain curve obtained from experiments or simulations; the fitness function means the sum of the mean square errors between the constitutive model values and the experimental or simulation values at k sampling points among m curves. In order to obtain the strain rate parameter c at one time, at least 3 or more experimental or simulation curves under different strain rates are used to determine the parameters simultaneously, that is, m≥3; the number of sampling points k should ensure that there are at least 100 sampling points within the waveform pulse width.

[0094] 3. Selection, Crossover and Mutation

[0095] Use the basic operations of the genetic algorithm to evolve the population:

[0096] Selection: Select better individuals according to the fitness values to enter the next generation.

[0097] Crossover: Perform crossover operations on the selected individuals to generate new individuals.

[0098] Mutation: Randomly mutate the individuals generated by crossover to increase the diversity of solutions.

[0099] Generate new parameters (A, B, n, c) based on the above process.

[0100] 4 Convergence Judgment

[0101] Iteratively execute the selection, crossover and mutation operations until the convergence condition is met (the fitness value reaches a certain threshold, such as 1%, or the number of iterations reaches the upper limit, such as 100 times). The convergence iteration process takes a curve at one strain rate as an example Figure 12 As shown, a set of parameters (A1, B1, c1, n1) is overall on the small side. After iterative optimization, another set of parameters (A2, B2, c2, n2) is obtained. At this time, the curve of the constitutive model fits the curve of the parameters to be determined better. As the number of genetic generations increases, the fitness function further decreases, and parameters (A3, B3, c3, n3) are obtained. At this time, the Johnson-Cook constitutive curve fits the experimental or simulation curve of the parameters to be determined well. The result image of the determination of the constitutive parameters of a group of curves is as Figure 13 shown. The Johnson-Cook model and the simulation results are in good agreement on multiple groups of curves, achieving the expected effect.

[0102] Example 2:

[0103] Based on Example 1 but with differences. In this example, taking the SHPB experiment of a certain ductile metal as an example, combined with the specific process and module proposed by the present invention, the working principle and technical advantages are described as follows.

[0104] 1. Data Acquisition:

[0105] The original waveform data obtained through the SHPB test platform is as follows Figure 14 shown, including incident wave, reflected wave, and transmitted wave signals. These data are collected by strain gauges with a sampling frequency of 1 MHz, ensuring that there are more than 1000 sampling points within the waveform pulse width.

[0106] Interpolation: Cubic spline interpolation is used for non-uniformly sampled data to ensure the uniformity of the waveform time series.

[0107] Cropping: The effective waveform segment is intercepted, and the invalid signal segment is removed.

[0108] 2. Waveform processing:

[0109] Definition of wave variables by genetic algorithm: The movement of the reflected wave (x r ) and the movement of the transmitted wave (x t ) are used as optimization variables to ensure that the balance relationship of incident wave + reflected wave ≈ transmitted wave is satisfied after movement.

[0110] Fitness function: The root mean square error (RMSE) is defined as the objective function:

[0111]

[0112] Genetic algorithm parameters: The population size is 50, the crossover probability is 0.8, the mutation probability is 0.05, and the maximum number of iterations is 200.

[0113] The algorithm converges at the 135th generation of iteration (fitness value < 1%), and the wave processing results are as follows Figure 15 shown.

[0114] 3. Determination of Johnson-Cook constitutive parameters:

[0115] Through three independent SHPB experiments (strain rates are 1200 s-1, 1600 s-1, 3000 s-1 respectively), three groups of true stress-strain curves (as shown in Figure 16 ) are obtained as the input of the genetic algorithm.

[0116] Definition of optimization variables for genetic algorithm parameters: The Johnson-Cook model parameters are A (initial yield stress), B (hardening coefficient), n (hardening exponent), and c (strain rate sensitivity coefficient).

[0117]

[0118] Among them, m is the number of strain rate groups (m = 3), and k is 100 sampling points for each curve.

[0119] Algorithm parameters: The population size is 100, the adaptive crossover and mutation probabilities, and the maximum number of iterations is 200 times.

[0120] As Figure 16 shown, the genetic algorithm converges after 150 iterations (fitness value < 1%).

[0121] The root mean square error (RMSE) between the fitted curve and the experimental data is 1.8%, indicating that the parameters are globally optimal.

[0122] In summary, the present invention has the following beneficial effects:

[0123] (1) Full-process automation: From waveform processing to parameter determination, no manual intervention is required throughout the process, and the data processing time is shortened from the traditional 30 minutes to 5 minutes.

[0124] (2) Precision and consistency: The global optimization of the genetic algorithm makes the parameter error < 2%, and the result is independent of the operator.

[0125] (3) Scalability: The same framework can be applied to SHTB (Split Hopkinson Pressure Bar) or different constitutive models (such as Zerilli - Armstrong), and only the fitness function needs to be modified.

[0126] It should be noted that in this invention patent, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations.

[0127] The above is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. An SHPB data automatic processing system based on genetic algorithm, characterized in that It includes a data acquisition module, a waveform processing module and a parameter determination module, where: The data acquisition module acquires waveform data through an SHPB test platform or a simulation platform. The waveform data of the test platform is obtained through sensors or measuring devices, and the waveform data of the simulation platform is obtained by setting the unit strain history variable; The waveform processing module includes a preprocessing unit and an automatic wave alignment unit, which are used to preprocess the waveform data and perform wave alignment through a genetic algorithm; The parameter determination module is used to encode the constitutive parameter determination problem as an optimization problem, find a set of parameters so that the fitting curve of the constitutive parameters is close to the stress-strain curve obtained from experiments or simulations; according to the stress-strain curves obtained from multiple experiments or simulations, use the genetic algorithm to solve the constitutive parameters and determine the optimal solution.

2. A method for automatically processing SHPB data based on a genetic algorithm implemented using the system described in claim 1, characterized in that, It includes the following: S1. Use the data acquisition module to acquire waveform data, which is obtained through an SHPB test platform or a simulation platform. Among them, the waveform data of the test platform is obtained through sensors or measuring devices, and the waveform data of the simulation platform is obtained by setting the unit strain history variable; S2. Use the waveform processing module to preprocess the waveform data obtained in S1 and perform wave alignment through a genetic algorithm; S3. Use the parameter determination module to encode the constitutive parameter determination problem as an optimization problem, find a set of parameters so that the fitting curve of the constitutive parameters is close to the stress-strain curve obtained from experiments or simulations; according to the stress-strain curves obtained from multiple experiments or simulations, use the genetic algorithm to solve the constitutive parameters and determine the optimal solution.

3. The automated SHPB data processing method based on genetic algorithm according to claim 2, characterized in that, The preprocessing of the waveform data in S2 specifically refers to automatically smoothing, interpolating and clipping the input waveform by the preprocessing unit to obtain separate incident waves, reflected waves and transmitted waves for subsequent wave alignment processing.

4. A SHPB data automatic processing method based on genetic algorithm according to claim 3, characterized in that The wave alignment through the genetic algorithm in S2 specifically includes the following: Use the automatic wave alignment unit to take the movement amounts of the transmitted wave and the reflected wave as optimization parameters, and transform the problem into an optimization problem: find a set of movement amounts of the transmitted wave and the reflected wave so that the sum of the incident wave and the reflected wave satisfies the balance relationship with the transmitted wave; Use the genetic algorithm to solve the above optimization problem: Encode the independent variables, and then generate an initial population as the starting point of the algorithm; Calculate the fitness function to evaluate the effect of the initial population corresponding to the independent variables; Perform selection, crossover and mutation operations on the population to optimize the independent variables through the genetic algorithm; During the optimization process, the fitness function continuously decreases, and the independent variables corresponding to the population gradually converge to the optimal solution; After the wave alignment is completed, use the three-wave method for processing to obtain the stress-strain curve; The specific process of wave alignment by the genetic algorithm is as follows: during the process of using the genetic algorithm for wave alignment, taking the movement amounts of the reflected wave and the transmitted wave as variables, when "incident wave + reflected wave = transmitted wave", it means that the wave alignment is successful.

5. A method for automatically processing SHPB data based on genetic algorithm according to claim 4, characterized in that, During the process of using the genetic algorithm to solve the constitutive parameters, taking the parameters of the constitutive model as independent variables, when a set of parameters is found so that the curve of the constitutive model fits the stress-strain curve obtained from experiments or numerical simulations, the determination of the constitutive parameters is successful.

6. A computer device, characterized in that, The computer device includes a processor and a memory, and at least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the instruction, program, code set or instruction set is loaded and executed by the processor to implement the genetic algorithm-based SHPB data automatic processing method according to any one of claims 2-5.

7. A computer-readable storage medium, characterized in that, At least one instruction, at least one program, a code set or an instruction set is stored in the computer-readable storage medium, and the instruction, program, code set or instruction set is loaded and executed by the processor to implement the genetic algorithm-based SHPB data automatic processing method according to any one of claims 2-5.

Images