Method and equipment for predicting mechanical property of light material part and storage medium

Abstract

The invention discloses a method and device for predicting the mechanical property of a light material component and a storage medium, and the method comprises the steps: obtaining a first parameter of a hardening model and a second parameter of a failure model corresponding to a light material; taking the first parameter and the second parameter as parameters of an initial prediction model, and optimizing the parameters of the initial prediction model based on a back propagation neural network and a genetic algorithm to obtain a third parameter; updating the parameters of the initial prediction model to the third parameters to obtain a target prediction model; and predicting the plastic deformation index and failure strain of the light material part by using the target prediction model. The parameters of the prediction model for the mechanical properties of the light material part are optimized based on the back propagation neural network and the genetic algorithm, so that a more accurate prediction model can be obtained, the application range of the prediction model is enlarged, and the reliability of a prediction result is improved.

Description

Technical Field

[0001] The present application belongs to the technical field of performance detection of lightweight material components, and in particular, relates to a method, device and storage medium for predicting the mechanical properties of lightweight material components. Background Art

[0002] With the increasing demand for lightweight components in the fields of aerospace, vehicle manufacturing, etc., lightweight materials (such as high-strength steel, magnesium-aluminum alloys, composite materials, engineering plastics, pore-filling foams, etc.) and composite connection technologies (such as spot welding, gluing, riveting, etc.) are widely used in the production of components. Components made of one lightweight material and components made of multiple lightweight materials connected together are prone to deformation and fracture in the event of a collision.

[0003] In the related technology, a prediction model that best fits the test is usually obtained based on the material's test data and parameter reverse calibration method, and the prediction model is then used to predict material deformation and fracture. This solution relies on test data samples, has a small scope of application, and the reliability of the predicted results is poor. Summary of the invention

[0004] The present application provides a method, device and storage medium for predicting the mechanical properties of lightweight material components, thereby at least to a certain extent increasing the scope of application of the prediction model for the mechanical properties of lightweight material components and improving the reliability of the prediction results.

[0005] Other features and advantages of the present application will become apparent from the following detailed description, or may be learned in part by the practice of the present application.

[0006] According to a first aspect of an embodiment of the present application, a method for predicting mechanical properties of a lightweight material component is provided, comprising:

[0007] Obtaining a first parameter of a hardening model and a second parameter of a failure model corresponding to the lightweight material;

[0008] The first parameter and the second parameter are used as parameters of an initial prediction model, and the parameters of the initial prediction model are optimized based on a back propagation neural network and a genetic algorithm to obtain a third parameter;

[0009] Updating the parameters of the initial prediction model to the third parameters to obtain a target prediction model;

[0010] The target prediction model is used to predict the plastic deformation index and failure strain of lightweight material components.

[0011] In some embodiments, the step of obtaining a first parameter of a hardening model and a second parameter of a failure model corresponding to the lightweight material includes:

[0012] Performing a tensile test and a simulation on a sample of the lightweight material, and correspondingly obtaining a load-displacement curve of the sample and a first simulation result;

[0013] According to the load-displacement curve and the first simulation result, a first parameter of the hardening model of the sample and a second parameter of the failure model of each sample combination are determined.

[0014] In some embodiments, the hardening model includes a saturation characteristic sub-model and an unsaturation characteristic sub-model, and the first parameter is a weight of the saturation characteristic sub-model or the unsaturation characteristic sub-model.

[0015] In some embodiments, the second parameters include stress triaxiality, Lode parameter, and failure strain.

[0016] In some embodiments, the optimization of the parameters of the initial prediction model based on the back propagation neural network and the genetic algorithm to obtain the third parameter includes:

[0017] Inputting multiple groups of parameters of the initial prediction model into the back propagation neural network respectively to obtain the plastic deformation index prediction error and failure strain prediction error corresponding to each group of parameters;

[0018] The genetic algorithm is used to optimize the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error to obtain the third parameter.

[0019] In some embodiments, the method for predicting mechanical properties of a lightweight material component further includes:

[0020] The number of hidden layer nodes of the back propagation neural network is determined according to the sum of the first parameter and the second parameter.

[0021] In some embodiments, the method for predicting mechanical properties of a lightweight material component further includes:

[0022] Perform a tensile simulation of the sample and an axial compression-feed simulation of the thin-walled tube according to the third parameter to obtain a second simulation result;

[0023] The target prediction model is verified according to the second simulation result.

[0024] According to a second aspect of an embodiment of the present application, a device for predicting the mechanical properties of lightweight material components is provided, comprising a processor and a memory, wherein the memory stores computer program instructions that can be executed by the processor, and when the processor executes the computer program instructions, the steps of the method described in any one of the first aspects above are implemented.

[0025] According to a third aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor is prompted to implement the steps of the method described in any one of the first aspects above.

[0026] According to a fourth aspect of an embodiment of the present application, a computer program product is provided, including a computer program. When the computer program product is executed by a processor, it prompts the processor to implement the steps of the method described in any one of the first aspects above.

[0027] In the present application, the first parameter of the hardening model and the second parameter of the failure model corresponding to the lightweight material are obtained; the first parameter and the second parameter are used as the parameters of the initial prediction model, and the parameters of the initial prediction model are optimized based on the back propagation neural network and the genetic algorithm to obtain the third parameter; the parameters of the initial prediction model are updated to the third parameter to obtain the target prediction model; the plastic deformation index and failure strain of the lightweight material component are predicted using the target prediction model. By optimizing the parameters of the prediction model of the mechanical properties of the lightweight material component based on the back propagation neural network and the genetic algorithm, a more accurate prediction model can be obtained, which increases the scope of application of the prediction model and improves the reliability of the prediction results.

[0028] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] The drawings herein are incorporated into the specification and constitute a part of the specification, showing embodiments consistent with the present application, and together with the specification, are used to explain the principles of the present application. Obviously, the drawings described below are only some embodiments of the present application, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0030] Figure 1 A schematic flow chart of a method for predicting mechanical properties of a lightweight material component in one embodiment is shown;

[0031] Figure 2 A schematic diagram of a load-displacement curve in one embodiment is shown;

[0032] Figure 3 A schematic diagram of a saturated feature sub-model and a non-saturated feature sub-model in one embodiment is shown;

[0033] Figure 4 A schematic diagram of the principle of a back propagation neural network in one embodiment is shown;

[0034] Figure 5 A schematic diagram of an axial crushing force curve of a metal thin-walled tube in one embodiment is shown;

[0035] Figure 6 A block diagram of a device for predicting mechanical properties of a lightweight material component in one embodiment is shown;

[0036] Figure 7 A schematic structural diagram of a device for predicting the mechanical properties of a lightweight material component in one embodiment is shown. DETAILED DESCRIPTION

[0037] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0038] In addition, described feature, structure or characteristic can be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present application. However, those skilled in the art will appreciate that the technical scheme of the present application can be put into practice without one or more of the specific details, or other methods, components, devices, steps, etc. can be adopted. In other cases, known methods, devices, realizations or operations are not shown or described in detail to avoid blurring the various aspects of the application.

[0039] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0040] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0041] It should also be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the objects used in this way can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those shown or described.

[0042] Figure 1 FIG. 1 is a flow chart showing a method for predicting the mechanical properties of a lightweight material component in one embodiment. Figure 1 As shown, a method for predicting the mechanical properties of a lightweight material component is provided, which may include the following steps 101 to 104.

[0043] In step 101, a first parameter of a hardening model and a second parameter of a failure model corresponding to a lightweight material are obtained.

[0044] Among them, lightweight materials refer to materials with low density and small weight, which can be high-strength steel, magnesium-aluminum alloy, composite materials, engineering plastics, pore-filling foam, etc. Lightweight material components composed of lightweight materials are widely used in the fields of vehicles, aerospace, ships, etc.

[0045] In some embodiments, a tensile test and simulation can be performed on a sample of a lightweight material to obtain a load-displacement curve and a first simulation result of the sample; based on the load-displacement curve and the first simulation result, a first parameter of a hardening model of the sample and a second parameter of a failure model of each sample combination are determined.

[0046] Among them, specimens of lightweight materials can be divided into many types, such as arc notch type, shear notch type, triangular notch type, etc.

[0047] Figure 2 FIG. 2 shows a schematic diagram of a load displacement curve in one embodiment. Figure 2 As shown, in some embodiments, a load-displacement curve can be obtained by performing a tensile test on a sample of a lightweight material. After the load-displacement curve is converted into a stress-strain curve, relevant parameters of the hardening model and the failure model can be determined according to the stress-strain curve.

[0048] In some embodiments, the hardening model includes a saturated characteristic sub-model and a non-saturated characteristic sub-model, and the first parameter is a weight of the saturated characteristic sub-model or the non-saturated characteristic sub-model.

[0049] The hardening model is a mixed hardening model σ=(1-α)σ A +ασ B For example, where σ A and σ B There are two sub-models, which usually show "saturation" and "unsaturation" characteristics respectively. The weight α is between [0 and 1]. Through the linear combination of weight α, it can meet the actual usage requirements. Figure 3 FIG. 2 shows a schematic diagram of a saturated feature sub-model and a non-saturated feature sub-model in one embodiment. Figure 3 The sub-model A(σ A ) and sub-model B(σ B) The fitting situation in the small strain range is basically the same, and there are differences in the extended large strain range. The 0° experimental point is the curve of the experimental plastic hardening stage. The embodiment of the present application uses two sub-models to fit and extend the curve, and the weight α is used as the first parameter.

[0050] In some embodiments, the second parameter includes stress triaxiality, Lode parameter, and failure strain.

[0051] The failure model is ε i =f(η i ,L i ), i∈[1,m] as an example, where m is the type of sample, ε i is the failure strain of the i-th type of specimen, η i is the stress triaxiality of the i-th type of specimen, L i is the Lode parameter of the i-th type of specimen. The stress-strain curve can reflect the fracture failure under different stress triaxiality and Lode parameters, and the stress triaxiality, Lode parameter and corresponding failure strain can be extracted from the curve.

[0052] It should be noted that due to the limitations of the test data, in order to provide sufficient training samples for the back propagation neural network, the embodiments of the present application can also extract the stress triaxiality, Lode parameters and failure strain of each sample during the entire deformation process by means of simulation. The displacement at which the sample loses its bearing capacity in the test is determined as the fracture displacement, and the same displacement moment in the corresponding simulation is the fracture moment. The maximum value of the equivalent plastic strain at the fracture moment is extracted as the simulated equivalent failure strain, and the stress triaxiality and Lode parameters of the entire process from plastic deformation to fracture are extracted. The corresponding average stress triaxiality and average Lode parameters are obtained by integration and averaging, and then the stress triaxiality and Lode parameters corresponding to the simulated equivalent failure strain are obtained. According to the principle of permutations and combinations, it can be constructed samples (n is the type of the second parameter, m is the type of the specimen, m is greater than n, for example, when the second parameter is stress triaxiality, Lode parameter and failure strain, n=3, when n=3 and m=10, 120 samples are needed), each sample represents a specimen combination, and the second parameter corresponding to each sample is obtained by the equation system method, and the second parameter solution of the failure model under the sample is completed.

[0053] In step 102, the first parameter and the second parameter are used as parameters of an initial prediction model, and the parameters of the initial prediction model are optimized based on a back propagation neural network and a genetic algorithm to obtain a third parameter.

[0054] It is understandable that the initial prediction model is used to predict the plastic deformation index and failure strain of lightweight material components. However, since the parameters of the initial prediction model are obtained based on relevant data of lightweight materials, when predicting lightweight material components (such as components composed of one lightweight material, and components composed of multiple lightweight materials connected together), the prediction results may not be accurate. Therefore, it is necessary to optimize the parameters of the initial prediction model to obtain a target prediction model with higher prediction accuracy.

[0055] In some embodiments, multiple groups of parameters of the initial prediction model can be input into the back propagation neural network respectively to obtain the plastic deformation index prediction error and failure strain prediction error corresponding to each group of parameters; the genetic algorithm is used to optimize the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error to obtain the third parameter.

[0056] The sample size is The data sample set is divided into two parts: training samples and test samples. The training samples are used to train the parameters of the initial prediction model based on the back propagation neural network, while the test samples are used to test the accuracy of the trained prediction model. The various parameters of the back propagation neural network are initialized, including the number of nodes in the input layer, hidden layer and output layer, the activation function, the output function, the number of training times, the learning rate, the setting of the expected error parameters, and the initialization of the weight threshold; the parameters of the initial prediction model are trained based on the training samples and the back propagation neural network until the training error or the number of iterations reaches the set target value; the accuracy of the trained prediction model is assessed based on the test samples. If the test error meets the requirements, the test passes, otherwise, the prediction model needs to be retrained.

[0057] In some embodiments, the number of hidden layer nodes of the back-propagation neural network can be determined based on the sum of the first parameter and the second parameter.

[0058] Figure 4 FIG. 2 shows a schematic diagram of the principle of a back propagation neural network in one embodiment. Figure 4 As shown, the back propagation neural network adopts a single hidden layer structure. The number of hidden layer nodes z can be obtained according to the following formula:

[0059]

[0060] Among them, x is the number of input layer nodes, that is, the number of first parameters (weight α) and second parameters (b1, b2....b n ); y is the number of output layer nodes; w is an integer between 1 and 10. Set the number of hidden layer nodes to By calculating the relative error of the test samples output when the number of hidden layer nodes is different, a more reasonable number of hidden layer nodes can be determined.

[0061] It can be understood that each sample (i.e., sample combination) corresponds to a set of parameters, which consists of a first parameter and a second parameter. The first parameter of the sample can be randomly selected from all weights α. Multiple sets of parameters of the initial prediction model are input into the back propagation neural network respectively. The back propagation neural network will output the failure strain prediction error δ corresponding to each set of parameters by comparing the test failure strain of the sample in the tensile test with the failure strain output by the initial prediction model. i By comparing the crushing force curve of the thin-walled tube corresponding to the sample in the quasi-static axial crushing test (such as Figure 5 The experimental plastic deformation indicators (such as peak crushing force and absorbed energy) extracted from the initial prediction model and the plastic deformation indicators output by the initial prediction model are used to output the plastic deformation indicator prediction error φ corresponding to each set of parameters.

[0062] It should be noted that thin-walled tubes are an application form of lightweight material components. In order to facilitate testing and simulation, the embodiment of the present application uses the experimental plastic deformation index obtained by the quasi-static axial compression test of thin-walled tubes to determine the prediction error of the plastic deformation index, so that the prediction model can predict the plastic deformation index of lightweight material components.

[0063] When using genetic algorithm to optimize the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error, the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error can be used as individual fitness values, the crossover probability is selected as 0.8, the mutation probability is selected as 0.2, and the range of each parameter is set as follows:

[0064]

[0065] Among them, each b i The boundary value of is given by the empirical value under the specified failure model.

[0066] In the above embodiment, the size of the sample set is related to the type of the sample and the type of the second parameter of the failure model. Each sample is a group of sample combinations, and the parameters corresponding to each sample include the second parameters obtained by solving a group of sample combinations, which ensures that the sample set covers all sample selection schemes, so that the optimization results can better meet the failure prediction of all types of samples. By applying the back propagation neural network and genetic algorithm to the parameter optimization of the prediction model, the problems of the small application range of the traditional parameter reverse calibration method, poor reliability of the calculation results, and poor prediction effect can be solved.

[0067] In step 103, the parameters of the initial prediction model are updated to third parameters to obtain a target prediction model.

[0068] In some embodiments, a tensile simulation of the sample and an axial compression-feed simulation of the thin-walled tube are performed according to the third parameter to obtain simulation results; and the target prediction model is verified according to the simulation results.

[0069] During the implementation process, the third parameter can be substituted into the material tensile simulation model and the thin-walled tube crushing simulation model to perform numerical simulations of the M-type specimen tensile and the quasi-static axial crushing of the thin-walled tube, respectively. The reliability of the optimization can be verified by comparing the simulation results with the prediction results of the target prediction model. Usually, the optimization result can be considered reliable if the prediction result error is within 5%.

[0070] In step 104, the plastic deformation index and failure strain of the lightweight material component are predicted using the target prediction model.

[0071] During the implementation process, the steps for determining the target prediction model can be referred to as follows:

[0072] The first step: According to the geometric dimensions of the lightweight material specimen (including length, width, gauge length, etc.) and the crush dimensions of the thin-walled tube corresponding to the lightweight material (including length, width, height, wall thickness, etc.), crush displacement, the crush force curve obtained by the crush test, and the load-displacement curve obtained by the tensile test, the test plastic deformation index and test failure strain are obtained.

[0073] Step 2: Determine the functional relationship between the hardening model and the failure model, and transform the first parameter of the hardening model (weight α) and the second parameter of the failure model (b1, b2, ..., b n ) as the input of the neural network to be solved and the subsequent back-propagation neural network.

[0074] Step 3: Establish an interface with Abaqus software for parametric modeling, automatically divide the grid according to the size of the first step, couple and initialize the first parameter and the second parameter, and solve the stress triaxiality, Lode parameter and failure strain of each specimen during the entire deformation process.

[0075] Step 4: Randomly extract a first parameter from the first parameters of the hardening model of each sample, combine the samples according to the type of the second parameter and the sample type, and calculate the second parameter corresponding to each sample to obtain a sample set. The sample set is input into the back propagation neural network to obtain the plastic deformation index prediction error and failure strain prediction error corresponding to each sample set output by the back propagation neural network.

[0076] Step 5: Use genetic algorithm to optimize the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error, and obtain the parameter with the minimum error as the third parameter.

[0077] Step 6: Substitute the third parameter into the material tensile simulation model and the thin-walled tube crushing simulation model, obtain the crushing force curve and the load-displacement curve by simulation calculation, extract the plastic deformation index and failure strain, and conduct an error assessment with the prediction results of the target prediction model. If the error is within 5%, it proves that the prediction results of the target prediction model are reliable.

[0078] The embodiment of the present application obtains the first parameter of the hardening model and the second parameter of the failure model corresponding to the lightweight material; uses the first parameter and the second parameter as the parameters of the initial prediction model, and optimizes the parameters of the initial prediction model based on the back propagation neural network and the genetic algorithm to obtain the third parameter; updates the parameters of the initial prediction model to the third parameter to obtain the target prediction model; and uses the target prediction model to predict the plastic deformation index and failure strain of the lightweight material component. By optimizing the parameters of the prediction model of the mechanical properties of the lightweight material component based on the back propagation neural network and the genetic algorithm, a more accurate prediction model can be obtained, which increases the scope of application of the prediction model and improves the reliability of the prediction results.

[0079] The following describes an apparatus embodiment of the present application, which can be used to execute the method for predicting the mechanical properties of a lightweight material component in the above-mentioned embodiment of the present application. For details not disclosed in the apparatus embodiment of the present application, please refer to the embodiment of the method for predicting the mechanical properties of a lightweight material component in the above-mentioned embodiment of the present application.

[0080] Figure 6 FIG. 1 is a block diagram of a device for predicting mechanical properties of a lightweight material component in one embodiment. Figure 6 As shown, the device for predicting the mechanical properties of a lightweight material component of an embodiment of the present application includes: a parameter acquisition module 601, used to obtain a first parameter of a hardening model and a second parameter of a failure model corresponding to the lightweight material; a parameter optimization module 602, used to use the first parameter and the second parameter as parameters of an initial prediction model, and optimize the parameters of the initial prediction model based on a back propagation neural network and a genetic algorithm to obtain a third parameter; a model determination module 603, used to update the parameters of the initial prediction model to the third parameter to obtain a target prediction model; a model prediction module 604, used to predict the plastic deformation index and failure strain of the lightweight material component using the target prediction model.

[0081] In some embodiments of the present application, based on the aforementioned scheme, the parameter acquisition module 601 is also used to perform tensile tests and simulations on the specimens of the lightweight material, and obtain the corresponding load-displacement curve and the first simulation result of the specimen; according to the load-displacement curve and the first simulation result, determine the first parameters of the hardening model of the specimen and the second parameters of the failure model of each specimen combination.

[0082] In some embodiments of the present application, based on the aforementioned solution, the hardening model includes a saturated characteristic sub-model and an unsaturated characteristic sub-model, and the first parameter is a weight of the saturated characteristic sub-model or the unsaturated characteristic sub-model.

[0083] In some embodiments of the present application, based on the aforementioned solution, the second parameter includes stress triaxiality, Lode parameter and failure strain.

[0084] In some embodiments of the present application, based on the aforementioned scheme, the parameter optimization module 602 is also used to input multiple groups of parameters of the initial prediction model into the back propagation neural network respectively, to obtain the plastic deformation index prediction error and failure strain prediction error corresponding to each group of parameters; and to use the genetic algorithm to optimize the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error to obtain the third parameter.

[0085] In some embodiments of the present application, based on the aforementioned solution, the parameter optimization module 602 is further used to determine the number of hidden layer nodes of the back propagation neural network according to the sum of the first parameter and the second parameter.

[0086] In some embodiments of the present application, based on the aforementioned scheme, the model determination module 603 is also used to perform tensile simulation of the sample and axial compression feeding simulation of the thin-walled tube according to the third parameter to obtain a second simulation result; and verify the target prediction model according to the second simulation result.

[0087] Based on the same inventive concept, the present application embodiment also provides a prediction device for the mechanical properties of a lightweight material component, referring to Figure 7 , showing a schematic diagram of the structure of a device for predicting the mechanical properties of a lightweight material component in an embodiment of the present application, wherein the device for predicting the mechanical properties of a lightweight material component comprises one or more memories 704, one or more processors 702, and at least one computer program (computer program instruction) stored in the memory 704 and executable on the processor 702, and the processor 702 implements the method described above when executing the computer program.

[0088] Among them, Figure 7In the embodiment of the present invention, a bus architecture (represented by bus 700) is shown, which may include any number of interconnected buses and bridges, and bus 700 links various circuits including one or more processors represented by processor 702 and memory represented by memory 704. Bus 700 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and are therefore not further described herein. Bus interface 705 provides an interface between bus 700 and receiver 701 and transmitter 703. Receiver 701 and transmitter 703 may be the same element, namely a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 702 is responsible for managing bus 700 and general processing, while memory 704 may be used to store data used by processor 702 when performing operations.

[0089] Based on the same inventive concept, an embodiment of the present application provides a computer-readable storage medium, in which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor is prompted to implement the steps of the method as described above.

[0090] Based on the same inventive concept, an embodiment of the present application provides a computer program product, including a computer program. When the computer program product is executed by a processor, it prompts the processor to implement the steps of the method as described above.

[0091] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on a computer-readable medium or transmitted via a computer-readable medium. Other examples and implementations are within the scope and spirit of the present application and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hard wiring, or a combination of any of these. In addition, each functional unit may be integrated into a processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit.

[0092] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0093] The units described as separate components may or may not be physically separated, and the components of the control device may or may not be physical units, that is, they may be located in one place or distributed in multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0094] If the integrated unit is implemented in the form of 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 the present application can be essentially or partly or all or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or optical disk, etc., which can store computer program instructions.

[0095] The above description is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present application shall be included in the scope of the claims of the present application.

Claims

1. A method for predicting the mechanical properties of a lightweight material component, characterized in that: include: Obtaining a first parameter of a hardening model and a second parameter of a failure model corresponding to the lightweight material; The first parameter and the second parameter are used as parameters of an initial prediction model, and the parameters of the initial prediction model are optimized based on a back propagation neural network and a genetic algorithm to obtain a third parameter; Updating the parameters of the initial prediction model to the third parameters to obtain a target prediction model; The target prediction model is used to predict the plastic deformation index and failure strain of lightweight material components.

2. The method for predicting the mechanical properties of a lightweight material component according to claim 1, characterized in that: The step of obtaining a first parameter of a hardening model and a second parameter of a failure model corresponding to the lightweight material includes: Performing a tensile test and a simulation on a sample of the lightweight material, and correspondingly obtaining a load-displacement curve of the sample and a first simulation result; According to the load-displacement curve and the first simulation result, a first parameter of the hardening model of the sample and a second parameter of the failure model of each sample combination are determined.

3. The method for predicting the mechanical properties of lightweight material components according to claim 2, characterized in that: The hardening model includes a saturated characteristic sub-model and an unsaturated characteristic sub-model, and the first parameter is a weight of the saturated characteristic sub-model or the unsaturated characteristic sub-model.

4. The method for predicting the mechanical properties of a lightweight material component according to claim 2, characterized in that: The second parameters include stress triaxiality, Lode parameter and failure strain.

5. The method for predicting the mechanical properties of a lightweight material component according to claim 2, characterized in that: The method of optimizing the parameters of the initial prediction model based on the back propagation neural network and the genetic algorithm to obtain the third parameters includes: Inputting multiple groups of parameters of the initial prediction model into the back propagation neural network respectively to obtain the plastic deformation index prediction error and failure strain prediction error corresponding to each group of parameters; The genetic algorithm is used to optimize the parameters corresponding to the minimum plastic deformation index prediction error and failure strain prediction error to obtain the third parameter.

6. The method for predicting the mechanical properties of a lightweight material component according to claim 5, characterized in that: Also includes: The number of hidden layer nodes of the back propagation neural network is determined according to the sum of the first parameter and the second parameter.

7. The method for predicting the mechanical properties of a lightweight material component according to claims 2 to 6, characterized in that: Also includes: Perform a tensile simulation of the sample and an axial compression-feed simulation of the thin-walled tube according to the third parameter to obtain a second simulation result; The target prediction model is verified according to the second simulation result.

8. A device for predicting the mechanical properties of a lightweight material component, comprising a processor and a memory, characterized in that: The memory stores computer program instructions that can be executed by the processor, and when the processor executes the computer program instructions, the steps of the method according to any one of claims 1 to 7 are implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, which, when executed by a processor, prompt the processor to implement the steps of the method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that When the computer program product is executed by a processor, the processor is prompted to implement the steps of the method according to any one of claims 1 to 7.

Images