Hyperbolic plate rebound prediction deep learning method based on constitutive relation

Abstract

The invention discloses a hyperbolic plate rebound prediction deep learning method based on a constitutive relationship, and the method comprises the steps: collecting hyperbolic plate rebound amount sample data of a multi-material formed plate test piece, and summarizing the hyperbolic plate rebound amount sample data of multiple materials to form a sample data set; constructing a deep convolutional neural network; adding a constitutive relation custom layer into the deep convolutional neural network; the deep convolutional neural network is trained according to the training data set, and a plate forming springback prediction model is obtained; and inputting the preprocessed rebound prediction input sample into the plate forming rebound prediction model, selecting a stress-strain curve corresponding to the rebound prediction input sample, adjusting the weight and offset parameters of the hyperbolic plate rebound prediction model according to the stress-strain curve of the material, and outputting predicted rebound data. The method can effectively solve the problem that an existing algorithm cannot consider material characteristics, and rebound prediction of various novel materials and composite material plates is achieved by embedding a large number of material constitutive models.

Description

Technical Field

[0001] The present invention belongs to the technical field of hyperbolic plate forming springback prediction, and specifically relates to a hyperbolic plate springback prediction deep learning method based on constitutive relationship. Background Art

[0002] At present, with the development of science and technology, the requirements for sheet metal forming in various industrial fields have also increased. At the same time, different structures have different requirements, and different materials can be used to meet different requirements. However, the material properties of many composite materials or other advanced materials currently used are relatively complex. Therefore, it is difficult to improve the accuracy of sheet metal forming when selecting materials to meet other requirements.

[0003] At present, the application of deep learning in the field of sheet metal forming for springback prediction has achieved good results. However, due to the complex constitutive relations of many composite materials, when the three-dimensional forming feature information extracted from these different materials is used for springback prediction, the prediction accuracy is reduced, which seriously affects the forming accuracy of the sheet metal. Summary of the invention

[0004] The purpose of the present invention is to provide a deep learning method for hyperbolic plate springback prediction based on constitutive relations, which can effectively overcome the problem that existing algorithms cannot consider material properties, and realize springback prediction of various new materials and composite materials by embedding a large number of material constitutive models.

[0005] In order to solve the above technical problems, the technical solution of the present invention is: a deep learning method for predicting the springback of a hyperbolic plate based on a constitutive relationship, comprising the following steps:

[0006] S1. Take various material formed sheet specimens for tensile tests to obtain tensile force displacement curves of corresponding materials, obtain stress strain curves of corresponding materials according to the tensile force displacement curves of corresponding materials, collect the springback of various material formed sheet specimens in forming experiments and numerical simulations, take the stress strain curves and springback of various materials as components of sample data, and aggregate the sample data of various materials to form a sample data set; the sample data at least includes the springback, stress strain curves, sheet thickness and forming variation of various materials; pre-process the sample data set to obtain a training data set;

[0007] S2. Based on convolutional neural network, a deep convolutional neural network is constructed for end-to-end springback prediction of sheet metal forming of multiple materials;

[0008] S3. Add a custom constitutive relationship layer to the deep convolutional neural network to map the constitutive relationship of the material, train the deep convolutional neural network based on the material constitutive relationship according to the training data set, and obtain a sheet metal forming springback prediction model;

[0009] S4. Input the preprocessed springback prediction input sample into the sheet metal forming springback prediction model, select the stress-strain curve corresponding to the springback prediction input sample, adjust the weight and bias parameters of the hyperbolic plate springback prediction model according to the stress-strain curve of the material, and output the predicted springback data.

[0010] The specific method of preprocessing in S1 is:

[0011] Set the hyperbolic plate plane as the xoy plane, the direction perpendicular to the plate surface as the z direction, discretize the xoy plane into an n×n grayscale image, and the grayscale value range is [0, n-1]. The maximum value of the plate thickness of the corresponding material corresponds to n-1, and the minimum value corresponds to 0. The plate thickness of the corresponding material is scaled, and the plate thickness data is written into the thickness file;

[0012] In RGB color mode, the maximum value of the forming variation and springback of the corresponding material corresponds to n-1, and the minimum value corresponds to 0. The forming variation of the corresponding material is scaled, and the forming variation data and springback data of the corresponding material are written into the input file and the output file respectively; wherein the thickness file, the input file and the output file are all stored in the same preset format;

[0013] The thickness file is converted into a grayscale image, and the input file and the output file are converted into RGB images as the input and output data sets; the input data set and the output data set are randomly divided into a training set, a validation set and a test set according to a preset ratio, and used as the training data set; wherein the input data includes the plate thickness and the forming change, and the output data includes the springback.

[0014] The structure of the deep convolutional neural network in S2 is: including convolution layer, activation layer, maximum pooling layer, and transposed convolution layer; among them, the convolution layer is used for linear product summation, the activation layer is a corrected linear activation function, the maximum pooling layer is used for nonlinear downsampling, and the transposed convolution layer is used for upsampling.

[0015] S3 is specifically as follows: adding material tags to the RGB images in the training data set, adding a custom constitutive relationship layer after the deep convolutional neural network completes the first pooling, identifying the stress-strain curve of the corresponding material in the custom constitutive relationship layer through the material tag, and performing element-by-element multiplication, and the output is passed to the subsequent convolutional layer and pooling layer, and after each round of training according to the training data set, the weight of the custom constitutive relationship layer is updated according to the training accuracy feedback, and after the preset rounds of training, the optimal weight of the custom constitutive relationship layer and the bias parameters of the sheet metal forming springback prediction model are determined to obtain the sheet metal forming springback prediction model.

[0016] In S4, the preprocessed springback prediction input sample is input, and the corresponding stress-strain curve is selected according to its material. The weight and bias parameters of the sheet metal forming springback prediction model are adjusted according to the stress-strain curve of the material. The sheet metal forming springback prediction model outputs an RGB image with a pixel size of n×n to characterize the springback amount of the springback prediction input sample. The RGB image is reverse mapped according to the preprocessing method, and the RGB value of the RGB image is converted into an actual springback value and stored in a preset format, and it is used as the predicted springback data.

[0017] n is 256.

[0018] The preset ratio of the training set, validation set, and test set is 8:1:1.

[0019] The deep convolutional neural network is set to 42 layers.

[0020] The thickness file, input file, output file and actual springback value are all stored in .xlsx file format.

[0021] A computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of any of the above methods are implemented.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The present invention applies deep learning to the field of hyperbolic plate forming by embedding a material curve deep convolutional neural network algorithm. It is suitable for the processing prediction of more composite materials and advanced material hull plates. The difference from the existing deep learning is that it can simultaneously predict the springback of multiple materials and more complex material plates; the deep convolutional neural network algorithm is used to predict the springback of the plate, and the influence of the material constitutive relationship on the forming and springback can be considered. By matching the shape input features of the plate and the material features of the plate with corresponding labels, the shape features of different materials and the corresponding material features are multiplied element by element, and the material features are brought into the subsequent training. The influence of the complex constitutive relationship of composite materials and advanced materials on the processing and forming of hull plates is fully considered. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of a flow chart of an embodiment of the present invention;

[0025] Figure 2 Schematic diagram of the overall framework of a deep convolutional neural network in an embodiment of the present invention;

[0026] Figure 3 It is a schematic diagram of the process of defining a custom layer of a constitutive relationship in an embodiment of the present invention;

[0027] Figure 4 A schematic diagram of the working process of the springback prediction model for sheet metal forming in an embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram of the shape data processing flow in an embodiment of the present invention;

[0029] Figure 6 Schematic diagram of the matching process of the springback prediction input sample and the corresponding stress-strain curve in an embodiment of the present invention. DETAILED DESCRIPTION

[0030] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0031] See also Figure 1-Figure 6 , the technical solution of the present invention is:

[0032] A deep learning method for predicting springback of a hyperbolic plate based on constitutive relations comprises the following steps:

[0033] S1. Take various material formed sheet specimens for tensile test to obtain the force displacement curve of the corresponding material, obtain the stress strain curve of the corresponding material according to the force displacement curve of the corresponding material, and then collect the springback of various material formed sheet specimens in the sheet forming test and numerical simulation, take the stress strain curve and springback of various materials as the components of sample data, and aggregate the sample data of various materials to form a sample data set; the sample data at least includes the springback, stress strain curve, sheet thickness and forming change of various material sheets; pre-process the sample data set to obtain a training data set;

[0034] S2. Based on convolutional neural network, a deep convolutional neural network is constructed for end-to-end springback prediction of sheet metal forming of multiple materials;

[0035] S3. Add a custom constitutive relationship layer to the deep convolutional neural network to map the constitutive relationship of the material, train the deep convolutional neural network based on the material constitutive relationship according to the training data set, and obtain a sheet metal forming springback prediction model;

[0036] S4. Input the preprocessed springback prediction input sample into the sheet metal forming springback prediction model, select the stress-strain curve corresponding to the springback prediction input sample, adjust the weight and bias parameters of the hyperbolic plate springback prediction model according to the stress-strain curve of the material, and output the predicted springback data.

[0037] The specific method of preprocessing in S1 is:

[0038] The hyperbolic plate plane is set as the xoy plane, the direction perpendicular to the plate surface is the z direction, the xoy plane is discretized into a 256×256 grayscale image, the grayscale value range is [0,255], the maximum value of the plate thickness of the corresponding material corresponds to 255, and the minimum value corresponds to 0. The plate thickness of the corresponding material is scaled, and the plate thickness data is written into the thickness file;

[0039] In RGB color mode, the maximum value of the forming variation and springback of the corresponding material corresponds to 255, and the minimum value corresponds to 0. The forming variation of the corresponding material is scaled, and the forming variation data and springback data of the corresponding material are written into the input file and the output file respectively; wherein, the thickness file, the input file and the output file are all stored in the same preset format;

[0040] The thickness file is converted into a grayscale image, and the input file and the output file are converted into RGB images as the input and output data sets; the input data set and the output data set are randomly divided into a training set, a validation set, and a test set in a ratio of 8:1:1, and used as the training data set; the input data includes the plate thickness and the forming change, and the output data includes the springback.

[0041] The structure of the deep convolutional neural network in S2 is: including convolutional layer, activation layer, maximum pooling layer, and transposed convolutional layer; among them, the convolutional layer is used for linear product summation, the activation layer is a corrected linear activation function, the maximum pooling layer is used for nonlinear downsampling, and the transposed convolutional layer is used for upsampling. The deep convolutional neural network is set to 42 layers.

[0042] S3 is specifically as follows: adding material tags to the RGB images in the training data set, adding a custom constitutive relationship layer after the deep convolutional neural network completes the first pooling, identifying the stress-strain curve of the corresponding material in the custom constitutive relationship layer through the material tag, and performing element-by-element multiplication, and the output is passed to the subsequent convolutional layer and pooling layer, and after each round of training according to the training data set, the weight of the custom constitutive relationship layer is updated according to the training accuracy feedback, and after the preset rounds of training, the optimal weight of the custom constitutive relationship layer and the bias parameters of the sheet metal forming springback prediction model are determined to obtain the sheet metal forming springback prediction model.

[0043] In S4, the preprocessed springback prediction input sample is input, and the corresponding stress-strain curve is selected according to its material. The weight and bias parameters of the sheet metal forming springback prediction model are adjusted according to the stress-strain curve of the material. The weight and bias parameters of the springback prediction model change with the stress-strain curves of different materials. The sheet metal forming springback prediction model outputs an RGB image with a pixel size of 256×256 to characterize the springback amount of the springback prediction input sample. The RGB image is reverse mapped according to the preprocessing method, and the RGB value of the RGB image is converted into the actual springback value and stored in the .xlsx file format, and it is used as the predicted springback data.

[0044] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A deep learning method for predicting springback of a hyperbolic plate based on constitutive relations, characterized in that: The following steps are involved: S1. Take various material formed sheet specimens for tensile tests to obtain tensile force displacement curves of corresponding materials, obtain stress strain curves of corresponding materials according to the tensile force displacement curves of corresponding materials, collect the springback of various material formed sheet specimens in forming experiments and numerical simulations, take the stress strain curves and springback of various materials as components of sample data, and aggregate the sample data of various materials into a sample data set; The sample data at least includes the springback amount, stress-strain curve, plate thickness and forming variation of various material plates; the sample data set is preprocessed to obtain a training data set; S2. Based on convolutional neural network, a deep convolutional neural network is constructed for end-to-end springback prediction of sheet metal forming of multiple materials; S3. Add a custom constitutive relationship layer to the deep convolutional neural network to map the constitutive relationship of the material, train the deep convolutional neural network based on the material constitutive relationship according to the training data set, and obtain a sheet metal forming springback prediction model; S4. Input the preprocessed springback prediction input sample into the sheet metal forming springback prediction model, select the stress-strain curve corresponding to the springback prediction input sample, adjust the weight and bias parameters of the hyperbolic plate springback prediction model according to the stress-strain curve of the material, and output the predicted springback data.

2. According to the deep learning method for predicting springback of a hyperbolic plate based on constitutive relations of claim 1, it is characterized in that: The specific method of preprocessing in S1 is: Set the hyperbolic plate plane as the xoy plane, the direction perpendicular to the plate surface as the z direction, discretize the xoy plane into an n×n grayscale image, and the grayscale value range is [0, n-1]. The maximum value of the plate thickness of the corresponding material corresponds to n-1, and the minimum value corresponds to 0. The plate thickness of the corresponding material is scaled, and the plate thickness data is written into the thickness file; In RGB color mode, the maximum value of the forming variation and springback of the corresponding material corresponds to n-1, and the minimum value corresponds to 0. The forming variation of the corresponding material is scaled, and the forming variation data and springback data of the corresponding material are written into the input file and the output file respectively; wherein, the thickness file, the input file and the output file are all stored in the same preset format; The thickness file is converted into a grayscale image, and the input file and the output file are converted into RGB images as the input and output data sets; the input data set and the output data set are randomly divided into a training set, a validation set and a test set according to a preset ratio, and used as the training data set; wherein the input data includes the plate thickness and the forming change, and the output data includes the springback.

3. A deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 2, characterized in that: The structure of the deep convolutional neural network in S2 is: including convolution layer, activation layer, maximum pooling layer, and transposed convolution layer; among them, the convolution layer is used for linear product summation, the activation layer is a corrected linear activation function, the maximum pooling layer is used for nonlinear downsampling, and the transposed convolution layer is used for upsampling.

4. A deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 3, characterized in that: S3 is specifically as follows: adding material tags to the RGB images in the training data set, adding a custom constitutive relationship layer after the deep convolutional neural network completes the first pooling, identifying the stress-strain curve of the corresponding material in the custom constitutive relationship layer through the material tag, and performing element-by-element multiplication, and the output is passed to the subsequent convolutional layer and pooling layer, and after each round of training according to the training data set, the weight of the custom constitutive relationship layer is updated according to the training accuracy feedback, and after the preset rounds of training, the optimal weight of the custom constitutive relationship layer and the bias parameters of the sheet metal forming springback prediction model are determined to obtain the sheet metal forming springback prediction model.

5. A deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 4, characterized in that: In S4, the preprocessed springback prediction input sample is input, and the corresponding stress-strain curve is selected according to its material. The weight and bias parameters of the sheet metal forming springback prediction model are adjusted according to the stress-strain curve of the material. The sheet metal forming springback prediction model outputs an RGB image with a pixel size of n×n to characterize the springback amount of the springback prediction input sample. The RGB image is reverse mapped according to the preprocessing method, and the RGB value of the RGB image is converted into an actual springback value and stored in a preset format, and it is used as the predicted springback data.

6. A deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 5, characterized in that: n is 256.

7. The deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 2 is characterized in that: The preset ratio of the training set, validation set, and test set is 8:1:

1.

8. The deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 3 is characterized in that: The deep convolutional neural network is set to 42 layers.

9. The deep learning method for predicting springback of a hyperbolic plate based on constitutive relations according to claim 5, characterized in that: The thickness file, input file, output file and actual springback value are all stored in .xlsx file format.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images