A multi-porous structure cross-domain reverse design method based on property constraint collaborative training

Abstract

The present application relates to a kind of porous structure cross-domain reverse design method based on property constraint collaborative training, mechanical property information and random noise are used as the input of reverse generation model after training, obtain encoding vector, reconstruct the porous structure data using the variational autoencoder model after training.This method directly carries out mechanical property prediction collaborative training to the feature vector distribution in encoding space in the training process of variational autoencoder, improves the feature distribution in encoding space, establishes more direct relationship between " encoding space→property ".By increasing the constraint of forward relationship, in the training process of reverse generation model, it is easier to establish " property→encoding space ", compared with ordinary reverse design scheme, the present application can improve the accuracy of prediction structure, support accurate porous structure cross-domain reverse design.The present application fully considers the characteristics of different distribution of cross-domain structure, realizes cross-domain reverse design with high accuracy.

Description

Technical Field

[0001] The present invention relates to the field of material structure design, and in particular to a cross-domain inverse design method for porous structures based on collaborative training of property constraints. Background Art

[0002] Porous structures are material structures that can be used in aerospace, medical and other fields. Designing a workpiece into a porous structure can reduce the mass of the workpiece as much as possible while retaining the required mechanical properties, such as strength and stiffness. In addition, through artificial design, porous materials can meet the mechanical properties required by people under specific conditions, such as specific buffering materials. Current porous materials are usually implemented using repeated porous units (generally cubic units). Researchers design the geometric structure within the unit to meet the requirements of mechanical properties. Common structures include rod structures, shell structures, spinodal structures, etc. Researchers change the structural design to achieve specific performance of the workpiece with the least amount of material. This type of design method is a forward design, and the direction of optimization by researchers is relatively blind.

[0003] Reverse design has also developed in recent years. Some solutions analyze stress distribution and adjust the dimensions of the workpiece at stress concentration points to accurately improve the mechanical properties of the workpiece. However, this type of optimization method relies on finite element analysis to obtain stress distribution, which is computationally intensive. Current neural network-based reverse design solutions still have some limitations:

[0004] (1) It mainly supports structural design within a specific domain, such as rod structure, shell structure, random spinodal structure, etc., and cannot generate structures that meet the requirements across domains, which greatly limits the diversity of reverse design;

[0005] (2) Some neural network inverse design schemes based on generative model structures (generative adversarial models, diffusion models, etc.) support cross-domain generation in terms of data format. However, since the distribution of cross-domain data is usually very different, such schemes often have poor results.

[0006] In summary, there is currently a lack of a cross-domain inverse design method for porous structures to solve or partially solve the aforementioned problems. Summary of the Invention

[0007] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and provide a cross-domain inverse design method for porous structures based on collaborative training of property constraints, so as to solve the problem that cross-domain inverse design cannot be realized or the effect is poor.

[0008] The purpose of the present invention can be achieved by the following technical solutions:

[0009] One aspect of the present invention provides a cross-domain inverse design method for porous structures based on collaborative training of property constraints. Mechanical property information and random noise are used as inputs of a trained inverse generative model to obtain an encoding vector. The trained variational autoencoder model is used to obtain reconstructed porous structure data. The training process of the inverse generative model and the variational autoencoder model includes the following steps:

[0010] Step S1, constructing cross-domain porous structure-mechanical property data pairs as training samples;

[0011] Step S2, using the porous structure data in the training sample as input to a variational autoencoder model to obtain an encoding vector, calculating a first property loss using an encoding space-mechanical property mapper, and implementing training of the variational autoencoder model based on the first property loss;

[0012] Step S3, using the noisy coding vector as the input of the inverse generation model to obtain the reconstructed coding vector, using the coding space-mechanical property mapper to calculate the second property loss, and realizing the training of the inverse generation model based on the second property loss.

[0013] As a preferred technical solution, in step S1, the cross-domain porous structure-mechanical property data pairs include porous structure-mechanical property data pairs corresponding to rod structure, shell structure and spinodal structure.

[0014] As an optimal technical solution, the training of the variational autoencoder model is achieved by combining the first property loss, porous structure reconstruction loss and coding space loss.

[0015] As a preferred technical solution, the calculation process of the porous structure reconstruction loss includes:

[0016] The porous structure data in the training sample is used as the input of the encoder in the variational autoencoder model to obtain an encoding vector, and the encoding vector is input into the decoder in the autoencoder model to obtain reconstructed porous structure data. The porous structure reconstruction loss is calculated based on the porous structure data in the training sample and the reconstructed porous structure data.

[0017] As a preferred technical solution, the calculation process of the coding space loss includes:

[0018] Calculate the loss between the encoding vector output by the variational autoencoder model and the preset standard distribution as the encoding space loss.

[0019] As a preferred technical solution, the process of calculating the primary property loss and the secondary property loss using the coding space-mechanical property mapper includes the following steps:

[0020] The encoding vector obtained by the variational autoencoder model or the inverse generation model is used as the input of the encoding space-mechanical property mapper to obtain the predicted mechanical property information. Based on the predicted mechanical property information and the mechanical property information in the training samples, the first property loss or the second property loss is calculated.

[0021] As an optimal technical solution, the second property loss and noise loss are combined to realize the training of the inverse generation model.

[0022] As a preferred technical solution, the noise loss calculation process includes the following steps:

[0023] The coding vector output by the variational autoencoder model is subjected to denoising to obtain the noisy coding vector, which is used as the input of the inverse generative model to obtain predicted noise information. A reconstructed coding vector is obtained based on the noise information, and the noise loss is calculated based on the predicted noise information and the actual noise.

[0024] Another aspect of the present invention provides an electronic device comprising: one or more processors and a memory, wherein the memory stores one or more programs, and the one or more programs include instructions for executing the aforementioned cross-domain inverse design method for porous structures based on collaborative training of property constraints.

[0025] Another aspect of the present invention provides a computer-readable storage medium comprising one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the aforementioned cross-domain inverse design method for porous structures based on collaborative training of property constraints.

[0026] Compared with the prior art, the present invention has at least one of the following beneficial effects:

[0027] (1) Fully considering the different characteristics of cross-domain structural distribution, achieving highly accurate cross-domain inverse design: During the training process of the variational autoencoder, the distribution of feature vectors in the encoding space is directly trained for mechanical property prediction, improving the distribution of encoding space features and establishing a more direct relationship between "encoding space → properties". By adding constraints on the positive relationship, it is easier to establish "properties → encoding space" during the training process of the inverse generative model. Compared with ordinary inverse design schemes, this invention can improve the accuracy of predicted structures and support accurate cross-domain inverse design of porous structures.

[0028] (2) Low computational resource and time cost burden, easy to promote and apply: Compared with the inverse design method that does not use property prediction collaborative training, the training process of the present invention only adds additional loss function calculation and a simple network structure (encoding space-mechanical property mapper), which hardly increases the time cost of training. The reasoning process (inverse design process) does not increase any computational or time cost. No additional loss is added, which is conducive to the implementation and use of the inverse design algorithm.

[0029] (3) Wide range of application scenarios: Compared with the inverse design method that does not use collaborative training of property prediction, the present invention adds a "coding space → property" mapper, so the model can be simultaneously applied to multi-task scenarios of forward property prediction and inverse design. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A six-layer diagram of a cross-domain inverse design method for porous structures based on collaborative training of property constraints in an embodiment;

[0031] Figure 2 This is an architecture diagram of the cross-domain reverse design of porous structures based on property constraint collaborative training in an embodiment;

[0032] Figure 3 is a schematic diagram of an electronic device in an embodiment;

[0033] Figure 4 This is an example diagram of the reverse design effect of the cross-domain reverse design of the porous structure based on property constraint collaborative training in the embodiment. DETAILED DESCRIPTION

[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0035] Example 1

[0036] To address the aforementioned issues with cross-domain reverse design in existing technologies, which are either unattainable or ineffective, this embodiment provides a cross-domain reverse design method for porous structures based on collaborative training of property constraints. This method relies on a variational autoencoder, an inverse generative model, and an encoding space-to-mechanical property mapper. In this embodiment, cross-domain refers to cross-structural domains, and common structures include rods, shells, and spinodals.

[0037] Preferably, the reverse generation model in this embodiment is implemented using an implicit diffusion model or a generative adversarial model.

[0038] See also Figure 2 ,The variational autoencoder includes an encoder and a decoder mapper, and the encoding space-mechanical property mapper can be composed of several layers of fully connected layers.

[0039] See also Figure 1 and Figure 4 ,This method mainly includes the training data preparation stage, ,variational autoencoder training stage, inverse generative model training stage and ,inverse design stage. Each stage will be described separately ,below.

[0040] Training data preparation stage: Construct a cross-domain porous structure-mechanical property data pair as a training sample set, which contains a total of N "porous structure data-property information" data pairs {Si, Pi} (i=0,1,...N) from k domains, where Si refers to the porous structure, expressed using a signed distance field, and the data format is a three-dimensional array of shape [r,r,r], where r represents the resolution. Pi refers to the mechanical properties, which can be stiffness, yield strength, stress-strain curve, etc. In this embodiment, Pi specifically refers to the stiffness matrix, which can characterize the linear relationship between stress and strain under small deformation, and its data format is a [6,6] matrix.

[0041] Variational Autoencoder Training Phase: The variational autoencoder consists of an encoder and a decoder. During training, the encoder maps the structure Si into the encoding space to obtain the encoding vector li, and the decoder restores the encoding vector li to the structure Si*. Three loss functions are constructed during the training phase: the reconstruction loss between the original structure Si and the reconstructed structure Si*, the loss function between the distribution of the encoding vector li and the standard Gaussian distribution, and the encoding space loss. By setting the first two loss functions, an encoder that satisfies the "structure → encoding space" mapping relationship and a decoder that satisfies the "encoding space → structure" mapping relationship can be obtained. At the same time, to establish a more direct connection between the encoding space and the properties, a third loss function is designed: the loss function between li and Pi—the property loss. The property loss is calculated by the encoding space-mechanical property mapper, which takes the encoding vector li as input and outputs the predicted property value Pi*. The loss function is the difference between Pi and Pi*.

[0042] Inverse generative model training phase: The input to the inverse generative model is the sample's encoding vector li, obtained by encoder mapping during the variational autoencoder training phase, the property condition Pi, and random Gaussian noise n. The output is the inversely generated encoding vector li*. The sample's encoding vector is first weighted and summed with random Gaussian noise to obtain a noisy encoding vector. The noisy encoding vector is then input into the inverse generative model, which predicts the noise information in the noise encoding and reconstructs the sample's encoding vector based on the noise value. This phase involves two losses: the first loss is the difference between the noise information predicted by the inverse generative model and the actual noise, and the second loss is the property loss of the reconstructed encoding vector. The second loss is calculated by the encoding space-mechanical property mapper, which takes the reconstructed li* as input and outputs the predicted property value Pi**. The loss function is the difference between Pi** and Pi.

[0043] Inverse design phase: The property condition information Pj and random Gaussian noise N are used as the input of the inverse generation model, and the inverse generation model reconstructs the encoding vector. The reconstructed encoding vector is input into the decoder trained in the variational autoencoder training phase to obtain the reconstructed porous structure. The inverse design process is as follows: Figure 4 As shown in Figure 3, this scheme can generate porous structures of different domains, including rod structures, shell structures, etc., according to a control condition (i.e., stiffness matrix).

[0044] Specifically, the inverse generative model is used to predict the noise information in the noise code, and the process of reconstructing the sample's code vector based on the noise value is as follows: during the noise addition process, the sample's code vector is weighted and summed with random Gaussian noise to obtain a noisy code vector (noise code vector = a * random Gaussian noise + b * code vector). The noisy code vector is input into the inverse generative model, which, after several inference iterations, predicts the random Gaussian noise added to the code vector. Given the known noise code vector and Gaussian noise, the initial sample code vector is obtained based on the weighted relationship (code vector = (noise code vector - a * random Gaussian noise) ÷ b), which is the reconstructed code vector.

[0045] Compared to training using the "structure → encoding space → structure" training mapping relationship, the encoding space usually has a more direct relationship with the structure. However, the relationship between structure and properties is relatively implicit, so this method usually requires a lot of computation to obtain. The connection between the "encoding space" and "properties" obtained in this way is also relatively implicit. At the same time, in the encoding space, structures within the same domain are usually similar, while structures across domains often have different distributions, which is not conducive to generating cross-domain structural encodings under the same property conditions.

[0046] Compared to the above approach, the variational autoencoder of this embodiment directly performs property prediction and collaborative training on the distribution of feature vectors in the encoding space during the training phase. This improves the distribution of features in the encoding space and establishes a more direct relationship between "encoding space → properties." By adding constraints on the forward relationship, establishing the "properties → encoding space" relationship is much easier during the inverse generative model training phase. Compared to conventional inverse design approaches, this method can improve the accuracy of predicted structures and support accurate cross-domain inverse design of porous structures.

[0047] Example 2

[0048] This embodiment provides Figure 3 The one shown corresponds to Figure 1 A schematic diagram of the electronic device, such as Figure 3 As mentioned above, at the hardware level, the electronic device includes a processor, an internal bus, a network interface, a memory and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to achieve the above Figure 1 The method of the cross-domain reverse design method of porous structures based on property constraint collaborative training. Of course, in addition to software implementation, the present invention does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0049] Example 3

[0050] This embodiment provides a computer-readable storage medium, including one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the cross-domain inverse design method for porous structures based on collaborative training of property constraints as described in Example 1.

[0051] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0052] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0053] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0054] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and such modifications or substitutions are intended to be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

Claims

1. A cross-domain inverse design method for porous structures based on property constraint collaborative training, characterized by: The mechanical property information and random noise are used as inputs of the trained inverse generative model to obtain an encoding vector, and the trained variational autoencoder model is used to obtain reconstructed porous structure data. The training process of the inverse generative model and the variational autoencoder model includes the following steps: Step S1, constructing cross-domain porous structure-mechanical property data pairs as training samples; Step S2, using the porous structure data in the training sample as input to a variational autoencoder model to obtain an encoding vector, calculating a first property loss using an encoding space-mechanical property mapper, and implementing training of the variational autoencoder model based on the first property loss; Step S3: Using the noisy code vector as the input of the inverse generative model to obtain a reconstructed code vector, calculating the second property loss using the code space-mechanical property mapper, and implementing the inverse generative model training based on the second property loss. In the step S1, the cross-domain porous structure-mechanical property data pairs include the porous structure-mechanical property data pairs corresponding to the rod structure, the shell structure and the spinodal structure. The process of calculating the primary property loss and the secondary property loss using the encoding space-mechanical property mapper includes the following steps: The encoding vector obtained by the variational autoencoder model or the inverse generation model is used as the input of the encoding space-mechanical property mapper to obtain the predicted mechanical property information. Based on the predicted mechanical property information and the mechanical property information in the training samples, the first property loss or the second property loss is calculated.

2. The cross-domain inverse design method for porous structures based on property constraint collaborative training according to claim 1 is characterized in that: The first property loss, the porous structure reconstruction loss and the coding space loss are combined to achieve the training of the variational autoencoder model.

3. The cross-domain inverse design method for porous structures based on property constraint collaborative training according to claim 2 is characterized in that: The calculation process of the porous structure reconstruction loss includes: The porous structure data in the training sample is used as the input of the encoder in the variational autoencoder model to obtain an encoding vector, and the encoding vector is input into the decoder in the autoencoder model to obtain reconstructed porous structure data. The porous structure reconstruction loss is calculated based on the porous structure data in the training sample and the reconstructed porous structure data.

4. The cross-domain inverse design method for porous structures based on property constraint collaborative training according to claim 2 is characterized in that: The calculation process of the coding space loss includes: Calculate the loss between the encoding vector output by the variational autoencoder model and the preset standard distribution as the encoding space loss.

5. The cross-domain inverse design method for porous structures based on property constraint collaborative training according to claim 1 is characterized in that: The second property loss and the noise loss are combined to realize the training of the reverse generation model.

6. The cross-domain inverse design method for porous structures based on property constraint collaborative training according to claim 5 is characterized in that: The calculation process of the noise loss includes the following steps: The coding vector output by the variational autoencoder model is subjected to denoising to obtain the noisy coding vector, which is used as the input of the inverse generative model to obtain predicted noise information. A reconstructed coding vector is obtained based on the noise information, and the noise loss is calculated based on the predicted noise information and the actual noise.

7. An electronic device, characterized in that: include: One or more processors and a memory, wherein the memory stores one or more programs, wherein the one or more programs include instructions for executing the cross-domain inverse design method for porous structures based on property constraint collaborative training as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that The invention comprises one or more programs for execution by one or more processors of an electronic device, wherein the one or more programs include instructions for executing the cross-domain inverse design method of porous structure based on property constraint collaborative training as described in any one of claims 1 to 6.

Images