# US 12,118,645 B2 · Generating microstructures for materials based on machine learning models > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 12,118,645 B2 - Document ID: US12118645B2 - Status: Granted - Kind: B2 - Application number: 17/553,524 - Filed: 2021-12-16 - Published: 2023-06-22 - Granted: 2024-10-15 - Earliest claimed: 2021-12-16 - Assignees: Volkswagen Aktiengesellschaft - Applicants: Volkswagen Aktiengesellschaft - Inventors: Wesley Teskey - Web page: https://patentgrep.com/patent/US12118645B2 ## CPC - G06T 11/00: Two-dimensional [2D] image generation - G06N 3/045: Combinations of networks - G06N 3/0455: Auto-encoder networks; Encoder-decoder networks - G06N 3/047: Probabilistic or stochastic networks - G06N 3/0475: Generative networks - G06N 3/094: Adversarial learning - G06T 7/0004: {Industrial image inspection} - G06T 2207/20084: Artificial neural networks [ANN] - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract In one embodiment, a method is provided. The method includes determining a set of spheres for a volume of a material. The volume of the material comprises the set of spheres and additional materials. The sizes of the set of spheres are based on a Gaussian mixture model (GMM). The method also includes determining a set of locations for the set of spheres within the volume of the material. The method further includes generating a set of images of the volume of the material based on a first generative adversarial network and a second generative adversarial network. The set of images depict a microstructure of the volume of material. ## Claims ### Claim 1 A method, comprising: determining a set of spheres for a volume of a material, wherein: the volume of the material comprises the set of spheres and additional materials; and sizes of the set of spheres are based on a Gaussian mixture model (GMM); determining a set of locations for the set of spheres within the volume of the material; generating a set of images of the volume of the material based on a first generative adversarial network and a second generative adversarial network, wherein the set of images depict a microstructure of the volume of material and wherein a first subset of set of images is generated by the first generative adversarial network and a second subset of the set of images is generated by the second adversarial network using the first subset as an input; and determining, by a machine learning model, one or more material properties of the volume of the material, wherein the set of images are provided as an input to the machine learning model. ### Claim 2 (depends on 1) The method of claim 1, wherein determining the set of locations for the set of spheres comprises: selecting a set of initial locations for the spheres within the volume of the material, wherein the set of initial locations are selected non-deterministically. ### Claim 3 (depends on 2) The method of claim 2, wherein determining the set of locations for the set of spheres comprises: displacing one or more spheres from the one or more initial locations to one or more different locations based on one or more overlaps between the one or more spheres. ### Claim 4 (depends on 2) The method of claim 2, wherein determining the set of locations for the set of spheres comprises: displacing one or more spheres from the one or more initial locations towards a central portion of the volume of the material. ### Claim 5 (depends on 2) The method of claim 2, wherein determining the set of locations for the set of spheres comprises: disposing a sphere from the set of spheres between two or more other spheres from the set of spheres. ### Claim 6 (depends on 1) The method of claim 1, wherein each image in the second subset of images is generated based on a previously generated image. ### Claim 7 (depends on 6) The method of claim 6, wherein locations of the additional material in a current image are based on the previous locations of the additional material in the previously generated image. ### Claim 8 (depends on 1) The method of claim 1, further comprising: providing the set of images to a material generation system, wherein the material generation system determines one or more material properties of the volume of the material. ### Claim 9 (depends on 1) The method of claim 1, wherein: the additional materials comprise one or more of binder material and electrolyte material; and the set of spheres comprise an active material. ### Claim 10 (depends on 1) The method of claim 1, wherein: the first generative adversarial network comprises a first conditional generative adversarial network; and the second generative adversarial network comprises a second conditional generative adversarial network. ### Claim 11 (depends on 1) The method of claim 1, wherein each image of the set of images depicts a cross section of the volume of the material at a respective depth of the volume of the material. ### Claim 12 An apparatus, comprising: a memory configured to store data; and a processing device coupled to the memory, the processing device configured to: determine a set of spheres for a volume of a material, wherein: the volume of the material comprises the set of spheres and additional materials; and sizes of the set of spheres are based on a Gaussian mixture model (GMM); determine a set of locations for the set of spheres within the volume of the material; generate a set of images of the volume of the material based on a first generative adversarial network and a second generative adversarial network, wherein the set of images depict a microstructure of the volume of material and wherein a first subset of set of images is generated by the first generative adversarial network and a second subset of the set of images is generated by the second adversarial network using the first subset as an input; and determine, based on a machine learning model, one or more material properties of the volume of the material, wherein the set of images are provided as an input to the machine learning model. ### Claim 13 (depends on 12) The apparatus of claim 12, wherein to determine the set of locations for the set of spheres the processing device is to: select a set of initial locations for the spheres within the volume of the material, wherein the set of initial locations are selected non-deterministically. ### Claim 14 (depends on 13) The apparatus of claim 13, wherein to determine the set of locations for the set of spheres the processing device is to: displace one or more spheres from the one or more initial locations to one or more different locations based on one or more overlaps between the one or more spheres. ### Claim 15 (depends on 13) The apparatus of claim 13, wherein to determine the set of locations for the set of spheres the processing device is to: displace one or more spheres from the one or more initial locations towards a central portion of the volume of the material. ### Claim 16 (depends on 13) The apparatus of claim 13, wherein to determine the set of locations for the set of spheres the processing device is to: dispose a sphere from the set of spheres between two or more other spheres from the set of spheres. ### Claim 17 (depends on 12) The apparatus of claim 12, wherein each image in the second subset of images is generated based on a previously generated image. ### Claim 18 (depends on 12) The apparatus of claim 12, wherein: the first generative adversarial network comprises a first conditional generative adversarial network; and the second generative adversarial network comprises a second conditional generative adversarial network. ### Claim 19 (depends on 12) The apparatus of claim 12, wherein each image of the set of images depicts a cross section of the volume of the material at a respective depth of the volume of the material. ### Claim 20 A non-transitory computer-readable storage medium including instructions that, when executed by a processing device, cause the processing device to perform operations comprising: determining a set of spheres for a volume of a material, wherein: the volume of the material comprises the set of spheres and additional materials; and sizes of the set of spheres are based on a Gaussian mixture model (GMM); determining a set of locations for the set of spheres within the volume of the material; generating a set of images of the volume of the material based on a first generative adversarial network and a second generative adversarial network, wherein the set of images depict a microstructure of the volume of material and wherein a first subset of set of images is generated by the first generative adversarial network and a second subset of the set of images is generated by the second adversarial network using the first subset as an input; and determining, based on a machine learning model, one or more material properties of the volume of the material, wherein the set of images are provided as an input to the machine learning model.