Abstract: Embodiments provide technologies for improving structural properties of a finite element structure. A computing system builds, from a plurality of graph representations of a corresponding finite element mesh, a graph neural network (GNN). The computing system may use the GNN or other methods to traverse the graph representation. In traversing the graph representation, the computing system evaluates, for each grouping of nodes in the representation, one or more properties associated with the structural element. The computing system determines whether the structural element satisfies a specified condition. Upon so determining, the computing system modifies the grouping of nodes and outputs the resulting graph representation.

Abstract: Devices, systems, and methods related to design of structural portions of a vehicular component can include estimating noise within a 3D noise tensor via a 3D denoising-diffusion model and removing noise from the 3D noise tensor based on the estimated noise. Structural portions can be generated including a 3D cellular structure. Design of the vehicular component can be adapted based on the generated structural portion.

Abstract: Devices, systems, and methods accelerated, variable load finite element analysis for vehicle design can provide a graphical representation of a finite element mesh and enter the graphical representation as an input to a recurrent neural network (RNN). Such solutions can develop a time series mesh (TSM) as an output from the RNN. Design of a component of the vehicle can be adapted based on the output.

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.

Abstract: A method, apparatus and system are provided to generate and analyze material structures. A first machine learning model may generate material structures and a second machine learning model may determine stress values and strain values for the generated material structures. The material structures are generated based on material parameters.

Abstract: Approaches, techniques, and mechanisms are disclosed for assessing battery states of batteries. According to one embodiment, raw sensor data are collected from a battery module. The battery module includes multiple battery cells. Input battery features are extracted from the raw sensor data collected from the battery module. The input battery features are used to update node states of a GNN. The GNN include multiple GNN nodes each of which representing a respective battery cell in the multiple battery cells. Estimation of one or more battery state of health (SoH) indicators is generated based at least in part on individual output states of individual GNN nodes in the multiple GNN nodes. The individual output states of individual GNN nodes in the multiple GNN nodes are determined based at least in part on the updated node states of the GNN.

Abstract: In one embodiment, a method is provided. The method includes obtaining a set of graphs for a set of material structures. Each graph of the set of graphs is associated with a material structure of the set of material structures. The method also includes determining first sets of stress values and first sets of strain values for the set of graphs based on a neural network. The method further includes obtaining second sets of stress value and second sets of strain values for a subset of the set of material structures. The method further includes updating the neural network based on the first sets of stress values, the first sets of strain values, the second sets of stress values, and the second sets of strain values.

Abstract: Approaches, techniques, and mechanisms are disclosed for assessing battery states of batteries. According to one embodiment, raw sensor data are collected from a battery module. The battery module includes multiple battery cells. Input battery features are extracted from the raw sensor data collected from the battery module. The input battery features are used to update node states of a GNN. The GNN include multiple GNN nodes each of which representing a respective battery cell in the multiple battery cells. Estimation of one or more battery state of health (SoH) indicators is generated based at least in part on individual output states of individual GNN nodes in the multiple GNN nodes. The individual output states of individual GNN nodes in the multiple GNN nodes are determined based at least in part on the updated node states of the GNN.

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.