Abstract: A system and method generates a mesh representing a flow domain around a surface of a physical object. The system receives an initial mesh representing the flow domain. The initial mesh has a corner vertex corresponding to a corner grid point of a grid in a parameterized space. The system updates the corner grid point to obtain an updated grid in the parameterized space according to angles formed by boundary edges and a border edge of the initial mesh. The system determines an updated mesh from the initial mesh according to the updated grid to reduce a difference between the angles formed by the boundary edges and the border edge. The system generates the mesh representing the flow domain for simulating a physical characteristic of the physical object.

Abstract: Methods and apparatuses for simulating an electromechanical object are described. An electromagnetic solver is invoked for an electromagnetic model of the object to generate an electromagnetic loss of the object an operating temperature. A temperature dependent electromagnetic loss of the object is determined based on the electromagnetic loss at the operating temperature. The temperature dependent electromagnetic loss varies according to a temperature of the object. A thermal solver is invoked for a thermal model of the object to estimate a temperature of the object based on the temperature dependent electromagnetic loss of the object and the operating temperature.

Abstract: Coupling of physics solutions (e.g., a first physics solution and a second physics solution) in a multi-physics simulation that includes non-conformal spaces can use extrapolation to determine data values for unmapped nodes by computing extrapolated values based on both mapped values from mapped nodes and data for unmapped nodes. The extrapolated values for an unmapped node are computed using a set of equations that include mapped values from nearby mapped nodes and data for nearby unmapped nodes. The methods and systems can provide improved computational efficiency by reducing the amount of memory required for the computations.

Abstract: A computer-implemented method for adaptive mesh refinement includes generating a first mesh for a first layer of a stacked model representing a 3D integrated circuit, the first mesh including a first grid for the first layer, performing a simulation for the stacked model using the first mesh to obtain first results, calculating an interpolated value for a particular grid node according to the first results, the particular grid node corresponding to a node of the first grid in the grid hierarchy. The method further includes generating a second mesh for the layer of the stacked model, the second mesh including a second grid for the first layer, performing a simulation for the stacked model using the second mesh to obtain a simulated value of the particular grid node, comparing the interpolated and simulated value, and generating a refined mesh for the first layer including a refined grid.

Abstract: Computational fluid dynamics simulations of objects with many similar but distinct flow boundary regions on a larger surface can use an approach that creates geometries and meshes for the larger surface (e.g., a first mesh) independently of geometries and meshes for the distinct flow boundary regions (second mesh). The second mesh can be imprinted onto the first mesh prior to a CFD simulation, and if a revision to the design of the regions is desired, a revised mesh for the regions (based on a revised geometry for the regions) can be imprinted on the first mesh without having to redo the first mesh. Thus, revisions of the design can avoid recreating the first mesh.

Abstract: Systems and methods are provided for A computer-implemented method for simulating behavior of an electrical circuit across a range of frequencies from 0 to fmax. A base function is determined that outputs magnitudes across the range of frequencies. A correction function is determined that outputs magnitudes across the range of frequencies. The base function is combined with the correction function to generate a circuit behavior model that provides magnitudes across the range of frequencies. Behavior of the electrical circuit is simulated using the circuit behavior model.

Abstract: Systems and methods for automatically training a machine learning model are described herein. An example method includes performing a set of computational simulations; and assembling a data set associated with the set of computational simulations. The data set includes data associated with at least one simulation result for at least one computational simulation in the set of computational simulations. The method also includes training a machine learning model with the data set. At least one feature and at least one target for the machine learning model are part of the data set.

Abstract: This disclosure provides modelling techniques for accurately and reliably analyzing printed parts. More specifically, a model with a powder representation may be used for the thermal analysis, the powder representation may then be removed from the model for performing the structural analysis. Although the powder representation is removed from the model for the structural analysis, the structural analysis still factors in the thermal effects of the powder on the part determined by the thermal analysis. The computer-based techniques of the current disclosure improve the functioning of a computer system as compared to conventional approaches by facilitating analyses (e.g., thermal, mechanical, design, and heat treatment analyses) that are more accurate, more efficient (e.g., faster, smaller memory requirements, etcetera), and/or have a reduced processing burden versus conventional approaches.

Abstract: Computer-implemented systems and methods are described herein for determining additive manufacturing parameters in a simulation. Input data regarding a product to be generated via additive manufacturing and a beam diameter are received. Based on the input data, a characteristic dimension is determined. The beam diameter is normalized based on the characteristic dimension. Additive manufacturing parameters, such as penetration depth and absorptivity, are determined based on the normalized beam diameter and experimentally-determined trends. The manufacturing of the product is then simulated according to the additive manufacturing parameters.

Abstract: Provided is an improved method, system, and computer program product to implement simulation for photonic devices. A composite, multi-domain simulation model is disclosed, with connected domain-specific representations that allow the use of the most relevant simulator technology for a given domain. The model has external connection points either expressed as actual ports or virtual ones, embodied by simulator API calls in the model.

Abstract: Methods and systems for improved simulation of thermal characterization and thermal modeling of devices, such as smart phones, are described. In one embodiment, a method can characterize center, edge, and corner thermal decay behavior at different locations on a simulated IC. For each location, near and far field thermal effects are captured at the same time. A simulation system can generate a steady state thermal decay curve for each selected location that shows how the temperature changes with distance to a heat source. The system can then use a set of location dependent thermal decay curves to compute, based on an inputted power profile for the IC, a steady state thermal profile of the IC.

Abstract: In one embodiment, use of virtual thermal cameras in simulations systems can rapidly increase the deployment of vehicles with autonomous driving systems that include thermal cameras as a part of the sensor systems that detect people, other cars, etc. This use can be achieved by calibrating virtual thermal cameras using a virtual object or scenes with predetermined temperatures of objects (humans) in the scenes. The virtually calibrated thermal camera can then be used in virtual simulations of driving scenarios to generate learning data, such as virtual thermal images, to test autonomous driving systems and autonomous driving algorithms. The calibration can use a temperature conversion process, and the generation of the learning data can also use this temperature conversion process that includes the addition of noise, represented by temperature values.

Abstract: A method and apparatus of a device of predicting a heat map for a die model is described. In an exemplary embodiment, the device receives a die model for representing the die and a boundary condition applicable to the die, the die model including dimensions of the die. In addition, the device may include scaling the dimensions of the die model and the boundary condition, wherein the dimensions of the die model are scaled to a target die size, the scaled dimensions including a scaled die thickness. Furthermore, each trained die model including a die surface of the target die size, different trained die models having different die thickness, the plurality of trained die models trained under a plurality of boundary conditions, each trained die model having a plurality of decay curves associated with the plurality of boundary conditions. The method may additionally include determining a relationship between the scaled boundary condition and the plurality of boundary conditions.

Abstract: A software tool used to assess and/or simulate an integrated circuit (“IC”) layout employs heterogeneous resistance extraction and/or meshing, to combine advantages of a relatively compact parasitics file size and robustness for effective simulation and electrical performance analysis. In one embodiment, filters are automatically applied (or are manually specified by a user) to select structures, layers, geographic regions, nets or other areas of a design which will receive additional extraction and/or meshing scrutiny, with the remainder of the design being subjected to default practices. For example, software can be scripted (or rule driven) to perform 1D meshing for in-layer electrical pathways while providing 2D, 3D, or other complex meshing for vias, contact pads, device instances, selected nets and/or other structures. Techniques can be applied for reassessing impact of structural redesign, and reports and/or visualization can be generated, each enhanced by the use of heterogeneous meshing.

Abstract: A three-dimensional model (i.e., MCAD model) for additively manufacturing a physical object with a laser is received in a computer system. The laser is controlled or driven by a set of instructions. The instructions specify scan properties for scanning each of multiple deposit layers during the additive manufacturing. Two-dimensional (2D) maps representing the scan properties for the deposit layer are generated. Each 2D map includes values in grid points representing the deposit layer. Each 2D map corresponds to a respective scan property for the deposit layer. A trained machine learning model is invoked to predict a physical behavior of a layer of the physical object based on the 2D maps as an input. The input further includes a state map showing a state of an immediately prior deposit layer. The layer of the physical object corresponds to the deposit layer.

Abstract: Techniques for security vulnerability assessment of security-sensitive circuit designs are described. A placement of security-sensitive components of a design may be based on constraints related to how far apart a relevant set of security-sensitive components are allowed without consuming too much power and how to optimize the placement to minimize electromagnetic side-channel leakage or other security vulnerabilities. In one embodiment, a method may receive data that includes a representation of a design of an IC and may identify security-sensitive components of the design from the data. The method may determine a placement for the design based on constraints on a level of security vulnerabilities of the security-sensitive components and may perform a power simulation for the design based on the placement. The method may generate an assessment of the level of security vulnerabilities of the security-sensitive components based on the power simulation to adjust the placement for the design.

Abstract: A CAD method for capacitance extraction includes decomposing a semiconductor structure model representing a three-dimensional, into virtual layers. Each virtual layer has polygons corresponding to conductors in the semiconductor structure model. The semiconductor structure model or includes a geometric point. The method includes creating spatial indexes respectively for the virtual layers. Each spatial index indicates respectively for each spatial index cell of a corresponding virtual layer at least one candidate conductor of the corresponding virtual layer. The method includes creating optimal transition squares for the geometric point respectively for one or more consecutive virtual layers according to spatial indexes of the one or more consecutive virtual layers. The geometric point is located in one of the one or more consecutive virtual layers. The optimal transition squares are constrained by one or more candidate conductors of the one or more consecutive virtual layers.

Abstract: A model representing a physical object in a shape optimization according to a design objective, and a control point for altering a shape of the physical object are received. The shape is defined by a set of nodes in the model. Sibling models are generated from the model according to a perturbation scheme. The control point is perturbed with respective perturbed values for the sibling models. Each sibling model contains nodal location changes for the nodes. The nodal location changes are determined based on a respective shape function formulated according to a respective perturbed value at the control point and one or more simulated physical behaviors of the model. The model is updated to have an optimal value for the control point. The optimal value is identified from a relationship according to the design objective. The relationship correlates a physical characteristic of the sibling models to the respective perturbed values.

Abstract: A generative machine learning model, such as a convolutional neural network (CNN), can be trained with solutions from a topology optimization solver for a solution for a topology of a set of structures so that the generative machine learning model can generate a plurality of alternative designs for a structure that are alternative topology optimizations (for the structure) for a set of initial setup parameters. The generative model when being trained includes a generative network and a discriminator network. The generative model can be trained using outputs from a CNN autoencoder for densities and a CNN autoencoder for strain energies.

Abstract: Simulations of products during the design of the products can use solvers that are based on trained neural networks, and these solvers can provide results about the design of the product that can predict performance, failures, fatigue and other potential problems with the design. The neural network can include a generative neural network that is trained with a discretized version of a partial differential equation (PDE) that provides a model of the product in the simulation, and this discretized version acts as a discriminator that trains the neural network to provide solutions to the PDE.