公开(公告)号：US20220293270A1

公开(公告)日：2022-09-15

申请号：US17646061

申请日：2021-12-27

Applicant: DASSAULT SYSTEMES

Inventor： Aziliz COTTIN ,  Nicolas PECUCHET ,  Agathe GUILLOUX ,  Sandrine KATSAHIAN

IPC: G16H50/20 ,  G06N3/08 ,  G06N3/04

Abstract: A computer-implemented method for machine-learning a function configured, based on input covariates representing medical characteristics of a patient with respect to a multi-state model of an illness having states and transitions between the states, to output a distribution of transition-specific probabilities for each interval of a set of intervals, the set of intervals forming a subdivision of a follow-up period. The machine-learning method including obtaining a dataset of covariates and time-to-event data of a set of patients, and training the function based on the dataset. This forms an improved solution for determining accurate patient data with respect to a multi-state model of an illness.

公开(公告)号：US11442111B2

公开(公告)日：2022-09-13

申请号：US17184249

申请日：2021-02-24

Applicant: Dassault Systemes Americas Corp.

Inventor： Johan Carlsson ,  Kwan Skinner ,  Michael Doyle ,  Nick Reynolds ,  Lalitha Subramanian ,  Felix Hanke

IPC: G01R31/392 ,  G01R31/367 ,  H01M10/42 ,  H01M10/48

Abstract: An automated system and method to investigate degradation of cathode materials in batteries via atomistic simulations, and in particular by simulating the creation of atomistic defects in the cathode material, which occurs during charge cycling. A systematic procedure relates the degradation of battery performance metrics to underlying structural changes due to atomic rearrangements within the material, for example through density functional theory simulations. The performance metrics modeled with this approach include the Open Cell Voltage (OCV) as well as the discharge capacity curve.

公开(公告)号：US20220215510A1

公开(公告)日：2022-07-07

申请号：US17557330

申请日：2021-12-21

Applicant: DASSAULT SYSTEMES

Inventor： Andreas WEINMANN ,  Holger DAMMERTZ

IPC: G06T5/00 ,  G06T3/40 ,  G06T5/50

Abstract: A computer-implemented method for forming a dataset configured for learning a Convolutional Neural Network (CNN) architecture including an image feature extractor. It comprises providing pairs of images, each pair comprising a reference image and a respective denoised image. For each pair of images, the method provides the pair of images to a pre-trained CNN architecture similar to the one the formed dataset will be configured for. The method computes an error map representing a difference between a first normalized feature of the denoised image and a second normalized feature of the reference image, the first and second normalized features being the output of a same layer of the pre-trained CNN architecture and adds the respective denoised image and the error map to the dataset. This constitutes an improved solution with respect to forming a dataset for learning a CNN architecture to identify areas of degradation generated by a denoiser.

公开(公告)号：US11373015B2

公开(公告)日：2022-06-28

申请号：US16123866

申请日：2018-09-06

Applicant: Dassault Systemes SolidWorks Corporation

Inventor： Utpal Kumar ,  Satish Rao Kanjarkar ,  Sachin Darwatkar ,  Rajendrasing Pardeshi

IPC: G06F30/00 ,  G06F111/04 ,  G06F111/20

Abstract: The invention notably relates to a computer-implemented method for designing a 3D modeled object by interaction of a user with a feature-based CAD system, the 3D modeled object representing a mechanical structure. The method comprises creating structural member features, each structural member feature representing a respective structural member of the mechanical structure, and displaying to the user a graphical representation of the mechanical structure based on the structural member features. The method further comprises creating corner features, each corner feature representing a respective corner of the mechanical structure, the creation of the corner features being performed automatically by the system, the corner features being editable by the user. This provides improved ergonomics for structural design.

公开(公告)号：US20220199261A1

公开(公告)日：2022-06-23

申请号：US17131905

申请日：2020-12-23

Applicant: Dassault Systemes Simulia Corp.

Inventor： Jonathan T. Jilesen ,  Gregory M. Laskowski

IPC: G16H50/50 ,  G06F30/25

Abstract: Described are computer aided techniques to simulate a human respiratory event. The computer aided techniques access a model including a portion of a person's respiratory tract, which models the respiratory tract as a volumetric region, initiate a respiratory event into the volumetric regions, which respiratory event originates in the accessed model at a depth that is inside of the modeled respiratory tract, simulate movement of elements of the respiratory event within the volumetric region, with the elements representing particles of the respiratory event, at an inlet boundary condition representing an area of the model that is at the threshold depth inside the respiratory tract, and obtain from the simulation, a representation of a trajectory of particles of the respiratory event.

公开(公告)号：US11341296B2

公开(公告)日：2022-05-24

申请号：US16215070

申请日：2018-12-10

Applicant: Dassault Systemes

Inventor： Pierre Denis Voirin

IPC: G06T17/20 ,  G06F30/20 ,  G06F30/17 ,  G06T19/20 ,  G06F113/24 ,  G06F119/18

Abstract: A computer-implemented method computes an unfolded part of a modeled bended 3D object in a 3D scene of a computer-aided design system. The method a) provides the 3D object; b) selects a fixed portion (FP) of the 3D object; c) selects a mobile portion (MP) of the 3D object; d) determines a 1D interface (INT) forming an intersection between the fixed portion (FP) and the mobile portion; e) computes a transformed portion resulting from a linear transformation of the mobile portion (MP) according to an drawing direction (DD); f) trims the transformed portion in the vicinity of the 1D interface (INT), thereby forming a trimmed transformed portion (TTP); g) creates a fillet (FI) between the 1D interface (INT) and the trimmed transformed portion (TTP); and h) defines the unfolded part as an union of the fixed portion (FP), the trimmed transformed portion (TTP) and the created fillet (FI).

公开(公告)号：US11321605B2

公开(公告)日：2022-05-03

申请号：US15810840

申请日：2017-11-13

Applicant: Dassault Systemes SolidWorks Corporation

Inventor： Ameya Divekar ,  Makarand Apte ,  Shrikant Savant

IPC: G06F30/00 ,  G06N3/04 ,  G06N3/08 ,  G06T19/00 ,  G06F30/17 ,  G06F30/27 ,  G06N3/02 ,  G06F111/04 ,  G06F111/20

Abstract: Methods and systems identify frequently-used CAD components and apply machine learning techniques to predict mateable entities and corresponding mate types for those components to automatically add components to a CAD model. An example method includes accessing information regarding CAD model parts and related mate information stored in a computer database, and dividing parts into a plurality of clusters having parts with similar global shape signatures. In response to a new part being added, contextual signatures of entities of the new part are input into a mateability predictor neural network to determine a mateable entity of the new part. Input into a mate-type predictor neural network is (i) a contextual signature of the mateable entity and (ii) a contextual signature of an entity of another part of the CAD model to determine a mate type between the entities. A mate between the new part and the other part is automatically added based on the determined mate type.

公开(公告)号：US20220121819A1

公开(公告)日：2022-04-21

申请号：US17506475

申请日：2021-10-20

Applicant: DASSAULT SYSTEMES

Inventor： Xavier GREHANT ,  Benoit DUVAL ,  Samu TAMMINEN

IPC: G06F40/279 ,  G06F40/166

Abstract: A computer implemented method for comparing unsupervised embedding methods for a similarity based industrial component model requesting system including obtaining a text corpus relating to industrial component models and a list of testing words, modifying by altering some of the occurrences of each testing word, the modified text corpus containing, for each testing word, occurrences of a first version of each testing word, and occurrences of a second version of each testing word, running an unsupervised embedding method on the modified text corpus and obtaining vector representations, determining a scoring value, by comparing, for at least some of the testing words, the vector representations of the first version of these testing words, and the vector representations the second version of these testing words, running the obtaining, modifying with the text corpus and the list of testing words with another unsupervised embedding method and returning the respective scoring values.

公开(公告)号：US20220108055A1

公开(公告)日：2022-04-07

申请号：US17060264

申请日：2020-10-01

Applicant: Dassault Systemes Simulia Corp.

Inventor： Sivapalan Senthooran ,  Mohamed Ibrahim Eldesouky ,  Robert Powell ,  Ales Alajbegovic

IPC: G06F30/28

Abstract: Disclosed are computer implemented techniques for correcting for numerically generated pressure waves at an inlet of a simulation space. The techniques include receiving a model of a simulation space and applying an inlet pressure to an inlet of the simulation space. The applied inlet pressure generates fluctuating velocities that produce undesired, numerically-generated pressure waves. The numerically generated pressure waves are measured to establish a measured pressure history. The measured pressure history is subtracted from the applied inlet boundary pressure history to provide a set of boundary conditions. The process conducts a fluid simulation using the set of boundary conditions. The process repeats using a subsequent set of boundary conditions, until an iteration is reached where the measured pressures near the inlet are sufficiently small to compensate for undesired, numerically-generated pressure waves, and thereafter stores that subsequent set of boundary conditions to provide a corrected set of boundary conditions.

公开(公告)号：US20220101105A1

公开(公告)日：2022-03-31

申请号：US17486684

申请日：2021-09-27

Applicant: DASSAULT SYSTEMES ,  ECOLE POLYTECHNIQUE ,  CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE

Inventor： Mariem MEZGHANNI ,  Maks OVSJANIKOV ,  Malika BOULKENAFED

IPC: G06N3/04 ,  G06N3/08

Abstract: A computer-implemented method for training a deep-learning generative model configured to output 3D modeled objects each representing a mechanical part or an assembly of mechanical parts. The method comprises obtaining a dataset of 3D modeled objects and training the deep-learning generative model based on the dataset. The training includes minimization of a loss. The loss includes a term that penalizes, for each output respective 3D modeled object, one or more functional scores of the respective 3D modeled object. Each functional score measures an extent of non-respect of a respective functional descriptor among one or more functional descriptors, by the mechanical part or the assembly of mechanical parts. This forms an improved solution with respect to outputting 3D modeled objects each representing a mechanical part or an assembly of mechanical parts.