finite element method
PulseAugur coverage of finite element method — every cluster mentioning finite element method across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
New PI-GNN offers improved stress localization modeling
Researchers have developed a novel variational physics-informed graph neural network (PI-GNN) designed to more accurately model stress localization in heterogeneous solids. Unlike traditional physics-informed neural net…
-
AI diffusion model guides engineering design via text prompts
Researchers have developed a novel method for engineering design by merging a training-free diffusion model with topology optimization. This approach allows engineers to specify design intent through natural language pr…
-
PINNs struggle with noisy data compared to traditional methods, study finds
A new research paper investigates the effectiveness of Physics-Informed Neural Networks (PINNs) when dealing with noisy data in inverse problems. The study found that while PINNs may require less specialized knowledge, …
-
New mesh-free neural network models brittle fracture with high accuracy
Researchers have developed a novel mesh-free method using a single neural network to model brittle fracture, eliminating the need for explicit crack tracking. This approach employs a multiresolution feature encoding bas…
-
AI University framework enhances engineering education with LLM learning assistants
Researchers have developed "AI University" (AI-U), a framework designed to create AI-powered learning assistants tailored for engineering courses. This system utilizes a fine-tuned large language model (LLM) combined wi…
-
New framework boosts AI prediction accuracy with analytical priors
Researchers have developed an analytical-prior learning framework designed to enhance data efficiency in predicting sound-reduction frequencies for Helmholtz resonators. This approach leverages a low-cost analytical mod…
-
Neural ODEs forecast power transformer thermal behavior
Researchers have developed a novel framework using Neural Ordinary Differential Equations (Neural ODEs) to accurately model and forecast the thermal behavior of power transformers. This physics-aware approach integrates…
-
ProPINN architecture tackles propagation failures in physics-informed neural networks
Researchers have introduced ProPINN, a novel architecture designed to address propagation failures in physics-informed neural networks (PINNs). These failures occur when supervision signals from initial or boundary cond…
-
New method enables label-free training for finite-element surrogate models
A new research paper proposes a label-free training objective for finite-element surrogate models, utilizing discrete energy minimization. This method eliminates the need for reference solutions, directly using the asse…
-
New neural operator frameworks tackle complex partial differential equations · 2 papers
Two new research papers introduce novel neural operator frameworks for solving partial differential equations (PDEs). The first, FB-C2CNet, utilizes fixed bases to encode and decode function coefficients, reducing train…
-
New convex neural energy elements enable reusable finite-element analysis
Researchers have developed a new method for creating reusable, geometry-parameterized neural elements for finite element analysis. This approach addresses structural failures in previous methods by ensuring that the ass…
-
New PIKS method offers universal physics-informed kernel learning
Researchers have introduced Physics-Informed Kernel methodS (PIKS), a novel approach to physics-informed machine learning that aims to overcome the limitations of existing methods. Unlike physics-informed neural network…
-
New FF-PINN model offers faster, more accurate analysis for geomaterials
Researchers have developed a Fourier Feature Physics-Informed Neural Network (FF-PINN) to address limitations in analyzing elasto-plastic problems in geotechnical engineering. Traditional Finite Element Methods (FEM) ar…
-
Physics-informed neural networks model wave propagation in bimaterial systems
Researchers have developed a novel framework utilizing physics-informed neural networks (PINNs) to model elastodynamic wave propagation in bimaterial systems. This approach embeds physical laws directly into the neural …
-
New benchmark SoftVTBench evaluates robotic manipulation safety
Researchers have introduced SoftVTBench, a new benchmark designed to evaluate robotic manipulation of deformable objects, focusing on both task completion and physical safety. This benchmark, built using NVIDIA Isaac Si…
-
New Physics-aware Transformer Reconstructs Fusion Divertor Temperature Fields
Researchers have developed a Physics-aware Neural Operator Transformer (PNOT) to reconstruct the temperature field of tungsten monoblock divertors in fusion devices. This method aims to overcome the computational expens…
-
New transfer learning method enhances AI for lithium-ion battery state estimation
Researchers have developed a transfer learning framework for physics-informed neural networks (PINNs) to improve state estimation in lithium-ion batteries. This approach addresses the challenge of training PINNs from sc…
-
New neural operator MR-GVNO accelerates plate response prediction
Researchers have developed MR-GVNO, a novel geometry-aware variational neural operator designed to accelerate response predictions for Mindlin-Reissner plates on irregular domains. This method utilizes boundary point cl…
-
New Method Approximates Whittle-Matern Fields on Discretized Manifolds
Researchers have developed a new method for approximating Whittle-Matern fields using discrete Gauss Markov Random Fields (GMRFs) on discretized Riemannian manifolds. This approach offers a universal approximation schem…
-
PINNOCHIO framework enhances surgical simulation with physics-informed neural networks
Researchers have developed PINNOCHIO, a new framework using Physics-Informed Neural Networks (PINNs) to simulate facial soft-tissue deformation for orthognathic surgery planning. This approach addresses the accuracy-eff…