SuperMeshing: A New Deep Learning Architecture for Increasing the Mesh Density of Metal Forming Stress Field with Attention Mechanism and Perceptual Features

• URL: http://arxiv.org/abs/2104.09276v1
• Date: Fri, 12 Mar 2021 06:02:30 GMT
• Title: SuperMeshing: A New Deep Learning Architecture for Increasing the Mesh Density of Metal Forming Stress Field with Attention Mechanism and Perceptual Features
• Authors: Qingfeng Xu, Zhenguo Nie, Handing Xu, Haosu Zhou, Xinjun Liu
• Abstract summary: We propose a new data-driven mesh-density boost model named SuperMeshingNet. It strengthens the advantages of finite element analysis (FEA) with low mesh-density as inputs to the deep learning model. Compared to the baseline that applied the linear method, SuperMeshingNet achieves a prominent reduction in the mean squared error (MSE) and mean absolute error (MAE) on test data.
• Score: 1.0660502023086995
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In stress field analysis, the finite element analysis is a crucial approach, in which the mesh-density has a significant impact on the results. High mesh density usually contributes authentic to simulation results but costs more computing resources, leading to curtailing efficiency during the design process. To eliminate this drawback, we propose a new data-driven mesh-density boost model named SuperMeshingNet that strengthens the advantages of finite element analysis (FEA) with low mesh-density as inputs to the deep learning model, which consisting of Res-UNet architecture, to acquire high-density stress field instantaneously, shortening computing time and cost automatically. Moreover, the attention mechanism and the perceptual features are utilized, enhancing the performance of SuperMeshingNet. Compared to the baseline that applied the linear interpolation method, SuperMeshingNet achieves a prominent reduction in the mean squared error (MSE) and mean absolute error (MAE) on test data, which contains prior unseen cases. Based on the data set of metal forming, the comparable experiments are proceeded to demonstrate the high quality and superior precision of the reconstructed results generated by our model. The well-trained model can successfully show more excellent performance than the baseline and other methods on the multiple scaled mesh-density, including $2\times$, $4\times$, and $8\times$. With the refined result owning broaden scaling of mesh density and high precision, the FEA process can be accelerated with seldom cost on computation resources. We publicly share our work with full detail of implementation at https://github.com/zhenguonie/2021_SuperMeshing_2D_Metal_Forming

Related papers

• Towards Faster and More Compact Foundation Models for Molecular Property Prediction [44.64301507940171]
  Joint Multi-domain Pre-training (JMP) foundation model has demonstrated strong performance across various downstream tasks. Despite JMP's advantages, fine-tuning it on molecular datasets ranging from small-scale to large-scale requires considerable time and computational resources. Our study provides insights for developing lighter, faster, and more scalable foundation models for molecular and materials discovery.
  arXiv  Detail & Related papers  (2025-04-28T07:41:03Z)
• PCA-RAG: Principal Component Analysis for Efficient Retrieval-Augmented Generation [0.0]
  High-dimensional language model embeddings can present scalability challenges in terms of storage and latency. This paper investigates the use of Principal Component Analysis (PCA) to reduce embedding dimensionality. We show that PCA-based compression offers a viable balance between retrieval fidelity and resource efficiency.
  arXiv  Detail & Related papers  (2025-04-11T09:38:12Z)
• DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs [70.91804882618243]
  This paper proposes DSMoE, a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks. We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge. Experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches.
  arXiv  Detail & Related papers  (2025-02-18T02:37:26Z)
• Towards Scalable and Deep Graph Neural Networks via Noise Masking [59.058558158296265]
  Graph Neural Networks (GNNs) have achieved remarkable success in many graph mining tasks. scaling them to large graphs is challenging due to the high computational and storage costs. We present random walk with noise masking (RMask), a plug-and-play module compatible with the existing model-simplification works.
  arXiv  Detail & Related papers  (2024-12-19T07:48:14Z)
• Task-Oriented Real-time Visual Inference for IoVT Systems: A Co-design Framework of Neural Networks and Edge Deployment [61.20689382879937]
  Task-oriented edge computing addresses this by shifting data analysis to the edge. Existing methods struggle to balance high model performance with low resource consumption. We propose a novel co-design framework to optimize neural network architecture.
  arXiv  Detail & Related papers  (2024-10-29T19:02:54Z)
• Enhancing Microgrid Performance Prediction with Attention-based Deep Learning Models [0.0]
  This research aims to address microgrid systems' operational challenges, characterized by power oscillations that contribute to grid instability. An integrated strategy is proposed, leveraging the strengths of convolutional and Gated Recurrent Unit (GRU) layers. The framework is anchored by a Multi-Layer Perceptron (MLP) model, which is tasked with comprehensive load forecasting.
  arXiv  Detail & Related papers  (2024-07-20T21:24:11Z)
• Iterative Sizing Field Prediction for Adaptive Mesh Generation From Expert Demonstrations [49.173541207550485]
  Adaptive Meshing By Expert Reconstruction (AMBER) is an imitation learning problem. AMBER combines a graph neural network with an online data acquisition scheme to predict the projected sizing field of an expert mesh. We experimentally validate AMBER on 2D meshes and 3D meshes provided by a human expert, closely matching the provided demonstrations and outperforming a single-step CNN baseline.
  arXiv  Detail & Related papers  (2024-06-20T10:01:22Z)
• Reinforcement Learning as a Parsimonious Alternative to Prediction Cascades: A Case Study on Image Segmentation [6.576180048533476]
  PaSeR (Parsimonious with Reinforcement Learning) is a non-cascading, cost-aware learning pipeline. We show that PaSeR achieves better accuracy while minimizing computational cost relative to cascaded models. We introduce a new metric IoU/GigaFlop to evaluate the balance between cost and performance.
  arXiv  Detail & Related papers  (2024-02-19T01:17:52Z)
• Conformal Predictions Enhanced Expert-guided Meshing with Graph Neural Networks [8.736819316856748]
  This paper presents a machine learning-based scheme that utilize Graph Neural Networks (GNN) and expert guidance to automatically generate CFD meshes for aircraft models. We introduce a new 3D segmentation algorithm that outperforms two state-of-the-art models, PointNet++ and PointMLP, for surface classification. We also present a novel approach to project predictions from 3D mesh segmentation models to CAD surfaces using the conformal predictions method.
  arXiv  Detail & Related papers  (2023-08-14T14:39:13Z)
• Proximal Symmetric Non-negative Latent Factor Analysis: A Novel Approach to Highly-Accurate Representation of Undirected Weighted Networks [2.1797442801107056]
  Undirected Weighted Network (UWN) is commonly found in big data-related applications. Existing models fail in either modeling its intrinsic symmetry or low-data density. Proximal Symmetric Nonnegative Latent-factor-analysis model is proposed.
  arXiv  Detail & Related papers  (2023-06-06T13:03:24Z)
• MultiScale MeshGraphNets [65.26373813797409]
  We propose two complementary approaches to improve the framework from MeshGraphNets. First, we demonstrate that it is possible to learn accurate surrogate dynamics of a high-resolution system on a much coarser mesh. Second, we introduce a hierarchical approach (MultiScale MeshGraphNets) which passes messages on two different resolutions.
  arXiv  Detail & Related papers  (2022-10-02T20:16:20Z)
• Efficient Micro-Structured Weight Unification and Pruning for Neural Network Compression [56.83861738731913]
  Deep Neural Network (DNN) models are essential for practical applications, especially for resource limited devices. Previous unstructured or structured weight pruning methods can hardly truly accelerate inference. We propose a generalized weight unification framework at a hardware compatible micro-structured level to achieve high amount of compression and acceleration.
  arXiv  Detail & Related papers  (2021-06-15T17:22:59Z)
• Extrapolation for Large-batch Training in Deep Learning [72.61259487233214]
  We show that a host of variations can be covered in a unified framework that we propose. We prove the convergence of this novel scheme and rigorously evaluate its empirical performance on ResNet, LSTM, and Transformer.
  arXiv  Detail & Related papers  (2020-06-10T08:22:41Z)
• Widening and Squeezing: Towards Accurate and Efficient QNNs [125.172220129257]
  Quantization neural networks (QNNs) are very attractive to the industry because their extremely cheap calculation and storage overhead, but their performance is still worse than that of networks with full-precision parameters. Most of existing methods aim to enhance performance of QNNs especially binary neural networks by exploiting more effective training techniques. We address this problem by projecting features in original full-precision networks to high-dimensional quantization features.
  arXiv  Detail & Related papers  (2020-02-03T04:11:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.