Results 81 to 90 of about 8,116 (238)

NanoSentry: An Edge‐Deployable Unified Neural Architecture for Multi‐Domain Industrial Prognostics and Health Management

Artificial Intelligence for Engineering, EarlyView.
NanoSentry, a sub‐50 KiB unified architecture for multi‐domain prognostics, matches or outperforms models 4.5× $4.5\times $ larger across turbofan, battery, and bearing benchmarks while reducing end‐of‐life prediction error up to 5.8× $5.8\times $. Physical STM32H7 deployment confirms sub‐3 ms inference and up to 26 KiB peak SRAM without quantisation ...
Naem Azam Chowdhury, Abu Nasek, Khan Prottoy, Md Hamid Borkot Tulla, Jafor Sadik Chowdhury   +4 more
wiley   +1 more source

Space Correlation Constrained Physics Informed Neural Network for Seismic Tomography

Journal of Geophysical Research: Machine Learning and Computation
Physics‐informed neural networks (PINNs) integrate physical constraints with neural architectures and leverage their nonlinear fitting capabilities to solve complex inverse problems.
Yonghao Wang, Zhuo Jia, Bowu Jiang, Wenkai Lu   +3 more
doaj   +1 more source

Large‐Scale Applications of Plasma Technologies: Progresses and Challenges

High Voltage, EarlyView.
ABSTRACT Plasma technologies convert electrical power into energetic electrons, ions, reactive species, radiation and heat, thereby breaking through the bottleneck of purely thermal or chemical routes and bringing impressive changes to our world. This roadmap reviews the progress and challenges of plasma technologies from a large‐scale and application ...
Pankaj Attri, Liuyang Bai, Dmitry Batomunkuev, Xueke Che, Guoxing Chen, Zhitong Chen, Wei Cui, Liguang Dou, Zhi Fang, Yangyang Fu, Fausto Gallucci, Yiming Guo, Bangdou Huang, Yinghong Li, Sirui Li, Hua Liang, Liangliang Lin, Dingxin Liu, Xiangyang Liu, Xinpei Lu, Ruonan Ma, Lanlan Nie, Guanghui Niu, Yuge Ouyang, Tao Shao, Jing Sun, Yong Tang, Xin Tu, Sergey P. Vashchenko, Hang Wang, Xianjie Wang, Anke Weidenkaff, Yun Wu, Haohan Xie, Shaojun Xu, Zongxian Yang, Fangli Yuan, Cheng Zhang, Hao Zhang, Jishen Zhang, Shuai Zhang, Yejian Zhang, Zhe Zhang, Zhibo Zhang, Peng Zhao, Shicheng Zhao, Linlin Zhong, Renwu Zhou, Zhenhua Zhou, Xi Zhu, Yifei Zhu, Haohua Zong   +51 more
wiley   +1 more source

NDAWL-PINN: a new non-dimensionalization and multi-task learning approach for efficient training of physics-informed neural networks to solve the shallow water equations

Engineering Applications of Computational Fluid Mechanics
The exploration of deep learning methodologies has recently generated significant interest in the use of Physics-Informed Neural Networks (PINNs) to address complex physical problems governed by partial differential equations (PDEs).
Xin Qi, Dawei Zhang, Fan Wang, Wuxia Bi, Mingda Lu   +4 more
doaj   +1 more source

Review of Insulation Ageing and Lifecycle Management for Oil‐Immersed Power Equipment: From Conventional Testing Analysis to Artificial Intelligence Method

High Voltage, EarlyView.
ABSTRACT Oil‐immersed power equipment is a critical component for energy transmission and secure operation of modern power grids. The ageing of oil–paper insulation under combined electrical, thermal and mechanical stresses directly affects equipment lifetime and system reliability. This paper presents a comprehensive review of the ageing behaviour and
Ruijin Liao, Jian Hao, Jing Xu, George Chen, Huanchao Cheng, Jiacong Li, Xize Dai   +6 more
wiley   +1 more source

A Physics‐Informed Deep Learning Method With Adaptively Weighted Loss for Modeling Soil Water Flows

Water Resources Research
Richards' equation, widely used to model soil water flows, presents numerical challenges due to the high nonlinearity of its constitutive relationships.
Cunwen Li, Yan Zhu, Xiaoping Zhang, Lili Ju, Qiang Luo, Hui Feng   +5 more
doaj   +1 more source

Differentiable Randers‐Finsler Eikonal Solvers

Computer Graphics Forum, EarlyView.
Abstract Fast and differentiable solvers for anisotropic and asymmetric distance fields are a key primitive in geometry processing, enabling gradient‐based optimization over metrics, drift fields, and downstream objectives that depend on geodesic distances and geodesics.
Barak Gahtan, Jacob Shpund, Alex M. Bronstein   +2 more
wiley   +1 more source

Wave time scattering and duality using PINN.

, 2023
This program provides an example of PINN solving the time-domain time-disturbed wave equation.The input file is a snapshot of two earlier wavelengths and their coordinates.We use python3.6, tensorflow2.6, and matplot3.0.3.
yingzhen (16944387)
core   +1 more source

Data-driven solutions and parameter estimations of a family of higher-order KdV equations based on physics informed neural networks

Scientific Reports
Physics informed neural network (PINN) demonstrates powerful capabilities in solving forward and inverse problems of nonlinear partial differential equations (NLPDEs) through combining data-driven and physical constraints. In this paper, two PINN methods
Jiajun Chen, Jianping Shi, Ao He, Hui Fang   +3 more
doaj   +1 more source

Volumetric Inverse Rendering via Neural Radiative Transfer

Computer Graphics Forum, EarlyView.
Abstract Volumetric inverse rendering seeks to recover the optical properties of participating media from images. Existing approaches either rely on differentiable stochastic light transport simulation, which require substantial algorithmic effort, or use simplified models that fail to capture global illumination.
Ntumba Elie Nsampi, Adarsh Djeacoumar, Hans‐Peter Seidel, Tobias Ritschel, Thomas Leimkühler   +4 more
wiley   +1 more source