Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Accurate Ultrasonic Thickness Measurement for Arbitrary Time-Variant Thermal Profile.

Sensors (Basel, Switzerland)·2024
Same author

Large-area inspection of defects in metal plates using multi-mode guided acoustic waves and sparse sensor networks.

Ultrasonics·2024
Same author

Void-Engineered Metamaterial Delay Line with Built-In Impedance Matching for Ultrasonic Applications.

Sensors (Basel, Switzerland)·2024
Same author

Ultrasonic Testing of Mechanical Changes in a Water-Filled Pipe with Multi-Mode and Broadband Signals and Two-Level Compensation.

Sensors (Basel, Switzerland)·2022
Same author

Thermodynamics, dynamics, and structure of supercritical water at extreme conditions.

Physical chemistry chemical physics : PCCP·2020
Same author

Electrical conductivity, ion pairing, and ion self-diffusion in aqueous NaCl solutions at elevated temperatures and pressures.

The Journal of chemical physics·2019
Same journal

CFREE-uPIV: A contrast-free resolution-enhanced ultrasound particle image velocimetry for rapid microvascular flow imaging.

Ultrasonics·2026
Same journal

Ultrasonic-induced nucleation kinetics in directed energy deposition.

Ultrasonics·2026
Same journal

Grain size evaluation in polycrystalline materials using the interference effect of bounded ultrasonic beams at the Rayleigh critical angle.

Ultrasonics·2026
Same journal

Ultrasonic characterization of functionally graded materials using a continuously graded model and spectral inversion.

Ultrasonics·2026
Same journal

Frequency-wavenumber domain inversion for arterial viscoelasticity.

Ultrasonics·2026
Same journal

Pressure- and frequency-dependent acoustic behavior of second-generation acoustic reporter genes-expressing bacteria for optimized ultrasound imaging.

Ultrasonics·2026
查看所有相关文章

相关实验视频

Updated: Jun 22, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

515

使用稀疏的声学传感器网络和知识转移的深度学习进行多层次的结构损伤表征.

Rajendra P Palanisamy1, Do-Kyung Pyun1, Alp T Findikoglu1

  • 1Materials Physics and Applications (MPA), Los Alamos National Laboratory, Los Alamos, NM 87545, USA.

Ultrasonics
|June 30, 2024
PubMed
概括
此摘要是机器生成的。

本研究引入了一种机器学习方法,用于复杂结构中的结构健康监测,大大提高了缺陷诊断的准确性. 该方法增强了知识传输,减少了数据需求,提高了检查效率.

关键词:
声学传感器网络中的声学传感器网络.适应性的卷积.数据驱动的结构健康监测知识转移 深度学习网络空间助手 (NSA) 的使用培训数据的稀疏性 培训数据的稀疏性

更多相关视频

Data Acquisition Protocol for Determining Embedded Sensitivity Functions
07:46

Data Acquisition Protocol for Determining Embedded Sensitivity Functions

Published on: April 20, 2016

6.1K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.0K

相关实验视频

Last Updated: Jun 22, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

515
Data Acquisition Protocol for Determining Embedded Sensitivity Functions
07:46

Data Acquisition Protocol for Determining Embedded Sensitivity Functions

Published on: April 20, 2016

6.1K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.0K

科学领域:

  • 工程 工程师 工程师 工程师
  • 材料科学 材料科学 材料科学
  • 人工智能的人工智能

背景情况:

  • 标准的结构健康监测 (SHM) 方法与复杂的结构作斗争,需要先进的技术.
  • 用于SHM的数据驱动机器学习需要广泛的训练数据集,这些数据集对于各种缺陷具有挑战性.
  • 神经网络中的知识传输可以降低SHM系统的数据需求和计算成本.

研究的目的:

  • 展示基于机器学习的多层次损害表征方法,用于稀疏的传感器网络,具有知识传输能力.
  • 在深度学习算法中引入用于SHM的高效知识转移的新技术.
  • 评估拟议方法在复杂结构上的缺陷定位和严重性评估方面的有效性.

主要方法:

  • 开发基于机器学习的损害特征框架,利用稀疏的传感器网络.
  • 实施一个新的网络空间辅助和自适应卷积技术,以实现高效的知识传输.
  • 在带有诱导缺陷的板上进行实验验证,分析多个分散波.

主要成果:

  • 拟议的方法提高了知识转移损害的特征化,在局部化方面提高了50%,在严重程度评估方面提高了24%.
  • 发现多个分散波含有丰富的缺陷签名,提高了识别和量化准确性.
  • 通过使用固定传感器网络的多重分散波来实现所有损害特征级别的100%预测准确性.

结论:

  • 机器学习与知识转移为复杂结构中的SHM提供了强大的解决方案,克服了传统方法的局限性.
  • 拟议的自适应卷积和空间辅助技术促进了高效的知识传输,减少了对大型数据集的需求.
  • 在SHM中使用多重分散波显著提高了缺陷检测和量化准确度,为更可靠的检查系统铺平了道路.