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

Design Example: Designing a Residential Plumbing System01:25

Design Example: Designing a Residential Plumbing System

701
The design of residential plumbing systems requires carefully evaluating water demand, flow rates, and pressure dynamics to ensure both efficiency and reliability. The nature of water flow within pipes is defined by its Reynolds number, which classifies flow as either laminar (smooth) or turbulent.
701

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

On-Device IoT-Based Predictive Maintenance Analytics Model: Comparing TinyLSTM and TinyModel from Edge Impulse.

Sensors (Basel, Switzerland)·2022
Same author

Mobility-Aware Proactive Edge Caching Optimization Scheme in Information-Centric IoV Networks.

Sensors (Basel, Switzerland)·2022
Same author

Rainfall-Induced Landslide Prediction Using Machine Learning Models: The Case of Ngororero District, Rwanda.

International journal of environmental research and public health·2020
查看所有相关文章

相关实验视频

Updated: Jul 10, 2025

High-precision Electromagnetic Flowmeter with Empty Pipe Detection via Complex Programmable Logic Device-based Waveform Recognition
05:11

High-precision Electromagnetic Flowmeter with Empty Pipe Detection via Complex Programmable Logic Device-based Waveform Recognition

Published on: June 27, 2025

10

智能建筑:使用TinyML检测水泄漏

Othmane Atanane1, Asmaa Mourhir1, Nabil Benamar1,2

  • 1School of Science and Engineering, Al Akhawayn University in Ifrane, P.O. Box 104, Hassan II Avenue, Ifrane 53000, Morocco.

Sensors (Basel, Switzerland)
|November 25, 2023
PubMed
概括

这项研究介绍了一种由TinyML驱动的系统,用于使用声学数据检测建筑物中的水泄漏. 效率网模型的准确度超过97%,使得水资源管理能够高效,实时,干预最小.

关键词:
在美国,CNN是CNN.有效的网络有效的网络在TinyML中使用TinyML.加速度计的加速计是什么?声学数据 声学数据 声学数据深度学习是一种深度学习.标杆图是指一个标杆图.

更多相关视频

Design and Use of a Full Flow Sampling System FFS for the Quantification of Methane Emissions
08:18

Design and Use of a Full Flow Sampling System FFS for the Quantification of Methane Emissions

Published on: June 12, 2016

16.8K
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

559

相关实验视频

Last Updated: Jul 10, 2025

High-precision Electromagnetic Flowmeter with Empty Pipe Detection via Complex Programmable Logic Device-based Waveform Recognition
05:11

High-precision Electromagnetic Flowmeter with Empty Pipe Detection via Complex Programmable Logic Device-based Waveform Recognition

Published on: June 27, 2025

10
Design and Use of a Full Flow Sampling System FFS for the Quantification of Methane Emissions
08:18

Design and Use of a Full Flow Sampling System FFS for the Quantification of Methane Emissions

Published on: June 12, 2016

16.8K
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

559

科学领域:

  • 环境科学 环境科学
  • 计算机科学 计算机科学
  • 工程 工程师 工程师 工程师

背景情况:

  • 全球水资源短缺需要有效的水资源管理策略.
  • 建筑管道中未被检测到的泄漏造成的水浪费是一个重大问题.
  • 基础设施老化和效率低下的做法加剧了水损失.

研究的目的:

  • 为智能建筑物开发一种有效的,低干预的水泄漏检测方法.
  • 探索边缘计算和TinyML用于实时水资源管理的应用.
  • 提高水资源紧张地区的水利用效率.

主要方法:

  • 利用PVC管道中的水泄漏声学数据集.
  • 预先处理声学数据,将其转化为分析的声谱图.
  • 使用五个卷积神经网络 (CNN) 变体 (EfficientNet,ResNet,AlexNet,MobileNet V1,MobileNet V2) 进行应用转移学习.
  • 使用量子化优化了EfficientNet模型,以便在Arduino Nano 33 BLE边缘设备上部署.

主要成果:

  • 效率网模型实现了97.45%的最大测试精度,98.57%的回忆,96.70%的精度和97.63%的F1得分.
  • 量子化的EfficientNet模型展示了低推断时间 (1932 ms),最小的RAM使用 (255.3 KB) 和小闪存需求 (48.7 KB).
  • 提议的TinyML解决方案可实现水泄漏检测的高效,本地化决策.

结论:

  • TinyML和边缘计算为智能建筑中的实时水泄漏检测提供了可行的解决方案.
  • 开发的系统可以在最小的人类干预下显著减少水浪费.
  • 在边缘设备上的高效模型部署为具有成本效益和可扩展的智能水管理系统铺平了道路.