Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jan 9, 2026

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

996

Cable partial discharge identification network based on adaptive residual diffusion denoising and morphological

Long Chen1, Qiong Li2, Guohua Long3

  • 1School of Information Engineering, Nanchang Hangkong University, Nanchang, 330000, China.

Scientific Reports
|December 1, 2025
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Ligand-Receptor-Based Novel TWEAK CAR-T Exhibits Potent Antitumor Activity Against Liver Cancer by Targeting FN14.

Journal of immunotherapy (Hagerstown, Md. : 1997)·2026
Same author

Flexible Multifunctional Composite Phase Change Materials for Enhanced Thermal Safety of Battery Energy Storage Systems.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Pheochromocytoma complicated with acute myocardial infarction and cerebral infarction: a case report.

Frontiers in oncology·2026
Same author

Identification and diagnosis of flow instability in parallel channel systems using machine learning.

Communications engineering·2026
Same author

The relationship between lactate-to-albumin ratio and prognosis in patients with sepsis-induced cardiomyopathy: a retrospective cohort study.

Scientific reports·2026
Same author

The effect of internal jugular vein reflux on ischemic stroke with cerebral microbleeds: a prospective clinical study.

Annals of medicine·2026

A new deep learning model, ARDDMA-Net, accurately identifies partial discharge (PD) in power cables. This advanced method effectively reduces noise and enhances signal features for improved power grid reliability.

Area of Science:

  • Electrical Engineering
  • Signal Processing
  • Artificial Intelligence

Background:

  • Accurate identification of partial discharge (PD) is vital for power cable insulation health and grid reliability.
  • Low amplitude PD signals and noise interference complicate accurate identification.
  • Existing methods struggle with low signal-to-noise ratio (SNR) and multi-scale signal characteristics.

Purpose of the Study:

  • To propose a novel deep learning architecture, ARDDMA-Net, for accurate PD identification in power cables.
  • To enhance the detection of weak PD signals masked by noise.
  • To improve the classification accuracy of PD signals under challenging conditions.

Main Methods:

  • Developed a two-stage deep learning network: ARDDMA-Net.
Keywords:
Adaptive residualConvolutional neural networkDeep learningMulti-scale attentionPartial discharge

More Related Videos

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

725

Related Experiment Videos

Last Updated: Jan 9, 2026

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

996
Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique
04:48

Swin-PSAxialNet: An Efficient Multi-Organ Segmentation Technique

Published on: July 5, 2024

725
  • Stage 1: Adaptive residual diffusion denoising and morphological attention mechanism to suppress noise while preserving PD signal features.
  • Stage 2: Multi-scale attention-enhanced ResNet-1D CNN for feature extraction and classification.

    • Main Results:

    • ARDDMA-Net demonstrated stable recognition performance across various noise levels and data loss scenarios.
    • The model effectively mitigated noise masking and misclassification of weak PD pulses.
    • Achieved superior identification accuracy compared to traditional machine learning and existing deep learning models.

    Conclusions:

    • ARDDMA-Net offers a robust solution for accurate PD identification in power systems.
    • The proposed architecture effectively addresses limitations of traditional methods in low SNR and complex signal conditions.
    • This advancement contributes to enhanced power grid reliability through early defect detection.