Jove
Visualize
Contact Us

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

Machine Learning-Driven Techno-Economic Uncertainty Analysis in Batch Pharmaceutical Manufacturing.

ACS omega·2026
Same author

A Diagnosis-Based Siamese Network for Fault Detection Through Transfer Learning.

Journal of chemical information and modeling·2025
Same author

Comprehensive Characterization of Commercial Reverse Osmosis Membranes through High-Temperature Cross-Flow Filtration.

ACS omega·2024
Same author

Facile and Efficient Phase-Selective Powder Polymer Organogelator for Oil Spill Remediation.

Langmuir : the ACS journal of surfaces and colloids·2022
Same author

Unsupervised Machine Learning Applied to Seismic Interpretation: Towards an Unsupervised Automated Interpretation Tool.

Sensors (Basel, Switzerland)·2021
Same author

Flocculation Efficiency and Spatial Distribution of Water in Oil Sands Tailings Flocculated with a Partially Hydrophobic Graft Copolymer.

ACS applied materials & interfaces·2021
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: May 30, 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

455

Can Focusing on One Deep Learning Architecture Improve Fault Diagnosis Performance?

João G Neto1, Karla Figueiredo2, João B P Soares3

  • 1Department of Chemical and Materials Engineering, Pontifical Catholic University of Rio de Janeiro, 225, Marquês de São Vicente Street, Gávea, Rio de Janeiro, RJ 22451-900, Brazil.

Journal of Chemical Information and Modeling
|January 30, 2025
PubMed
Summary
This summary is machine-generated.

Focusing on a single deep learning architecture, like convolutional neural networks, significantly improved fault diagnosis performance on the Tennessee Eastman Process dataset. This targeted approach achieved higher accuracy than broader methods, demonstrating its potential for industrial applications.

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

8.9K
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

944

Related Experiment Videos

Last Updated: May 30, 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

455
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

8.9K
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

944

Area of Science:

  • Chemical Engineering
  • Computer Science
  • Artificial Intelligence

Background:

  • Machine learning model selection often involves broad exploration of diverse architectures.
  • This can limit in-depth analysis and optimization of established methods.
  • Developing efficient fault diagnosis systems is crucial for industrial process safety and optimization.

Purpose of the Study:

  • To investigate the effectiveness of a focused deep learning approach for fault diagnosis.
  • To assess performance improvements by concentrating on a single architecture type.
  • To evaluate a modified convolutional neural network on the Tennessee Eastman Process dataset.

Main Methods:

  • Concentrated on a single deep learning architecture: convolutional neural networks (CNNs).
  • Utilized the benchmark Tennessee Eastman Process dataset for case study.
  • Investigated modifications to a reference CNN-based model for enhanced fault diagnosis.

Main Results:

  • Achieved a maximum average F1-score of 89.85% for fault classification.
  • Demonstrated a 7.47% improvement over the baseline model's performance.
  • Outperformed other reported results in the literature for this benchmark dataset.

Conclusions:

  • A focused approach on a specific deep learning architecture can significantly enhance fault diagnosis performance.
  • Modified CNNs show strong potential for improving industrial process monitoring.
  • The findings suggest this focused strategy warrants further exploration on diverse datasets.