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: Sep 22, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.2K

Nonlinear Dynamic Soft Sensor Development with a Supervised Hybrid CNN-LSTM Network for Industrial Processes.

Jiaqi Zheng1, Lianwei Ma2, Yi Wu2

  • 1College of Science & Technology, Ningbo University, Ningbo 315300, People's Republic of China.

ACS Omega
|May 23, 2022
PubMed
Summary
This summary is machine-generated.

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

Frontotemporal cortical activity in Chinese male alcohol-dependent patients during a verbal fluency task: a functional near-infrared spectroscopy study.

Frontiers in psychiatry·2026
Same author

Exploring the impact of AI tools on filmmakers' creative satisfaction: the mediating role of creative efficiency and autonomy.

Frontiers in psychology·2026
Same author

Cortical dynamics of icon perception: effects of concreteness and attractiveness.

Cerebral cortex (New York, N.Y. : 1991)·2026
Same author

High Spatial Resolution In Situ Fe Isotope Analysis by Laser Ablation Collision/Reaction Cell MC-ICP-MS: Application to Chang'e-6 Lunar Samples.

Analytical chemistry·2026
Same author

Random Attention Mechanism for Deep Quality Modeling and Online Prediction in Chemical Processes.

ACS omega·2026
Same author

Beyond dormancy: Active DNA repair and protein synthesis drive antibiotic tolerance and enable synergistic eradication.

Microbiological research·2026

This study introduces a novel hybrid deep learning model for soft sensors, improving real-time industrial process monitoring with limited data by enhancing quality-relevant feature learning.

Area of Science:

  • Industrial Process Monitoring
  • Artificial Intelligence
  • Chemical Engineering

Background:

  • Soft sensors are crucial for real-time measurements in industrial processes where direct sensing is difficult.
  • Deep learning models, including Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks, show promise for complex, nonlinear industrial processes.
  • Conventional deep learning methods struggle to guarantee the inclusion of quality-relevant features in hidden states with limited training data.

Purpose of the Study:

  • To develop an advanced soft sensor model capable of effective quality-relevant feature learning even with limited industrial process data.
  • To address the limitations of conventional deep learning approaches in capturing essential quality features in hidden states.
  • To enhance the reliability and accuracy of soft sensor applications in complex industrial settings.

More Related Videos

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

652

Related Experiment Videos

Last Updated: Sep 22, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.2K
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

652

Main Methods:

  • A supervised hybrid network combining dynamic Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) was designed.
  • Multilayer dynamic CNN-LSTM architecture with improved structures was constructed.
  • Data augmentation via dynamic sample expansion was implemented at each time instant.
  • Quality variables were incorporated as input to supervised hidden units for enhanced feature learning.

Main Results:

  • The proposed hybrid CNN-LSTM soft sensor demonstrated improved learning of quality-related features.
  • The model effectively handled nonlinear and dynamic characteristics of industrial processes.
  • Validation in penicillin fermentation and debutanizer column applications confirmed the model's effectiveness.

Conclusions:

  • The developed supervised hybrid CNN-LSTM soft sensor model offers a robust solution for industrial processes with limited data.
  • The integration of dynamic CNN, LSTM, and quality variable supervision significantly enhances feature learning for soft sensors.
  • The approach provides a reliable method for real-time quality monitoring in complex industrial applications.