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 Concept Videos

  1. Home
  3. Research Domains
  5. Engineering
  7. Mechanical Engineering
  9. Mechanical Engineering Asset Management
  11. Prediction Of The Remaining Useful Life Of Bearings Through Cnn-bi-lstm-based Domain Adaptation Model

Prediction of the Remaining Useful Life of Bearings Through CNN-Bi-LSTM-Based Domain Adaptation Model

Feifan Li1, Zhuoheng Dai1, Lei Jiang1

  • 1School of Information Engineering, College of Science & Technology, Ningbo University, Ningbo 315000, China.

Sensors (Basel, Switzerland)
|November 9, 2024

Related Experiment Videos

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

2.6K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.7K
Analyzing Mitochondrial Morphology Through Simulation Supervised Learning
12:06

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Published on: March 3, 2023

3.9K

View abstract on PubMed

Summary
This summary is machine-generated.

This study introduces a novel approach for predicting the remaining useful life (RUL) of mechanical bearings using neural networks for feature engineering. The method enhances model generalization across different operating conditions, improving predictive accuracy.

Area of Science:

  • Mechanical Engineering
  • Artificial Intelligence
  • Predictive Maintenance

Background:

  • Accurate remaining useful life (RUL) prediction for mechanical bearings is vital for industrial maintenance and cost reduction.
  • Traditional RUL models struggle with generalization due to varying operating conditions and reliance on manual feature engineering.
  • Developing robust RUL prediction models that adapt to diverse operational environments is a significant challenge.

Purpose of the Study:

  • To develop a neural network-based feature engineering approach for RUL prediction.
  • To enhance the generalizability of RUL prediction models across different operating conditions.
  • To simplify the development and maintenance of bearing health assessment models.

Main Methods:

  • Utilized a Convolutional Neural Network (CNN) for automated feature engineering from sensor data.
Keywords:
Bi-LSTMCNNRULbearing

Related Experiment Videos

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

2.6K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.7K
Analyzing Mitochondrial Morphology Through Simulation Supervised Learning
12:06

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Published on: March 3, 2023

3.9K
  • Employed a Bidirectional Long Short-Term Memory (Bi-LSTM) network to capture time-series degradation patterns.
  • Integrated domain adaptation techniques to mitigate feature distribution discrepancies caused by varying operating conditions.

    • Main Results:

    • The proposed CNN-Bi-LSTM model effectively predicts the RUL of mechanical bearings.
    • Domain adaptation significantly improved model generalizability under different operating conditions.
    • Eliminated the need for manual feature selection and prior domain knowledge.

    Conclusions:

    • Neural network-based feature engineering offers a robust solution for RUL prediction.
    • The integrated domain adaptation enhances model performance and reliability in real-world industrial settings.
    • This approach streamlines predictive maintenance processes, leading to reduced costs and improved operational efficiency.
    domain adaptation