Search research articles

Search research articles

Related Experiment Videos

Degradation prediction model based on a neural network with dynamic windows.

Xinghui Zhang¹, Lei Xiao^2,3, Jianshe Kang⁴

¹Mechanical Engineering College, Shijiazhuang 050003, China. dynamicbnt@aim.com.

Sensors (Basel, Switzerland)

|March 26, 2015

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

Screening and Characteristics of Cadmium-Tolerant Microorganisms in Soil Remediation.

Bulletin of environmental contamination and toxicology·2025

Same author

1-benzyl-6-nitro-4-phenyl-4-(methoxycarbonyl)-2(1H)-pyridinone, a novel pirfenidone derivative, alleviate hepatic fibrosis through T cells.

Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie·2025

Same author

Effects of Longitudinal External Magnetic Field on Metal Transfer Behavior and Spatter Formation in CO<sub>2</sub> Arc Welding.

Materials (Basel, Switzerland)·2025

Same author

Dietary Effects of Different Proportions of Fermented Straw as a Corn Replacement on the Growth Performance and Intestinal Health of Finishing Pigs.

Animals : an open access journal from MDPI·2025

Same author

Research Progress on the Modification of B and Al High-Energy Fuels for Powder Fuel Ramjet Applications.

ACS applied materials & interfaces·2025

Same author

An interpretable machine learning model assists in predicting induction chemotherapy response and survival for locoregionally advanced nasopharyngeal carcinoma using MRI: a multicenter study.

European radiology·2025

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

This study introduces a novel neural network model with dynamic windows for Remaining Useful Life (RUL) estimation. The method effectively predicts component degradation even with limited data, improving maintenance decisions.

Area of Science:

Mechanical Engineering
Reliability Engineering
Data Science

Background:

Effective maintenance decision-making relies on tracking mechanical component degradation.
Remaining Useful Life (RUL) estimation is crucial for predictive maintenance.
Limited run-to-failure data for high-reliability components hinders traditional RUL prediction.

Purpose of the Study:

To develop a robust RUL prediction model addressing limitations of existing methods.
To overcome challenges posed by insufficient degradation data and poor model extrapolability.
To enhance the accuracy of RUL estimation for mechanical components.

Main Methods:

A neural network model incorporating dynamic windows for RUL prediction.
Key steps include window size determination, change point detection, and rolling prediction.

Related Experiment Videos

The model does not assume specific degradation trajectory distributions.

Main Results:

The proposed model demonstrates adaptability to varying degradation indicators.
It overcomes issues of prediction convergence and fluctuation seen in other methods.
Validation using real-world field and simulation data confirms performance.

Conclusions:

The dynamic window neural network model offers a significant advancement in RUL prediction.
This approach enhances reliability and maintenance strategies for critical components.
The method provides a more accurate and adaptable solution for degradation tracking.