Related Experiment Video
Updated: Sep 11, 2025

Author Spotlight: Simulation and Analysis of the Temperature Rise of Ring Main Unit Equipment
Published on: July 5, 2024
Research on Adaptive Identification Technology for Rolling Bearing Performance Degradation Based on
Zhenghui Li1,2, Lixia Ying1, Liwei Zhan2
1College of Mechanical and Electrical Engineering, Harbin Engineering Univesity, Harbin 150500, China.
This study introduces a vibration-temperature fusion method for accurate rolling bearing performance degradation assessment. The novel approach enhances health indicator characterization and life-stage identification, outperforming vibration-only methods.
Area of Science:
- Mechanical Engineering
- Condition Monitoring
- Data Science
Background:
- Traditional rolling bearing performance degradation assessment relies heavily on vibration signals, often leading to low accuracy in identifying transition states.
- Integrating multi-source sensor data, such as temperature, can potentially improve diagnostic capabilities.
Purpose of the Study:
- To develop a vibration-temperature fusion-based adaptive method for enhanced rolling bearing performance degradation assessment.
- To improve the accuracy of identifying bearing life-stage transitions.
Main Methods:
- Constructed a multidimensional time-frequency feature set from vibration and temperature signals.
- Introduced a composite sensitivity index (CSI) for feature screening, followed by mutual information clustering and entropy weight optimization for parameter reselection.
- Employed adaptive feature fusion using auto-associative kernel regression (AFF-AAKR) and temporal residual analysis to enhance health indicator (HI) characterization.
- Developed a dual-criteria adaptive bottom-up merging algorithm (DC-ABUM) for life-stage identification.
Main Results:
- The proposed vibration-temperature fusion method demonstrated superior performance compared to traditional vibration-only approaches.
- The integration of temperature data and advanced feature engineering improved the characterization of bearing degradation trends.
- The adaptive fusion and life-stage identification algorithms effectively captured transition states.
Conclusions:
- Vibration-temperature fusion offers a more robust and accurate approach to rolling bearing performance degradation assessment.
- The developed adaptive method, incorporating CSI, AFF-AAKR, and DC-ABUM, significantly enhances the identification of bearing health and remaining useful life.
- This fusion strategy provides a promising direction for improving the reliability of condition monitoring systems.
Related Concept Videos
Bearings: Problem Solving
Temperature Dependent Deformation
Bearing Stress
Due to the intricacy of these microforces, an average value, known as bearing stress, is often used by...
Thermal Stress
Collar Bearings
Design of Transmission Shafts - Stress Analysis

