Health Assessment of Rolling Bearings Based on Multivariate State Estimation and Reliability Analysis
Open Access
- 12 May 2025
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 15 (10) , 5396
Abstract
Rolling bearing is an indispensable part of mechanical rotating parts, which plays an important role in reducing friction and ensuring the rotation accuracy of rotating parts. It is necessary to carry out a health assessment of the bearing. Current health assessment methods for rolling bearings only extract strongly related feature indicators and input them into the health assessment model without considering the profound impact external conditions have on the fluctuation of feature indicators, which will lead to an inaccurate health assessment. Besides, most methods evaluating the health of rolling bearings only consider the real-time index data but do not make full use of bearing maintenance data for reliability modeling and analysis, actually reducing the hierarchy and rationality of the health assessment. Therefore, this paper combines multivariate state estimation (MSET) and reliability analysis to evaluate the health of rolling bearings. Firstly, the health baseline of the rolling bearing under multi-speed conditions is established based on MSET, which collects the history health data of rolling bearings under various working conditions and learns the impact of working conditions on health data. Subsequently, Mahalanobis distance is used to measure the degree of deviation from the health baseline, and calculated Mahalanobis distance is input into the health mapping function to get the initial health score. Finally, combined with the reliability analysis correcting the initial score, the final health score is obtained, which can provide data support for intelligent operation and maintenance and a decision-making basis for equipment maintenance. The proposed health assessment method is validated using the bearing dataset from Case Western Reserve University and historical failure data of rolling bearings. The proposed method reduces speed-related influences in bearing health evaluation, dynamically adjusting the health assessment result through the reliability model to track performance degradation throughout the bearing’s service life.Keywords
Funding Information
- National Natural Science Foundation of China (52372402, 51975487)
- National Natural Science Foundation of China (52372402, 51975487)
This publication has 12 references indexed in Scilit:
- Multivariate state estimation-based condition monitoring of slurry circulating pumps for wet flue gas desulfurization of power plantsEngineering Failure Analysis, 2024
- Hidden Markov Models based intelligent health assessment and fault diagnosis of rolling element bearingsPLOS ONE, 2024
- EMA Health Indicator Extraction Based on Improved Multivariate State Estimation Technique With a Composite OperatorIEEE Sensors Journal, 2023
- Continuous Health Monitoring of Bearing by Oscillatory Sparsity Indices Under Non Stationary Time Varying Speed ConditionIEEE Sensors Journal, 2022
- An Integrated Approach for Bearing Health Indicator and Stage Division Using Improved Gaussian Mixture Model and Confidence ValueIEEE Transactions on Industrial Informatics, 2021
- Correlation-Aware Sport Training Evaluation for Players With Trust Based on Mahalanobis DistanceIEEE Access, 2021
- Wind turbine condition monitoring based on a novel multivariate state estimation techniqueMeasurement, 2021
- Data-Driven Method for Predicting Remaining Useful Life of Bearing Based on Bayesian TheorySensors, 2020
- Bearing performance degradation assessment using long short-term memory recurrent networkComputers in Industry, 2019
- Rolling element bearing diagnostics using the Case Western Reserve University data: A benchmark studyMechanical Systems and Signal Processing, 2015