Motor Bearing Fault Detection Based on ICEEMDAN-EPO-SVM

DOI: https://doi.org/10.62517/jbdc.202501211

Author(s)

Zhi Wang*, Hao Wang, Yuan Wang

Affiliation(s)

Hohhot Power Supply Branch, Inner Mongolia Power (Group) Co., Ltd, Hohhot, Inner Mongolia, China *Corresponding Author

Abstract

To address the issues of frequent damage to motor bearings and the limitations of traditional diagnostic methods, such as time consumption and low accuracy, a novel bearing fault detection strategy is proposed. This strategy integrates the Eagle Perching Optimization (EPO) algorithm for optimizing key parameters of Support Vector Machine (SVM). Initially, the Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (ICEEMDAN) technique is utilized to analyze vibration signals, thereby obtaining a set of Intrinsic Mode Functions (IMFs). Subsequently, correlation analysis is conducted for feature selection, followed by reconstruction, where the energy moment serves as the fault feature vector. Given that SVM classification performance is significantly influenced by the configuration of its key parameters, the EPO algorithm is utilized to optimize and determine these parameters effectively. The fault feature vectors are then classified using the EPO-optimized SVM model. Experimental results demonstrate that the proposed ICEEMD-EPO-SVM approach achieves a comprehensive fault detection accuracy of 97.5%.

Keywords

Motor; Bearing; ICEEMDAN; Eagle Perching Optimization; SVM

References

[1]Wang X, Li J, Jing Z, et al. Fault diagnosis method of rolling bearing based on SSA-VMD and RCMDE. Scientific Reports, 2024, 14(1):30637-30637. [2]Song J, Nie X, Wu C, et al. A Novel Intelligent Fault Diagnosis Method of Rolling Bearings Based on the ConvNeXt Network with Improved DenseBlock. Sensors, 2024, 24(24):7909-7909. [3]Feng Y, Liu P, Du Y, et al. Cross working condition bearing fault diagnosis based on the combination of multimodal network and entropy conditional domain adversarial network. Journal of Vibration and Control, 2024, 30, (23-24):5375-5386. [4]Li K, Chen X, Hu X, et al. Failure Libration Model of Bearing And Its Fault Diagnosis Based on EMD//IEEE, IEEE Beijing Section. Proceedings of 2013 IEEE 4th International Conference on Software Engineering and Service Science. Joint Laboratory of Ocean-based Flight Vehicle Measurement And Control China Satellite Maritime Tracking and Control Department, 2013, 1028-1031. [5]Lin J. Improved Ensemble Empirical Mode Decomposition and its Applications to Gearbox Fault Signal Processing. International Journal of Computer Science Issues, 2012, 9(6):194-199. DOI:10.1007/978-3-642-27869-3_14. [6]Ralhan S, Singh M, Sahu N, et al. Comparative Analysis of FOPI Controller using Eagle Perching Optimization for Fuel Cell connected Distributed Generation System. 2021. DOI:10.1109/ICAECT49130.2021.9392443. [7]Hong T, Hongliang D, Yi D. Tapered Roller Bearing Failure Diagnosis Based on Improved Probability Box Model. IEEE ACCESS, 2020, 8151452-151464.