Enhancing Fault Diagnosis: A Hybrid Framework Integrating Improved SABO with VMD and Transformer–TELM

Lubricants, Journal Year: 2025, Volume and Issue: 13(4), P. 155 - 155

Published: March 31, 2025

Rolling bearings, as core components in mechanical systems, directly influence the overall reliability of equipment. However, continuous operation under complex working conditions can easily lead to gradual performance degradation and sudden faults, which not only result equipment failure but may also trigger a cascading effect, significantly amplifying downtime losses. To address this challenge, study proposes an intelligent diagnostic method that integrates variational mode decomposition (VMD) optimized by improved subtraction-average-based optimizer (ISABO) with transformer–twin extreme learning machine (Transformer–TELM) ensemble technology. Firstly, ISABO is employed finely optimize initialization parameters VMD. With strategy particle position update method, optimal parameter combination be precisely identified. Subsequently, are used model decompose signal through VMD, selected constructed two-dimensional evaluation system. Furthermore, diversified time-domain features extracted from these form initial feature set. deeply mine information, multi-layer Transformer introduced refine more discriminative representations. Finally, input into TELM fault diagnosis achieve precise rolling bearing faults. The experimental results demonstrate exhibits excellent terms noise resistance, accurate capture, classification. Compared traditional techniques such kernel (KELM), (ELM), support vector (SVM), Softmax, outperforms other models accuracy, recall, F1 score.

Language: Английский

An Intelligent Fault Diagnosis Model for Rolling Bearings Based on IGTO-Optimized VMD and LSTM Networks

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(8), P. 4338 - 4338

Published: April 14, 2025

To address the issue of rolling bearing fault diagnosis, this paper proposes a novel model combining Improved Gorilla Troop Optimization (IGTO) algorithm, Variational Mode Decomposition (VMD), Permutation Entropy (PE), and Long Short-Term Memory (LSTM) networks. The IGTO algorithm is used to optimize parameters VMD LSTM, enhancing signal decomposition feature extraction. proposed achieves classification accuracies 96.67% 98.96% in testing training phases, respectively, on Case Western Reserve University dataset, with minimal accuracy fluctuations. Furthermore, Jiangnan reaches an average 98.85%, highest reaching 99.48%. results also demonstrate high stability, as indicated by low standard deviations (1.2148 1.3217) narrow 95% confidence intervals ([95.75%, 97.58%] [96.73%, 97.49%]). Despite longer runtime 13.88 s per sample, model’s superior justifies computational cost. These excellent diagnostic performance, adaptability different datasets, practical applicability for diagnosis. This approach provides valuable reference predictive maintenance detection systems industrial applications.

Language: Английский