90 / 2025-04-12 16:13:17

Fault Diagnosis of Locomotive Bearings Based on Time-Frequency Fusion Using VMD-FFT

fault diagnosis,HXD1 locomotive,fast Fourier transform,ant colony optimization algorithm,RFCAConv-Transformer

A fault diagnosis method based on VMD-FFT time-frequency feature fusion is proposed for vibration signals collected from HXD1 locomotive motor bearings under strong noise. Initially, Variational Mode Decomposition (VMD) is applied to decompose the collected vibration signals into Intrinsic Mode Functions (IMFs) for noise reduction. To optimize the penalty factor $\alpha $ and the decomposition mode number $K$ in VMD, Ant Colony Optimization (ACO) algorithm is used to select the optimal parameter combination. Subsequently, the Fast Fourier Transform (FFT) is applied to extract frequency-domain information from the signal. Next, the time-frequency domain features extracted from the signal are fused, and a fault diagnosis model combining the cross-attention mechanism and Receptive Field Coordinate Attention Convolutional (RFCAConv)-Transformer is employed for rapid identification and classification of bearing faults. Experimental results demonstrate that the RFCAConv-Transformer model, optimized using the ACO algorithm, significantly outperforms other models in terms of classification accuracy. Furthermore, compared to models without ACO optimization, the proposed method also yields superior results.