Acoustic Fault Diagnosis of Train Bearings Using Gorilla Troops Optimizer Combined with Variational Mode Decomposition and Frost Algorithm

To tackle the challenges in bearing fault signal recognition, this paper proposes an advanced strategy for diagnosing bearing fault signals by combining the Improved Variational Mode Decomposition (IVMD) method with frost spatial filtering. Specifically, IVMD utilizes the Artificial Gorilla Troops Optimizer (GTO), which has outstanding performance in high-dimensional global optimization, to estimate the sensitive parameters of Variational Mode Decomposition (VMD). Then, the observed signal y is input into the anti-aliasing and highly separable VMD for decomposition. After decomposition, the Intrinsic Mode Functions (IMF1-IMF7) data are screened by sample entropy. Finally, envelope modulation is carried out after Frost spatial filtering to diagnose the fault signals. Experimental results demonstrate that the fault information can be accurately identified under different working conditions, proving that this method improves fault diagnosis accuracy, and the signal-to-noise ratio (SNR) is improved by at least 21.74%, demonstrating the recognition and noise reduction capabilities of this improved method for fault information.