Composite Fault Diagnosis of VMD-FastICA Bearing Based on Parameter Optimization

Rolling bearings are one of the most commonly used bearing components in rotating machinery and are also one of the most prone to failure. Therefore, the diagnosis of the condition of rolling bearings is of significant importance for safety and economy. In practical engineering, bearings are often interfered with by vibration signals from other components, which makes it difficult to extract the fault characteristic information of bearings due to noise or other signals. In this paper, to address the problem of difficulty in extracting compound fault characteristic information of rolling bearings under strong noise background, a feature extraction method based on sparrow algorithm optimized VMD parameters and energy entropy and envelope entropy mean minimization is proposed, and combined with FastICA to achieve signal separation, denoising, and feature enhancement. This method is applied to the fault feature extraction of simulated signals, achieving denoising and feature enhancement of separated signals. The results of simulation and experimental analysis demonstrate the feasibility of this method. Furthermore, a comparison with other methods is conducted to validate the superiority of this method.