Mechanical Incipient Fault Detection and Performance Analysis Using Adaptive Teager-VMD Method

For large rotating machinery with low speed and heavy load, the incipient fault characteristics of rolling bearings are particularly weak, making it difficult to identify them effectively by direct signal processing methods. To resolve this issue, we propose a novel approach detecting features that combines energy enhancement decomposition. First, structure conventional Teager algorithm is modified further increase micro-impact component hence impact amplitude. Then, kind composite chaotic mapping constructed extend original fruit fly optimization (FOA) framework, improving FOA’s randomness search power. The effective intrinsic mode functions (IMFs) determined searching for optimal combination values key parameters variational decomposition (VMD) improved FOA (ICFOA). kurtosis index then used select IMFs most relevant information. Finally, sensitive components analyzed multiple early determine detailed information about faults. Moreover, evaluated simulation measured signal. comprehensive evaluation indicates has clear advantages over other excellent methods in extracting feature equipment great potential application engineering.