Method for extracting fault characteristic frequency of main bearing of aero-engine in service environment

The invention discloses an aero-engine main bearing fault characteristic frequency extraction method in a service environment. The invention relates to a novel method for extracting a weak fault signal characteristic frequency of a main bearing of a military aero-engine in a strong noise service environment. The method mainly comprises the following steps: aiming at a collected aero-engine casing measuring point vibration acceleration signal, separating the signal by adopting a weak random signal extraction method technology to obtain a discrete component and a random component of the signal, and constructing a dimensionless characteristic quantity reflecting a bearing fault characteristic frequency local/overall characteristic significant level; the method can effectively extract random components in the signals, filter interference components of discrete parts such as the rotating speed and the gear meshing frequency in square envelope spectrum analysis, improve the amplitude corresponding to the frequency