Three-order tensor singular value decomposition method and system for mechanical fault diagnosis

The invention relates to the technical field of fault diagnosis, and discloses a singular value decomposition method and system of a third-order tensor for mechanical fault diagnosis, and the method comprises the following steps: 1) collecting signals generated when a machine works from a plurality of channels according to a set sampling frequency and a set sampling point number; 2) constructing a three-order tensor signal by signals acquired by a plurality of channels to be diagnosed; 3) performing adaptive decomposition on the constructed third-order tensor signal by adopting a tensor singular value decomposition formula to obtain a component signal; and 4) comparing whether the frequency of the obtained component signal is consistent with the mechanical fault characteristic frequency or not so as to diagnose whether the machine has a fault or not. According to the method, basic support can be provided for fault diagnosis and state monitoring of the mechanical equipment, effective decomposition and fault fe