Fan blade crack fault diagnosis method based on Hilbert-Huang transformation

Abstract

The invention discloses a fan blade crack fault diagnosis method based on Hilbert-Huang transformation, and belongs to the field of equipment fault diagnosis. The method mainly comprises the followingsteps: S1, establishing an expansion sample database of blade cracks; S2, acquiring vibration mode displacement parameters of the blade in normal and different crack states, and establishing a sampledatabase of blade cracks; S3, decomposing the vibration mode displacement parameters in the sample database into a plurality of basic intrinsic mode functions by utilizing an empirical mode decomposition method; S4, performing spatial Hilbert transform on an intrinsic mode function component to obtain corresponding energy distribution, namely a Hilbert spectrum, and judging the crack position according to mutation of singular points at the crack section; S5, defining a characteristic frequency representing the crack depth of the blade to realize judgment of the crack depth; S6, expanding thevibration mode displacement parameters in the database and inputting the parameters into the crack fault diagnosis model, and diagnosing faults.

Description

technical field [0001] The invention belongs to the field of fault diagnosis, in particular to a fault diagnosis method for fan blade cracks based on Hilbert-Huang transformation. Background technique [0002] The fan is a machine that uses the rotation of the impeller to increase the pressure of the gas and transport the gas. It is responsible for replacing the air and ensuring the health of the staff. It is widely used in mining, agriculture, metallurgy, petroleum, pharmaceuticals, chemicals, aerospace, navigation, energy and Performance and other engineering fields. The blade is an important part of the fan rotor system, and the working conditions are harsh and the load is complex during operation. Therefore, the long-term operation of fan blades will produce cracks in different degrees, and even cause the blades to break, which will threaten the safe operation of the unit and cause major safety accidents. [0003] Existing fan blade crack fault diagnosis methods have t...

AI Technical Summary

Problems solved by technology

[0003] Existing fan blade crack fault diagnosis methods have their own limitations: First, the signal processing-based diagnosis method—the on-site vibration situation is complex and unstable, and it is still a technical problem to separate the vibration signal from the noise signal. The signal is often annihilated by the noise signal or the overall vibration signal, and the current fault monitoring system mainly judges whether there is a fault based on whether the fault characteristics exceed the set threshold, and it is still unable to perform positioning detection; secondly, the fault diagnosis method based on artificial intelligence ——Although the fault diagnosis method based on artificial intelligence is fast and convenient, it cannot integrate empirical knowledge, and requires a large number of sample learning to ensure the reliability of diagnosis, and there is a problem that the diagnosis model learning and training time is too long due to the large scale problem, there is no general solution to the nonlinear problem of blade cracks, and it is difficult to diagnose the position and depth of fan blade cracks

However, there is a very important problem in wavelet analysis in the fault diagnosis of fan blade cracks, which is the selection of wavelet bases: different wavelet bases will produce different results for the same problem.

Images