Fault diagnosis method and diagnosis system for gearbox

Abstract

The invention provides a fault diagnosis method and system for a gearbox, and the method comprises the steps: dividing an original composite fault signal of the gearbox into different frequency bands through a 1 / 3 binary tree strategy, calculating the kurtosis values and singular value negentropies of different frequency bands, so as to define an SVN-Kurtosis diagram, and selecting an optimal frequency band from the SVN-Kurtosis diagram; acquiring filtering signals according to the center frequency and the bandwidth of the optimal frequency band, making multi-point kurtosis spectrum analysis on the filtering signals, and performing separation and extraction by an MOMEDA algorithm according to the fault period interval extracted from the multi-point kurtosis spectrum; and making fault feature enhanced analysis on the result of separation and extraction by adopting a composite cross-correlation spectrum. According to the fault diagnosis method and the diagnosis system, fault diagnosis can be better realized, and important information can be fully reserved.

Description

technical field [0001] The invention relates to the field of gearbox fault diagnosis, in particular to a gearbox fault diagnosis method and a diagnosis system. Background technique [0002] In large-scale rotating machinery, the gears and bearings in the gearbox are high-failure parts, and the operating environment of the gearbox is harsh, such as wind power generation systems, often resulting in multiple faults and complex faults. If not repaired in time, it will cause damage to the mechanical system. serious hazard. Due to the mutual interference, coupling and aliasing among the fault responses of composite faults, it is difficult to separate and extract the characteristic information, so that it is impossible to distinguish the fault type. How to extract the characteristic parameters that can effectively represent the fault state of the gearbox from the composite fault signal, and increase the discrimination between fault features is the key to realize the composite faul...

AI Technical Summary

Problems solved by technology

However, the commonly used methods cannot fully consider the characteristics of the fault signal, which may easily cause the loss of important information.

Images