Early fault diagnosis method for planetary gearbox based on TEO demodulation and stochastic resonance

Abstract

The invention relates to an early fault diagnosis method for a planetary gearbox based on TEO demodulation and stochastic resonance. The early fault diagnosis method comprises the steps: first, performing empirical mode decomposition on a planetary gearbox vibration signal, selecting a component signal containing fault information, and using TEO demodulation operation to obtain a demodulated signal of the component signal; secondly, in order to satisfy a small parameter condition of a stochastic resonance system, appropriately compressing the demodulated signal and performing frequency subsampling; thirdly, with a defined output signal-to-noise ratio of the stochastic resonance system as a fitness function, optimizing structural parameters of the stochastic resonance system by using a particle swarm optimization (PSO) algorithm, and then reconstructing the stochastic resonance system; finally, re-inputting the signal into the parameter-optimized stochastic resonance system to achieve enhanced extraction of fault features. According to the signal preprocessing method based on EMD+TEO disclosed by the invention, the difficulty of fault extraction is reduced, the PSO algorithm is introduced to adaptively realize stochastic resonance driven by parameter adjustment, and the weak early fault of the planetary gearbox is efficiently extracted.

Description

technical field [0001] The invention relates to the field of early fault diagnosis of rotating machinery, in particular to an early fault diagnosis method of a planetary gearbox based on TEO demodulation and stochastic resonance. Background technique [0002] Planetary gear transmission is widely used in various industrial machinery due to its advantages of small size, large transmission ratio and strong load capacity. In some important fields, once the planetary gearbox fails, it will cause extremely serious consequences. Therefore, it is imperative to carry out fault diagnosis of planetary gearbox and explore efficient fault diagnosis methods. [0003] Extracting the fault information contained in the vibration signal is an effective method for gearbox fault diagnosis. The early fault signals of gears are often very weak and easily submerged by noise, so conventional signal processing methods fail to extract weak fault signals of gears. Aiming at the problem of extractin...

AI Technical Summary

Problems solved by technology

However, when the energy of the fault signal is extremely small and the noise dominates, the result of signal decomposition will show obvious modal aliasing, and it is difficult to effectively extract the fault signal by using this method alone

However, blind analysis technology has problems such as underdetermination and difficulty in separating non-stationary signals, which limits its practical application effect.

Images