Planetary gear fault diagnosis method based on differential evolution probabilistic neural network

Abstract

A planetary gear fault diagnosis method based on a differential evolution neural network belongs to the field of rotating machinery fault diagnosis methods. The present invention includes the following steps: S1, determining the type of failure mode, and obtaining the vibration signal of the planetary gear through the sensor; S2, decomposing the vibration signal by using the empirical wavelet transform method; S3, using the time-frequency domain index to select the decomposed signal to form a characteristic matrix ; S4, through the t-SNE feature dimension reduction method to the feature matrix dimensionality reduction; S5, on the basis of the probabilistic neural network, a probabilistic neural network fault diagnosis model based on differential evolution optimization is proposed, and the differential evolution optimization algorithm is used to optimize the probability The smooth factor δ in the neural network is optimized, and the optimal value of δ is selected to improve the accuracy of fault diagnosis. Compared with the traditional fault diagnosis method, the present invention has higher fault diagnosis accuracy.

Description

technical field [0001] The invention relates to the field of fault diagnosis methods for rotating machinery, in particular to a fault diagnosis method for planetary gears based on a differential evolution neural network. Background technique [0002] As the most important part of rotating machinery, planetary gears are developing in the direction of small size, light weight and strong carrying capacity with the development of science and technology, making planetary gears play an important role in aerospace, mining machinery, wind power and other fields. Due to the complex industrial and mining environment and harsh environment in the actual working environment, different forms of failures occur in gears under impact loads or heavy loads. When the planetary gear breaks down, it will cause equipment downtime, and cause huge economic losses and casualties. Therefore, the research on the early fault diagnosis method of planetary gear is not only of great significance for maint...

AI Technical Summary

Problems solved by technology

However, artificial neural networks still have certain limitations. The existing mathematical theories cannot give quantitative explanations well, and require a large storage space and a large amount of calculation.

Images