Gearbox fault diagnosis method based on improved depth forest

Abstract

The invention discloses a gearbox fault diagnosis method based on an improved deep forest, relates to the field of mechanical equipment fault diagnosis, aims at solving the diagnosis problem caused by gearbox fault multi-mode and mode aliasing, and effectively solves the problem of calculation redundancy in the data transmission and processing process in a deep forest structure. And the diagnosis efficiency and stability are improved. The method comprises the following steps: firstly, preprocessing collected vibration signal data to obtain a data sample, and inputting the data sample into an improved multi-granularity scanning structure to complete multi-scale feature enhancement of the data; secondly, proposing a PCA-based feature dimension reduction algorithm, and optimizing the data representativeness of multi-scale features; then, in the cascade forest, cascading the dimension reduction features with output features of each level of the cascade forest; and finally, completing model training, inputting test data, and completing fault diagnosis of the gearbox. Effective features can be accurately extracted, the feature representativeness is increased, the problem of vector redundancy is solved, and the fault diagnosis accuracy and stability of the gearbox are improved.

Description

technical field [0001] The invention relates to the field of gearbox fault diagnosis, in particular to a gearbox fault diagnosis method based on improved deep forest. Background technique [0002] As the main transmission device in the mechanical system, the gearbox is mainly composed of gears, bearings, rotating shafts and other vibration components. It has the advantages of large transmission torque and compact structure. It is widely used in aviation machinery and agricultural machinery to transmit power and Change the speed. As an important component of the mechanical system, the gearbox usually works in high-speed, heavy-load and other environments, which greatly increases the probability of gearbox problems. Therefore, it is necessary to monitor the health status of the gearbox and conduct fault diagnosis research. [0003] The proposal of the deep forest model provides a new idea for the research in the field of fault diagnosis, which has become the frontier and hot ...

AI Technical Summary

Problems solved by technology

[0005] In view of the problems existing in the above-mentioned background technology, the present invention reduces the dimensionality of the transformed feature vector generated by multi-granularity scanning according to a certain ratio, and then inputs it into the cascade forest, which solves the problem of redundancy of the original deep forest vector and reduces the amount of calculation. Improves the efficiency of fault diagnosis

Images