Gear fault diagnosis method based on improving multivariable predictive models

Abstract

The invention provides a gear fault diagnosis method based on improving multivariable predictive models. The method comprises the following steps: measuring the vibration signal of a fault object; extracting a fault characteristic value from the vibration signal, namely the instantaneous amplitude entropy of local characteristic scale decomposition; dividing the fault characteristic value into a training sample and a test sample; respectively carrying out training of multivariable predictive models based on a support vector regression machine method to the training sample to establish the best variable predictive model, and classifying the test sample according to the best variable predictive model; and distinguishing the operating state and the fault type of the fault object according to the classification result. The gear fault diagnosis method based on improving the multivariable predictive models has higher resolution in the model recognition process.

Description

technical field [0001] The invention relates to a gear fault diagnosis method, in particular to a gear fault diagnosis method based on an improved multivariate prediction model. Background technique [0002] The diagnosis process of mechanical equipment includes three parts: diagnosis information acquisition, fault characteristic information extraction and state identification. [0003] In pattern recognition, commonly used methods include decision tree cluster analysis, gray cluster analysis, and fuzzy algorithm cluster analysis. Although these methods have been applied in mechanical fault diagnosis, they lack universality and require a large amount of calculation. In addition, pattern recognition also includes rough set theory, neural network pattern recognition method and support vector regression machine (Support vector machine, SVM). Rough set theory has great advantages in dealing with fuzzy and uncertain information, but its decision rules are very unstable, the acc...

AI Technical Summary

Problems solved by technology

Rough set theory has great advantages in dealing with fuzzy and uncertain information, but its decision rules are very unstable, the accuracy needs to be improved, and it is based on a complete information system. When processing data, it often encounters data loss. ; Although the neural network pattern recognition method has strong self-organization, self-learning and nonlinear pattern classification performance, it needs a large number of typical fault samples

In addition, there are no certain standards for the selection of the structure of the neural network and the initial value setting of the weights, and often need to rely on experience or prior knowledge, which will affect its classification accuracy

Images