Rolling bearing fault diagnosis method for optimizing random forest through improved differential evolution algorithm

Abstract

The invention relates to a rolling bearing fault diagnosis method for optimizing a random forest through an improved differential evolution algorithm. Comprising the steps of optimizing a fault diagnosis model of the random forest by adopting an improved differential evolution algorithm; and performing fault diagnosis according to the model, wherein the fault diagnosis model adopting the improved differential evolution algorithm to optimize the random forest is shown in description. In the formula, Ptrain is an input feature matrix for training the random forest model, and Qtrain is a one-dimensional column vector for training the random forest model; the fault diagnosis according to the model refers to inputting an input characteristic matrix P of a rolling bearing to be subjected to fault diagnosis into the fault diagnosis model to obtain a one-dimensional column vector Q, wherein 0 in the Q represents normal, 1 represents a rolling body fault, 2 represents an outer ring fault, 3 represents an inner ring fault, and 4 represents a retainer fault. According to the invention, the improved differential evolution algorithm is used for optimizing the random forest, so that adaptive adjustment of parameters can be realized, and the model has excellent robustness and accuracy.

Description

technical field [0001] The invention belongs to the technical field of fault intelligent diagnosis, and relates to a rolling bearing fault diagnosis method for optimizing a random forest by improving a differential evolution algorithm. Background technique [0002] Artificial intelligence algorithms have been greatly developed today, and intelligent diagnosis methods are also widely used in the fault diagnosis of rolling bearings. Random forest is one of the most typical intelligent diagnosis methods. Like other algorithms, Random Forest also has hyperparameters that need to be adjusted, and there are different optimal combinations of hyperparameters for different fault diagnosis problems. It takes a lot of time to manually adjust hyperparameters, and the field of fault diagnosis usually expects the algorithm to respond as soon as possible, so it is of great practical significance to introduce optimization methods into the field of fault diagnosis. Differential evolution a...

AI Technical Summary

Problems solved by technology

However, like other optimization methods, the differential evolution algorithm is also prone to fall into local optimum, unable to find the optimal combination of hyperparameters, which makes the random forest algorithm unable to obtain high-precision fault diagnosis results

Images