A Fault Diagnosis Method for Rotating Machinery with Variable Working Conditions Based on Equilibrium Distribution Adaptation

Abstract

The invention belongs to the technical field of simulation analysis, and particularly relates to a method for diagnosing faults of rotating machinery under variable working conditions with balanced distribution adaptation, including acquiring fault data of rotating machinery under variable working conditions, and dividing the fault data into source domains and domains according to different working conditions. The target domain data set; predict the pseudo-label of the target domain sample through the source domain data training model, and use the class conditional distribution to approximate the conditional distribution of the target domain; use the kernel function to map the source domain and the target domain feature set to the potential feature space, use The balance factor adjusts the source domain, the target domain conditional distribution and the edge distribution weight, realizes the source domain and the target domain sample distribution difference minimization; Outputs the fault diagnosis result under the variable working condition; The present invention uses the balance factor to weigh the source domain, the target domain conditional distribution With the edge distribution weight, the sample distribution difference between the source domain and the target domain is minimized, thereby improving the efficiency and accuracy of fault diagnosis under variable working conditions of rotating machinery.

Description

technical field [0001] The invention belongs to the technical field of simulation analysis, in particular to a fault diagnosis method of rotating machinery under variable working conditions with balanced distribution and adaptation. Background technology [0002] Major rotating mechanical equipment such as aeroengine, wind turbine generator set and steam turbine generator set often operate under complex working conditions such as variable speed and variable load. Under the action of alternating load, key parts such as gears and bearings are easy to fail. In recent years, although domestic and foreign scholars have carried out a lot of research work on mechanical equipment fault diagnosis technology based on artificial intelligence such as machine learning and deep learning, in practical engineering, due to the influence of variable speed, variable load and other factors, the distribution of fault characteristics under different working conditions is inconsistent, resulting in the...

AI Technical Summary

Problems solved by technology

However, the instance migration algorithm based on TradaBoost is usually only effective when the distribution difference between domains is small, while the feature migration algorithm based on TCA only considers the edge distribution of the source domain and the target domain, and ignores the conditional distribution of the source domain and the target domain. Adaptation, only adapting to the edge distribution cannot meet the diagnostic needs of mechanical equipment under variable working conditions

Images