Nonlinear fault detection method based on semi-supervised manifold learning

Abstract

The invention relates to a nonlinear fault detection method based on semi-supervised manifold learning, which belongs to the field of electromechanical equipment fault diagnosis. The method comprises the following steps that (1) vibration signal data acquisition and preprocessing are performed on monitored electromechanical equipment, and hybrid-domain feature extraction is performed to obtain an initial sample set which represents an operating state of the equipment; (2) a semi-supervised Laplacian Eigenmap algorithm is adopted to perform manifold feature extraction on an equipment sample, so as to obtain essential manifold features sensitive to faults; and (3) an intelligent diagnosis model based on an LS-SVM (Least Squares-Support Vector Machine) is established in low-dimensional manifold feature space, so as to realize mode recognition and diagnosis decision to the operating state of the equipment faults. By using a semi-supervised manifold learning algorithm adopted by the invention, nonlinear geometric manifold features of a vibration signal sample can be effectively extracted, the fault category of the equipment operating state is judged, and the fault detection pertinence and accuracy are improved. The nonlinear fault detection method can be widely used for fault detection and diagnostic analysis of all kinds of mechanical equipment.

Description

technical field [0001] The invention relates to a method for detecting a fault of electromechanical equipment, in particular to a method for detecting a non-linear fault of electromechanical equipment. Background technique [0002] Fault detection is the key technology to ensure the safe operation of electromechanical equipment, and it is one of the research focuses in the field of mechanical fault diagnosis. With the increasingly complex structure and functions of modern equipment, the operating state of equipment has strong non-stationary and nonlinear characteristics, and the diagnostic information obtained is also more abundant, making the fault detection and diagnosis process very difficult and complicated. To meet the needs of nonlinear data analysis, some nonlinear feature extraction methods have been proposed continuously. [0003] Manifold learning is a new type of nonlinear feature extraction method developed in recent years. It aims to discover the inherent geome...

AI Technical Summary

Problems solved by technology

But they all belong to unsupervised learning methods, which do not make full use of the category information of the acquired label samples

Images