Chaos-based early-stage single point of failure detection and classification method for mechanical part

Abstract

The invention discloses a chaos-based early-stage single point of failure detection and classification method for mechanical part. The method comprises the following steps that: firstly, processing existing sample fault signals, under different states, of the mechanical part, and establishing verification intervals of different fault types; secondly, obtaining fault feature frequency correspondingto all single-point failure states of the mechanical part to construct the frequency matrix of a Duffing chaotic oscillator; thirdly, solving the critical threshold of a corresponding periodic stimulating force amplitude under different fault feature frequencies, and constructing a frequency-threshold matrix; and finally, adding a signal to be detected to calculate a maximum Lyapunov index matrixM, carrying out verification on data in the M, calculating the correlation dimension of the signal to be detected if a fault signal exists, carrying out fault classification by contrasting with the established correlation dimension interval of different fault types, carrying out fault classification, and determining a fault mode. By use of the method, the detection and the classification of the early-stage single-point failure of the mechanical part are realized, anti-noise ability if high, and in addition, a fault detection success rate is high.

Description

technical field [0001] The invention relates to a chaos-based early single-point fault detection and classification method for mechanical parts, belonging to the technical field of fault diagnosis of mechanical parts. Background technique [0002] In modern industrial production, production equipment is developing in the direction of large-scale, complex, high-speed, automated and intelligent. Not only are the different parts of each equipment related to each other and tightly coupled, but there are also close connections between different equipment. Contact, forming a complete system during the operation of the equipment. For those large and complex electromechanical equipment that is usually difficult to grasp its operating status intuitively, whether the normal operation of some key equipment can be guaranteed is directly related to all levels of an enterprise's development. Serious or even catastrophic casualties and social impact will be produced. Since complex and ad...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problems of low detection success rate and difficulty in realizing early prediction in existing methods when detecting and classifying early single-point failures of mechanical parts, and the method of directly observing the phase trajectory has low work efficiency and cannot be automatically detected problems, as well as a large number of manual frequency adjustments and high professional requirements when using the Lyapunov index method for fault classification, combined with the respective advantages of the two detection methods of the Lyapunov index and the correlation dimension, a chaos-based mechanical Early single-point fault detection and classification method for components

Images