A data-driven real-time fatigue life prediction method for mechanical structures

Abstract

The present invention proposes a data-driven real-time fatigue life prediction method for mechanical structures, the steps of which are as follows: first, obtain the range of exponent m and coefficient c of the corresponding Paris model in the mechanical structure, and generate a series of exponents and coefficients ;Secondly, randomly obtain observation points on the mechanical structure, for a group (m q , c q ) cycle, use the dual reciprocal boundary element method to analyze the fatigue crack growth of the observation point, obtain the displacement and real-time fatigue life information of the observation point, and form a data information pair; until all (m, c) are traversed, the data set is obtained; and then Input the data set into the BP neural network for training to obtain the BP neural network model; finally, collect the displacement of the observation point in the mechanical structure, and input the displacement of the observation point into the BP neural network model to obtain the real-time fatigue life information of the observation point. The invention can predict the fatigue life of the mechanical structure only by observing the point displacement, saving a lot of time and cost.

Description

technical field [0001] The invention relates to the technical field of fatigue crack detection and prediction of mechanical products, in particular to a data-driven real-time fatigue life prediction method for mechanical structures. Background technique [0002] In the process of mechanical material processing, defects such as voids, inclusions and cracks will inevitably appear. Under complex working conditions and cyclic loads, excessive stress concentration will occur, causing fatigue fractures, and even catastrophic consequences. According to statistics, the damage caused by fatigue fracture accounts for about 50%-90% of all mechanical damage, which is related to the defects in engineering structures. In order to prevent safety accidents and reduce economic losses, predicting the remaining fatigue life of mechanical structures has become the focus of attention. [0003] With the development of computational mechanics and computer technology, more and more numerical metho...

AI Technical Summary

Problems solved by technology

[0007] Aiming at the deficiencies in the above-mentioned background technology, the present invention proposes a data-driven real-time fatigue life prediction method for mechanical structures, which solves the problem that the fatigue life of mechanical structures cannot be predicted in the prior art, resulting in a decrease in the detection accuracy of mechanical structures. question

Images