Method and related device for fatigue life prediction of a connection based on neural networks

By using a neural network-based fatigue life prediction method, combined with a fatigue simulation system and an improved MLP neural network model, the problems of high cost and long cycle in existing technologies are solved, and fast and accurate fatigue life prediction of connectors is achieved, which is applicable to fields such as automobiles and rail transportation.

CN122113484APending Publication Date: 2026-05-29HUNAN UNIVERSITY SUZHOU INSTITUTE +1
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HUNAN UNIVERSITY SUZHOU INSTITUTE
Filing Date
2026-01-15
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

Existing technologies are costly and time-consuming in predicting the fatigue life of connectors, and cannot quickly adapt to different plate thickness combinations, thus failing to meet the needs of rapid design iteration.

Method used

By combining a fatigue simulation system and a multilayer perceptron neural network model with a neural network-based approach, and utilizing interpolation and material correction coefficients, fatigue life prediction for arbitrary plate thickness combinations can be quickly obtained. An improved MLP neural network model is then constructed to achieve fast and accurate fatigue life prediction.

Benefits of technology

It achieves efficient and accurate fatigue life prediction, shortens the R&D cycle, reduces testing costs, is applicable to multiple fields, covers different plate thickness combinations, and meets the requirements of rapid design iteration.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122113484A_ABST
    Figure CN122113484A_ABST
Patent Text Reader

Abstract

The application relates to the technical field of structure fatigue life prediction, in particular to a connecting piece fatigue life prediction method based on a neural network and related equipment. The method comprises the following steps: obtaining S-N curves of upper and lower plate materials according to material body fatigue tests; fatigue life data under corresponding loads are obtained through connecting piece fatigue tests of multiple groups of connecting pieces with different thicknesses; a fatigue simulation system is constructed according to the fatigue life data; interpolation processing is performed with the total thickness of the plates as the independent variable and the correction parameters as the dependent variable, so that the optimal correction parameter of any plate thickness combination is obtained; a training set is obtained through simulation according to the optimal correction parameter, an improved multilayer perception machine neural network model is trained through the training set, a verification set is input into the model for prediction, and the fatigue life of the connecting piece is obtained. The optimal material correction parameter combination corresponding to different plate thicknesses can be quickly determined, full coverage of different plate thickness combinations is realized, and high-precision fatigue simulation data sets can be quickly obtained.
Need to check novelty before this filing date? Find Prior Art