Artificial Neural Network and Finite Element Investigation of Cracked Rotor Shaft

The rotary shaft is an important machine part. Due to cyclic loading and unloading fatigue cracks may appear in the rotary shaft. Crack generated in the rotary shaft causes sudden failure of machine parts. The presence of crack varies the dynamic properties of rotary shaft system such as an amplitud...

Full description

Saved in:
Bibliographic Details
Published in:Turkish journal of computer and mathematics education 2021-04, Vol.12 (7), p.1311-1319
Main Authors: Kekan, Abhijeet H, Kumar, B Raghu
Format: Article
Language:English
Subjects:
Amplitudes
Artificial neural networks
Crack propagation
Cyclic loads
Fast Fourier transformations
Fatigue cracks
Fatigue failure
Fourier transforms
Modal analysis
Neural networks
Resonant frequencies
Three dimensional models
Vibration analysis
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The rotary shaft is an important machine part. Due to cyclic loading and unloading fatigue cracks may appear in the rotary shaft. Crack generated in the rotary shaft causes sudden failure of machine parts. The presence of crack varies the dynamic properties of rotary shaft system such as an amplitude of vibration, natural frequency of vibration, and critical speed. In the present experiment, an experimental set up generated to identify the effect of crack severity on critical speed. Fast Fourier Transform set up used to plot frequency amplitude graph by varying crack depth. A model of a shaft with a single crack generated using 3D modelling software. Models with varying crack depth developed. Modal analysis of shaft models carried out to get natural frequencies of vibration. Variations in natural frequencies of shaft observed with respect to Severity of crack. Natural frequency and critical speed data used to identify crack location and severity using artificial neural network.
ISSN:1309-4653