ANN-Based Crack Identification in Rotor System with Multi-Crack in Shaft

Abstract:

Article Preview

Rotating machinery, such as steam turbo, compressor, and aeroengine etc., are widely used in many industrial fields. Among the important rotor faults, the fatigue crack fault, which can lead to catastrophic failure and cause injuries and severe damage to machinery if undetected in its early stages, is most difficult to detect efficiently with traditional methods. In the paper, based on the truth of the change of the mode shapes of the cracked structure, a new method by combining accurate finite element model of rotor with multi-crack in shaft and artificial neural network (ANN) is proposed to identify the location and depth of cracks in rotating machinery. First, based on fracture mechanics and the energy principle of Paris, the accurate FE model of the rotor system considering several localized on-edge non-propagating open cracks with different depth, is built to produce the specific mode shapes. Then a set of different mode shapes of a rotor system with localized cracks in several different positions and depths, which will be treated as the input of the designed ANN model, can be obtained by repeating the above step. At last, with several selected crack cases, the errors between the results obtained by using the trained ANN model and FEM ones are compared and illustrated. Meanwhile, the influences of crack in the different position on the identification success are analyzed. The method is validated on the test-rig and proved to have good effectiveness in identification process.

Info:

Periodical:

Key Engineering Materials (Volumes 353-358)

Edited by:

Yu Zhou, Shan-Tung Tu and Xishan Xie

Pages:

2463-2466

DOI:

10.4028/www.scientific.net/KEM.353-358.2463

Citation:

T. Yu et al., "ANN-Based Crack Identification in Rotor System with Multi-Crack in Shaft", Key Engineering Materials, Vols. 353-358, pp. 2463-2466, 2007

Online since:

September 2007

Export:

Price:

$35.00

In order to see related information, you need to Login.

In order to see related information, you need to Login.