Stress detection based on electromagnetic fusion and deep belief network optimised by a simulated annealing algorithm
Traditional methods of single-field electromagnetic detection do not meet the accuracy requirements of stress detection. In this paper, a new non-destructive testing (NDT) method based on multi-field electromagnetic signals combining the magnetic Barkhausen noise (MBN) signal and the eddy current (EC) signal is proposed. After extracting mixed features of the two electromagnetic signals, the principal component analysis (PCA) method is adopted, which reduces the multifield electromagnetic features to the intrinsic dimension of the mixed features. Then, the MBN-EC fusion features are obtained. After the signal is fused on the feature layer, it is input into the deep belief network (DBN) model optimised by the simulated annealing (SA) algorithm to accurately predict the stress value of the metal component. The accuracy of the stress detection method is verified by comparative analysis of equal-strength steel beams in the stress experiment.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
No Article Media
Document Type: Research Article
Publication date: July 1, 2019
More about this publication?
- Official Journal of The British Institute of Non-Destructive Testing - includes original research and devlopment papers, technical and scientific reviews and case studies in the fields of NDT and CM.
- Information for Authors
- Submit a Paper
- Subscribe to this Title
- Information for Advertisers
- Terms & Conditions
- Ingenta Connect is not responsible for the content or availability of external websites