Application of Metal Magnetic Memory Testing Technology in Pipeline Defect

Autor: Shujun Liu, Dean He, Ming Jiang
Rok vydání: 2018
Předmět:
Zdroj: DEStech Transactions on Engineering and Technology Research.
ISSN: 2475-885X
DOI: 10.12783/dtetr/ecame2017/18393
Popis: The metal magnetic memory testing technology can detect the stress concentration area of ferromagnetic materials, and then diagnose the micro defects and early damage. Based on the principle of artificial neural network, 3 single output three layer BP neural networks are designed by using metal magnetic memory testing technology. In this paper, the stress concentration, crack and other pipeline defects are detected and identified. The experimental results show that the recognition rate of pipeline defects is 97.5%.
Databáze: OpenAIRE