Computer Code for Materials Diagnosis Using Monte Carlo Method and Neural Networks

详细信息    查看全文

• 作者：Hocine Bendjama ; Djallel Mahdi
• 关键词：Non ; destructive testing ; X ; ray imaging ; Materials diagnosis ; Monte Carlo ; Neural network
• 刊名：Journal of Failure Analysis and Prevention
• 出版年：2016
• 出版时间：October 2016
• 年：2016
• 卷：16
• 期：5
• 页码：931-937
• 全文大小：604 KB
• 刊物类别：Chemistry and Materials Science
• 刊物主题：Chemistry
  Materials Science
  Tribology, Corrosion and Coatings
  Characterization and Evaluation Materials
  Mechanics
  Structural Mechanics
  Quality Control, Reliability, Safety and Risk
  
• 出版者：Springer Boston
• ISSN：1864-1245
• 卷排序：16

文摘

Non-destructive testing (NDT) is a highly valuable technique in evaluation and evolution of materials and products. X-ray imaging is an important NDT technique that is used widely in the metal industry in order to control the quality of materials. Sometimes it may be difficult to get a measurement. The simulation of X-ray imaging is often performed using computer codes. This paper presents a new simulation method for materials diagnosis. The simulation is based primarily on the X-ray attenuation law and it is performed using a combination between Monte Carlo method and multi-layer perceptron neural network. The main goal of the proposed method is to obtain more detailed information about the state of the materials.