ディフェクトレーダーによる配管減肉位置/形状同時評価手法確立への挑戦

2023 Fiscal Year Annual Research Report

• Project/Area Number: 23KJ0130
• Research Institution: Tohoku University
• Principal Investigator: 郭 一君 東北大学, 工学研究科, 特別研究員(PD)
• Project Period (FY): 2023-04-25 – 2024-03-31
• Keywords: Nondestructive testing / Pipe inspection / Microwave / Size evaluation

Outline of Annual Research Achievements

This research focuses on the size evaluation of pipe defect size using microwaves. The main achievements up to the current point are summarized as follows： (1)Established the simulation database for frequency spectra of microwaves from pipe wall thinning with different sizes. (2)Analyzed the relationship between the frequency spectrum and wall thinning size. (3)Developed a convolutional neural networks to evaluate the defect size according to the frequency spectrum. (4)Above results have been summarized to a manuscript, which will be submitted to a journal paper soon.
Consequently, the size evaluation of pipe wall thinning has been initially achieved using microwaves. Subsequent efforts may consider establishing a comprehensive database to assess the size of various defect shapes.