2023 Fiscal Year Final Research Report
Construction of digital twin nondestructive evaluation for laser ultrasonic visualization testing
Project/Area Number |
21K04231
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Multi-year Fund |
Section | 一般 |
Review Section |
Basic Section 22020:Structure engineering and earthquake engineering-related
|
Research Institution | Gunma University |
Principal Investigator |
|
Project Period (FY) |
2021-04-01 – 2024-03-31
|
Keywords | 超音波非破壊評価 / 時間反転法 / 深層学習 / 機械学習 / レーザー超音波可視化試験 / 数値シミュレーション |
Outline of Final Research Achievements |
This study aimed to enhance the advanced laser ultrasonic non-destructive evaluation method, which is one of the maintenance techniques for structures, by introducing the concept of digital twins. In this study, various inverse analysis techniques were developed, including advanced ultrasonic simulation techniques in a virtual space created by digital twins, time-reversal methods that retransmits received ultrasonic waves in the virtual space by time-reversing actual measured waveforms, and deep learning techniques such as style transfer, which serves as the basis for creating AI, to estimate the position, size, and shape of defects inside structures and materials. These developed methods were implemented using data such as received ultrasonic waveforms obtained from actual laser ultrasonic non-destructive evaluation, demonstrating the effectiveness and validity of the proposed approach.
|
Free Research Field |
応用力学,計算力学,非破壊評価
|
Academic Significance and Societal Importance of the Research Achievements |
近年,構造物内部の欠陥の有無等を検査する手法である非破壊検査に注目が集まっている.特に,土木構造物は,検査範囲が広範囲であり,橋梁裏側やトンネル上部等,検査しずらい箇所が検査対象となる場合も多い.そのため,検査を制御し非接触で実施できれば,検査効率を大幅に向上できる.また,検査の際には,検査対象内部の欠陥の有無,大きさ,位置等を検査員が推定することとなるが,その推定精度は,検査員の熟練度に依存するため,定量的に評価できる方法も検討できれば有意義であろう.そこで本研究では,非接触で広範囲な検査に対応可能なレーザー超音波可視化試験にデジタルツインやAIを導入し,検査を容易に実行する方法を開発した.
|