2023 Fiscal Year Final Research Report
Autonomous nondestructive evaluation using non-contact ultrasonic measurements and machine learning
| Project/Area Number |
21H01573
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 25020:Safety engineering-related
|
|---|
| Research Institution | Osaka University |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
森 直樹 大阪大学, 大学院工学研究科, 講師 (00802092)
|
|---|
| Project Period (FY) |
2021-04-01 – 2024-03-31
|
| Keywords | 非破壊検査 / 機械学習 / 非接触超音波計測 |
|---|
| Outline of Final Research Achievements |
The non-contact ultrasonic measurement and machine learning technologies for determining abnormal noise were investigated as technologies for the IoT and automation of non-destructive inspection technology. For non-contact measurement, techniques using lasers and microphones were investigated, and for machine learning, the automatic determination of abnormal noise from acoustic data leaking from pipes using One Class-SVM was investigated. The results showed that the use of remote acoustic measurement and machine learning enabled the determination of abnormal noise with very good accuracy.
|
| Free Research Field |
超音波非破壊検査
|
| Academic Significance and Societal Importance of the Research Achievements |
非破壊検査は，高所や高温箇所，高線量下などの過酷な環境で実施されることが多い上，検査技術者は非常に高度な技能が求められる．また，そのような技術者不足が進む一方であり，検査の自動化，無人化は喫緊の課題である．その社会的背景において本研究の技術は非常に重要な位置を占めている．
また，超音波検査で得られる計測データから機械学習を用いて判定する試みは最近始まったばかりであり，超音波計測の専門家がその知識を活かしてデータの前処理を適切に行った上で異音判定を高精度で可能にした本研究は学術的にも重要である．
|
