International Associating for Bridge and Structural Engineering
巻, 号, ページ
pp. 955-964
出版年月
2021年9月22日
出版者
和文:
英文:
会議名称
和文:
英文:
IABSE Congress Ghent 2021- Structural Engineering for Future Societal Needs
開催地
和文:
英文:
Ghent
アブストラクト
The hammer test is generally used as one of the non-destructive methods for detecting defects such
as voids and delamination in concrete structures like tunnels and bridges. It is necessary to eliminate
human mistakes and improve quantitative analysis so that Impact Acoustics Method (IAM) was
proposed and studied. IAM helps human decision of the defective concrete parts through comparing
waveform and frequency distribution between healthy and defective parts which are taken from
sensor or microphone. Hence, artificial intelligence (AI) is expected to replace or assist human
labor inspection by quantifying the defects. This research aims to inspect defects quickly and
efficiently the only microphone through promoting a machine learning AI analysis system flow which
mainly includes neural networks. Two experiments were held to achieve the purpose.