In this paper, damage assessment technology based on statistical pattern recognition technology was developed for maintenance of structure and the performance of the developed technology was verified by vibration test. The damage assessment technique uses the improved Mahalanobis distance theory, which is a statistical pattern recognition technique, and developed to take account of the variability between the measured data. In order to verify the damage evaluation performance of the developed technology, a cable damage test was conducted for a cable-stayed bridge. Experimental results show that the developed damage assessment technology has the capability of extracting information that can determine the location of damage due to cable damage.

• 허광희(건양대학교 해외건설플랜트학과) | Heo Gwang-Hee
• 김용석(건양대학교 의료IT공학과) | Kim Yong-Suk
• 고병찬(건양대학교 재난안전공학과) | Ko Byeong-Chan
• 이진옥(충남대학교 토목공학과) | Lee Chin-Ok
• 김충길(건양대학교 공공안전연구소) | Kim ChungGil 교신저자