논문 상세보기
pp.141-142
이 논문에서는 기존의 초음파 감쇠량을 측정하는 기법1과 이를 이용하여 콘크리트 내부 공극의 크기 분포를 추정2하는 논문을 요약하고자 한다.
초음파 속도법 등 일반적으로 사용하고 있는 초음파 탐촉자를 이용한 실험에서는, 탐촉자와 시편 사이의 커플링(coupling) 문제로 인해 시편의 감쇠량을 정량적으로 측정하는 데 한계가 있다. 따라서 감쇠량에 대한 신뢰성있는 측정을 위해, Fig. 1과 같은 실험 장치를 고안하여 개발하였다. 가장 중요한 특징으로는 기존의 탐촉자 앞에 얇은 PZT 세라믹을 배치하여 자기보상 주파수 응답함수 기법(self-compensating frequency response function, SC-FRF)를 적용하였다. 이 기법을 이용하여 시편의 고유한 감쇠량을 측정하면, Fig. 2와 같이 주파수에 따른 변화를 측정할 수 있다. 이를 Roney 식과 같은 매질 내 삽입된 입자 분포에 의한 초음파 산란과 비교하면, 시편의 공극에 대한 크기 분포를 Table 1과 같이 정량적으로 추정할 수 있다.
The dimension of air voids in concrete is one of the most important parameters to evaluate the resistance to freezing-and-thawing damage, but it is not easy to obtain the three-dimensional information in an existing structure. This paper summarizes the previous study for developing a ultrasonic wave attenuation measurement technique and its application for air voids characterization. The wave attenuation mainly due to inclusions in a medium can be used to characterize the volumetric distribution of air voids in a cement-based material. The measurement of cement paste indicates that the voids volume fraction is a dominant factor in the increase of high-frequency wave attenuation. A severe increase of wave attenuation was observed for a cement paste having a permeable voids volume fraction of 34%. The corresponding air voids volume evaluated by the proposed technique was 8.2%. The proposed technique analyzes the measured wave attenuation based on the modified Roney model, and finally obtains the volume fraction and size distribution of air voids via a nonlinear optimization for prediction error. The obtained voids volume fractions of tested samples were compared with the permeable voids volumes measured by a conventional test. Their size distributions were also compared with those evaluated from cross-sectional images, where image analysis was performed. The goodness-offit for the distribution measured by image processing was a lognormal distribution. When a log-normal distribution was used as a basis function for the nonlinear optimization, the inversely determined first and second moments of the distribution were not substantially different from those obtained from image processing. The volume fraction also gave better agreement with the permeable voids test results when the log-normal distribution was used for the proposed inverse analysis.
