Practical identification of favorable time windows for infrared thermography for concrete bridge evaluation
Practical identification of favorable time windows for infrared thermography for concrete bridge evaluation
复制标题
用于混凝土桥梁评估的红外热成像有利时间窗的实际识别
DOI：
10.1016/j.conbuildmat.2015.10.156
复制
发表时间：
2015
影响因子：
7.4
通讯作者：
F. Catbas
中科院分区：
文献类型：
--
作者：
Azusa Watase;R. Birgul;Shuhei Hiasa;M. Matsumoto;K. Mitani;F. Catbas
红外热成像技术是检测混凝土桥面板分层的无损检测技术之一。这些缺陷是通过捕捉混凝土表面的温度梯度来识别的。为了使该技术在损伤检测中有效，应在具有有利温度条件的特定时间窗口进行IRT检测，以在被检测表面上获得清晰的温度梯度。这项研究是一项实验性工作，检查周围环境条件的影响，在一天中的不同时间，以定位地下分层和空隙在一个较浅的深度，这是一个额外的影响因素。本研究亦试图找出周围环境条件与混凝土表面温度值之间的关系，以估计适合IRT检测环境条件的最佳时间窗。为此，专门设计了不同厚度的可重复使用的混凝土测试板，以收集热电偶传感器读数。采用多元回归分析来生成预测模型，该模型寻求环境条件与附接到目标桥梁的测试板上的温度梯度之间的关系。回归模型还利用了在不同于目标桥梁位置的另一个位置收集的传感器数据。发现传感器数据收集的最重要方面是实现测试板与混凝土桥面表面的完美接触，以获得可辨别的温度梯度。如果不满足这一条件，数据分析就会产生虚假的结果，从而得出无效的结论。另一方面，还观察到回归分析生成的预测模型遵循与传感器读数相同的模式。这使得有可能有预测方程的基础上传感器读数，以确定合适的时间窗口进行IRT检查。
Infrared Thermography (IRT) is one of the nondestructive inspection techniques to detect delaminations in concrete bridge decks. These defects are identified by capturing the temperature gradient of concrete surfaces. In order for this technique to be effective in damage detection, IRT inspections should be conducted at certain time windows with favorable temperature conditions to get clear temperature gradients on inspected surfaces. This study is an experimental work examining the effects of ambient environmental conditions at different times of a day to locate subsurface delaminations and voids at a shallow depth, which is an additional influencing factor. This study also attempts to figure out a relationship between ambient environmental conditions and the temperature values of concrete surfaces to estimate the best time window with appropriate environmental conditions for IRT inspections. To this end, specially designed reusable concrete test plates with different thicknesses were manufactured to collect thermocouple sensor readings. Multiple regression analyses were employed to generate prediction models that seek a relationship between environmental conditions and temperature gradients on the test plates attached to a target bridge. Regression models also utilized sensor data collected at another location different than the target bridge location. It was found out that the most important aspect of sensor data collection was to accomplish a perfect contact of test plates with concrete bridge deck surfaces to get discernible temperature gradients. When this condition is not met, data analyses yield spurious results leading to futile conclusions. On the other hand, it was also observed that prediction models generated by regression analyses followed the same pattern as that of sensor readings. This makes it possible to have prediction equations based on sensor readings to determine suitable time window for conducting IRT inspections.