Statistical analysis of favorable conditions for thermographic inspection of concrete slabs

A Review of Infrared Thermography for Delamination Detection on Infrastructures and Buildings

“…The reason was that Clark et al [95] reported delamination on concrete bridges and masonry bridges was recognized when ∆T was more than ±0.2-0.3 • C. Additionally, Hiasa et al [159] stated that ∆T of at least 10-20 times camera's NETD allowed inspectors to distinguish delamination from thermal noise. On the other hand, several studies [24,56,89,157,168] adopted ∆T of 0.5 • C as the criterion according to ASTM [47] for bridge deck inspection. Moreover, higher ∆T was used as the criterion.…”

“…Meanwhile, Scott et al [147] suggested that the recommended time window was 12 p.m. to 3 p.m. for up to 6.5 cm deep delamination because of a time lag between the maximum solar loading at noon and thermal contrast responses. Pozzer et al [24] statistically analyzed meteorological data and thermal contrast. They predicted favorable time windows from 12 p.m. to 3 p.m. due to high solar radiation, high ambient temperature, and low pressure.…”

“…IRT is defined as a process of measuring surface temperature distribution using IR cameras and processing and interpreting the data of IR images [24]. For infrastructures and buildings, IRT is used not only for delamination detection but also a wide range of inspections: moisture [25][26][27][28][29][30][31], thermal insulation [7,[32][33][34][35], internal structure [36], cracks [37][38][39], air leakage [40,41], and cultural heritage [42][43][44][45][46].…”

Experimental investigation of subsurface defect detection in concretes by infrared thermography and convection heat exchange

“…The thickness of concrete cover or the depth of void underneath the surface was 5, 10, 15, 20 and 25 mm for models SQ100-D5 to SQ100-D25, respectively. In the experimental studies [3,11,29,31,32] on the effect of geometry and depth of subsurface defects using IRT, void-like and delamination-like defects are simulated by casting pieces of polystyrene and timber of different sizes and at various depths in the concrete. Polystyrene-made inclusions are widely used in the simulation of artificial delamination and voids because the concrete over them has shown similar temperature contrast patterns to actual defects [29,33].…”

“…The Root Mean Squared Error (RMSE) is equal to 0.369 (close to 0) and the p value is equal to 1.86 × 10 -24 . This p-value means that at 5% significance level, all the independent variables contribute to the response variable (the null hypothesis that all parameters do not contribute is rejected) [32] and this model aligns to the data significantly better than a degenerate model of a constant term [41]. This preliminary analysis includes the data from all the experiment sets to distinguish the effective terms in the regression model.…”