Spatial variability assessment of structures from adaptive NDT measurements

Inspection by non-destructive testing (NDT) techniques is an effective way for assessing structures' condition state, their pathologies and locating potential critical areas. However, establishing accurate diagnoses requires numerous measurements while available budget is limited. Nevertheless, determining the spatial variability of the material properties leads fewer of the required measurements, as it characterizes zones with almost identical properties and helps identify weak areas. But assessing this spatial variability still requires numerous measurements. It should be limited by a rational criterion that could be expressed in terms of accuracy and cost minimization. In the present work, we develop an adaptive approach to characterize spatial variability of structures' material properties modelled by transformed Gaussian random fields. The methodology is performed in two steps. Firstly, the correlation length of the Gaussian random field is computed for a given accuracy. Secondly, the number of measurements is refined to estimate the first two moments of the random field with a given accuracy too. These two steps result in a minimization of the number of measurements (i.e the cost) for a target accuracy. The proposed methodology and theoretical results are illustrated using synthetic and real data.

Automated summary

Non-destructive testing techniques are effective for assessing structures' condition and pathologies, but accurate diagnoses require numerous measurements with limited budget. Spatial variability of material properties can reduce the required measurements and help identify weak areas, but it still requires substantial measurements. A rational criterion incorporating accuracy and cost minimization is necessary to limit and optimize these measurements.