Construction of Conversion Models of Observables into Indicators
Résumé
Non-destructive techniques can be used in concrete structures for detection, localization (1D, 2D or 3D) or quantification. These three sets of objectives correspond to objectives of increasing ambition:
– in the first case, it is about detecting the presence of a defect, a singularity or a local variation of property: presence of a rebar, thickness change, compactness defect, etc. It is based on the sensitivity of the non-destructive measurement to the desired change;
– in the second case, we aim to map an observable (the result of non-destructive measurement), for example the resistivity which varies according to the water content of the cover concrete, or the velocity of the radar wave. As in the first case, it exploits the sensitivity of the measurement to the expected contrast and can also take advantage of the velocity of acquisition of non-destructive measurements to rapidly map large areas;
– the third objective is more ambitious and more difficult to satisfy: it is a matter of estimating a quantity, called an Indicator (I), from the non-destructive measurement of an Observable (O). The sensitivity of the value of I to any change of O is of course necessary, but it is also necessary to have a model M which will make it possible to deduce the value of I from the measurement of O in the form I = M(O).