Microcracking based rock classification using ultrasonic and porosity parameters and multivariate analysis methods

• We present a method to characterize the microcracking state of rocks.
• We present a method to classify rock in microcracking based equivalent groups.
• We use ultrasonic velocity, attenuation coefficient and porosity.
• We use principal component method to study these parameters.
• We use Cluster Analysis to classify the investigated core samples.

This work aims at presenting a new methodology, based on NDT ultrasonic techniques and water porosity measurements, to characterize the microcracking state of rocks and classify them in microcracking based equivalent groups. The measurements of ultrasonic pulse velocity, the attenuation coefficient and the porosity by water saturation under vacuum conditions make it possible to compare and validate all these techniques as good practices to classify aggregates and ornamental stones with regard to their rock matrix compactness.Beyond the fact that these developments give new approaches to assess the rock microcracking, it was shown that these parameters have a direct relationship. Indeed, the classification methodology was applied to a database containing 56 cores coming from blocks sampled in an aggregate production quarry. For these cores, ultrasonic parameters (wave velocity, attenuation and anisotropy coefficients) and porosity parameters (total water, crack and pore porosities) were measured. Two multivariate statistical methods (Principal Component Analysis and Cluster Analysis) were applied on this database to assess the relationship between all these parameters and to classify the cores into micro-structurally similar groups.The application of the setup methodology on the core database allows us to study the main correlations between the measured microcracking rock properties. On the other hand, it was shown that the method can be used as an effective way to characterize the differences in terms of microstructure between rock samples.