Non-Destructive Evaluation of the Elastic Modulus in Large Concrete Structures

In several areas of Civil Engineering real challenges arising for the design, construction and maintenance of new structures and for the retrofitting of existing ones concern the correct selection of materials, along with their complete and reliable characterization, and the diagnostic identification of possible damages. In particular, the task of determining the deformability of Civil Engineering materials is assuming a growing importance, both in design processes and in diagnosis, monitoring and strengthening procedures. Indeed, owing to recent design and verification codes, careful quality control and accurate design and verification in serviceability conditions are required to the number of structures and infrastructures which are currently under construction in Europe: to this end, investigating the deformability of materials turns out to be crucial. Similarly, the evaluation of deformability is a primary requirement in case of restoration works, in order to guarantee the maximum compatibility between the damaged members and the restored parts, and hence to ensure the durability of repairs. Moreover, the analysis of the deformability properties of materials is essential in order to assess not only the damage level in RC structures, but also the conservation conditions of the ancient and wide European architectural and monumental heritage, which is currently subject to high risks of deterioration. In particular, the investigation on the decrease in the elastic properties generally provides significant diagnostic information for damage detection and quantification: it is well–known, indeed, that the local elastic modulus of materials can be correlated with their integrity and consequently with their damage level when subjected to static or dynamic loading and environmental effects. Therefore, a prompt observation of significant variations in the values of the elastic modulus in time or point–by–point in a structure could be associated to the presence of damage phenomena in progress and makes it possible to take appropriate safety measures. Unfortunately, determining the elastic modulus in existing structures is a complex task, especially when dealing with large structures, such as bridges, viaducts, etc. Indeed, on account of the unavoidable variability in experimental measurements and most of all due to local material inhomogeneities, the elastic modulus should be suitably modeled as a random variable, to be examined in a probabilistic context, and consequently a considerable number of experimental data should be required for a satisfactory material characterization. It stands to reason that a systematic use of standard methods which involve drilling a high number of cores to be subjected to laboratory tests would be very expensive and, in most cases, unfeasible. For this reason, in the last decades, some authors have proposed sophisticated correlations with other properties that could be more easily determined and many others have developed experimental techniques in the attempt of obtaining rapid and cost-effective NDE methodologies. Some of these techniques are based on the measurement of ultrasonic waves propagation velocity and reveal to be particularly fast and easy–to–perform, but, on the other hand, their accuracy deeply depends on the ability of the operator and on testing conditions in general. Conversely, other mechanical techniques make it possible to achieve a substantially higher accuracy, but their use is restricted to specific applications or has the drawback of requiring a longer time and causing a little damage to the structural member under investigation. Based on the foregoing considerations, the present study deals with the problem of experimentally determining the elastic modulus in existing concrete structures and appropriately processing the experimental data through statistical tools. A novel mechanical method, denoted as Impulse Method, has been proposed for the on–site estimation of the elastic modulus: it proves to be accurate, repeatable and sufficiently fast and easy–to–perform to make it possible to collect a suitable number of data in a very short time. Subsequently, the problems related to the evaluation and the practical use of such experimental data are addressed through a statistical approach based on Hypothesis Testing Theory: in particular, a sequential hypothesis testing procedure has been developed, which is optimal with respect to the error probabilities and the required number of experimental measurements. Finally, an extensive experimental campaign has been conducted at the Non–Destructive Testing Laboratory of the Politecnico di Torino, with the aim of assessing the effectiveness and the possible drawbacks of the proposed method: the results achieved proved to be fairly accurate and reliable, so that the Impulse Method reveals to have interesting possibilities of application in the field of diagnosis and monitoring, especially when used in conjunction with the proposed hypothesis testing procedure. These encouraging findings suggest to continue the research in view of the integration with other experimental techniques, with the final aim of minimizing the total number of tests to be performed for a complete material characterization. The present thesis is structured as follows: at first, the main methods currently used for the experimental determination of the elastic modulus of concrete, as well as some of the most recent methods proposed in the literature as potentially valuable techniques, will be presented and discussed, in the attempt to provide a comprehensive and critical view of the state–of–the–art in this field (Chapter 1); then, a specific technique denoted as Impulse Method is described, with regards to its physical foundations (Chapter 2); subsequently, the sequential hypothesis testing procedure to be used to process the experimental data is presented (Chapter 3); finally, the results of the experimental campaign implementing the Impulse Method and the proposed hypothesis testing procedure are reported and discussed (Chapter 4).