Correlações para determinação prévia do índice de compressão em solos brasileiros

Abstract

In Geotechnical Engineering there are uncertainties inherent to the soil due to its nature, considering that its composition is the result of non-homogeneous and highly variable matter. In addition, when subjected to loads, such uncertainties present in the preliminary design, design and during the execution of the work are intensified, as loads added to the soil influence the interaction between the particles that constitute it, and consequently, generate changes in its state of tensions. The alterations that the different types of loading cause in the soils originate the densification process, which can be understood as a compression of the saturated soil over a certain time, causing a deformation by the expulsion of water from its voids. This process triggers settlements in buildings and landfills, these in turn are known in the geotechnical environment for their complexity. For the analysis of settlements, a geotechnical investigation in the field and later in the laboratory is necessary, aiming at the knowledge of the parameters of the soil in question. The densification process occurs slowly and obtaining the results of the densification test (edometric) can take long periods of time, however, in technical feasibility studies, quick responses are required for project definitions and directions. For this, the present work aims to present solutions for preliminary projects using linear correlations in order to obtain the parameters of a more complex and expensive test (edometric) through simpler test parameters, in order to achieve quick responses and with good approximation. Through the analysis of 32 soil samples of brazilian origin, linear correlations were developed for the previous determination of the compression index (cc ) of the edometric densification test by means of parameters of other geotechnical tests of simpler realization. As a result of the analysis, linear correlation equations were obtained (used by means of an evaluation of the precision, in view of its empirical nature), in addition to the presentation of existing correlations studied by other authors.