Modelling a rock mass in an accurate and realistic way allows researchers to reduce the uncertainty associated with its characterisation and reproduce the intrinsic spatial variability and heterogeneities present in the rock mass. However, there is often a lack of a structured methodology to characterise heterogeneous rock masses using geotechnical information available from the prospection phase. This paper presents a characterization methodology based on the geostatistical simulation of geotechnical variables and the application of a scenario reduction technique aimed at selecting a reduced number of realisations able to statistically represent a large set of realisations obtained by the geostatistical approach. This type of information is useful for a further rock mass behaviour analysis. The methodology is applied to a gold deposit with the goal of understanding its main differences to traditional approaches based on a deterministic modelling of the rock mass. The obtained results show the suitability of the methodology to characterise heterogeneous rock masses, since there were considerable differences between the results of the proposed methodology, mainly concerning the theoretical tunnel displacements, and the ones obtained with a traditional approach.
Developing the second generation of the structural Eurocodes is now reaching its final steps. The title of the revised Eurocode EN 1990 (“Basis of structural and geotechnical design”) reflects the specific identity of geotechnical problems. While the original version of the geotechnical Eurocode EN 1997 (“Geotechnical design”) was developed largely based on soil mechanics, the aim is that the revised version treats soil and rock on an equal basis. Thus, rock engineering aspects have been substantially revised and expanded. The paper presents the main changes to the Eurocodes as far as they affect rock engineering design and lists improvements that will still be necessary after this critical revision. <!----> <!----> <!----> <!---->
Este artigo explora a aplicação de métodos probabilísticos na análise da fiabilidade e no dimensionamento do reforço de taludes rochosos sujeitos a mecanismos de rotura planar. O estudo incide no caso de um maciço rochoso intersetado por uma família de descontinuidades paralela à face do talude. O volume com potencial de instabilização é idealizado como um modelo de blocos, para o qual se desenvolve uma formulação em sistemas de componentes. Como termo de comparação, analisa-se igualmente um modelo simplificado que considera o mesmo volume como um bloco único. Os resultados evidenciam a influência da inclinação do plano de deslizamento. Verifica-se que o modelo constituído por múltiplos blocos conduz a índices de fiabilidade inferiores. Contudo, a sua utilização para o dimensionamento probabilístico do reforço tem vantagens apenas para inclinações baixas do plano de deslizamento. A análise sugere ainda que o modelo de blocos múltiplos pode ser adequadamente representado por um modelo simplificado de dois blocos.
In this paper, aspects concerning the geomechanical characterization related to the case study of the underground complex of Venda Nova II built in the North of Portugal are presented. Focus is drawn to the extensive in situ characterization performed by LNEC. A statistical study was carried out in order to establish reliable confidence intervals for the parameters. A 3D model for the powerhouse complex was developed and used together with the monitored displacements and an advanced optimization algorithm to back analyze geomechanical parameters. The back analysis study validated the parameters obtained by the in situ survey and proved to be a reliable complement in this characterization.