Prediction of construction performance of Beliche Dam

Abstract

Comparative predictions of the end of construction performance of a central core embankment dam located in southern Portugal are presented. The predictions are based on finite element analyses, one set being carried out at Swansea and the other at the Laboratório Nacional de Engenharia Civil (LNEC) in Lisbon. Except for collaboration over the interpretation of laboratory test results the work was independent. Different analytical models were used at the two centres: the ‘KG’ model at Swansea and one based partly on the ‘hyperbolic’ model of Duncan and Chang at LNEC. The analytical work leading to the predictions was completed by the early stages of the dam's construction thus making them class A. The Paper starts with a description of a four-year laboratory research programme at LNEC to determine the properties of the fill materials. This involved novel large-scale one-dimensional and triaxial testing. Predictions of settlements and total stresses from the two models are presented at instrument locations. Useful comparisons between the models are obtained. This Paper will be followed later with a Technical Note comparing predictions and measurements. L'article présente des prévisions comparatives du comportement à la fin de la construction d'un barrage en terre avec un noyau central au Portugal du sud. Ces prévisions sont basées sur des analyses par éléments finis effectuées d'un côté à Swansea et de l'autre côté au Laboratôrio Nacional de Engenharia Civil (LNEC) à Lisbonne. Ces deux recherches étaient indépendantes l'une de l'autre, à l'exception d'une collaboration dans l'interprétation des résultats des essais faits en laboratoire. On a utilisé des modèles analytiques différents, le modèle K–G à Swansea et un modèle basé en partie sur le modèle hyperbolique de Duncan et Chang au LNEC. Les prévisions étaient de catégorie A, parce qu'on a complété le travail analytique qui les a précédées au cours des premières étapes de la construction du barrage. L'article commence par décrire le programme de recherches dans le laboratoire qui a été suivi pendant quatre années au LNEC pour déterminer les propriétés des matériaux de remblayage. Cette procédure comprenait de nouveaux essais unidimensionnels et triaxiaux de grand envergure. Des prévisions de tassements et de contraintes totales déduites des deux modèles sont présentées pour les emplacements des instruments et des comparaisons utiles sont obtenues. On a l'intention de publier plus tard une note technique qui comparera les prévisions et les mesures.

An elastoplastic model for granular materials taking into account grain breakage

“…(i) factors which allow the degree of grain breakage to be quantified: Marsal (1967), Lee and Farhoomand (1967), Hardin (1985) and Lade and Yamamuro (1996); (ii) relations to determine the evolution of the grain size distribution: Fukumoto (1990Fukumoto ( , 1992, Mc Dowell et al (1996) and Mc Dowell and Bolton (1998); (iii) relations between the maximal friction angle and the friction angle at critical state or the interparticulate friction angle: Vesic and Clough (1968), Koerner (1970), Barton and Kjaernsli (1981), Bolton (1986), Naylor et al (1986), Maksimovic (1996); (iv) models to provide the evolution of the stress-strain relationship during loading: Loret (1982), Simonini (1996), Mc Dowell et al (1996) and Mc Dowell and Bolton (1998).…”

“…These relations are widely used. The relation proposed by Naylor et al (1986) connects the angle of friction to the confining pressure. Maksimovic (1996) also related the friction angle to the state of stress.…”

The Deformation Mechanism of a High Rockfill Dam during the Construction and First Impoundin

“…During the first impounding of central core rockfill dams, the collapse settlement inside the upstream shell can increase the settlement of rockfill shell relative to the core. for instance, during the first impounding of the Cherry Valley dam, the settlement of the upstream shell was four times greater than that of central core, which consequently led to formation of longitudinal cracks on the crest along the shell-core interface [40,41] [43][44][45][46]. Other researchers used the porous media mechanics and unsaturated soil frameworks to develop other methods with similar objectives [47][48][49].…”

Yielding of rockfill in relative humidity-controlled triaxial experiments

“…Wetting tests performed during triaxial testing (Veiga Pinto, 1983;Naylor et al 1986) also resulted in a transient collapse of the specimen, in a sudden reduction of deviatoric stress and in a permanent reduction of the shear strength. Similar results were already reported by Nobari & Duncan (1972) who tested crushed claystone from Pyramid dam.…”