No abstract

No abstract

No abstract

This study deals with development of artificial neural networks (ANNs) and multiple regression analysis (MRA) models for determining hydraulic conductivity of fine-grained soils. To achieve this, conventional falling-head tests, oedometer falling-head tests, and centrifuge tests were conducted on silty sand and marine clays compacted at different dry densities and moisture contents. Further, results obtained from ANN and MRA models were compared vis-à-vis experimental results. The performance indices such as the coefficient of determination, root mean square error, mean absolute error, and variance were used to assess the performance of these models. The ANN models exhibit higher prediction performance than the MRA models based on their performance indices. It has been demonstrated that the ANN models developed in the study can be employed for determining hydraulic conductivity of compacted fine-grained soils quite efficiently.Résumé : Cette étude touche le développement des réseaux de neurones artificiels (RNA) et de modèles d'analyse par ré-gression multiple (ARM) pour déterminer la conductivité hydraulique de sols fins. Des essais à charge variable conventionnels, des essais oedométriques à charge variable et des essais dans une centrifugeuse ont été réalisés sur du sable silteux et des argiles marines, compactés à des densités sèches et des teneurs en humidité différentes. De plus, les résultats obtenus par les RNA et par les modèles ARM ont été comparés aux résultats expérimentaux. Des indicateurs de performance comme le coefficient de détermination, l'erreur du moindre carré, l'erreur absolue de la moyenne et la variance ont été utilisés pour évaluer la performance de ces modèles. Les modèles RNA présentent une meilleure performance en pré-diction que les modèles ARM selon les indicateurs de performance. Ainsi, il a été démontré que les modèles RNA déve-loppés dans cette étude peuvent être utilisés pour déterminer efficacement la conductivité hydraulique des sols fins compactés.Mots-clés : réseaux de neurones artificiels, sols fins, modélisation avec centrifugeuse, essais à charge variable, conductivité hydraulique.[Traduit par la Rédaction]

No abstract

No abstract

In this study, the swell pressure versus soil suction behaviour was investigated using artificial neural networks (ANNs). To achieve this, the results of the total suction measurements using thermocouple psychrometer technique and constant-volume swell tests in oedometers performed on statically compacted specimens of Bentonite–Kaolinite clay mixtures with varying soil properties were used. Two different ANN models have been developed to predict the total suction and swell pressure. The ANNs results were compared with the experimental values and found close to the experimental results. Moreover, several performance indices such as correlation coefficient, variance account for (VAF), and root mean square error (RMSE) were calculated to check the prediction capacity of the ANN models developed. Both ANN models have shown a high prediction performance based on the performance indices. Therefore, it can be concluded that the initial soil suction is the most relevant state of suction that characterizes the potential swell pressures.

Expansive soils exhibit significantly high volumetric deformations and so pose a serious threat to stability of the structures and foundations. Thus, determination of their swelling properties (i.e. swelling potential and swell pressure) becomes essential. However, measurement of the swelling properties is time-consuming and requires special and expensive equipment. With this in view, efforts were made to develop artificial neural network (ANN) and multiple regression analysis (MRA) models that can be employed for estimating swell percent and swell pressure. To achieve this, the results of free swell tests performed on statically compacted specimens of Kaolinite-Bentonite clay mixtures with varying soil properties were used. Two different ANN (ANN-1 and ANN-2) and MRA (MRA-1 and MRA-2) models have been developed: ANN-1 and MRA-1 models for predicting swell percent and ANN-2 and MRA-2 models for predicting swell pressure. The results obtained from ANN and MRA models were compared vis-à-vis those obtained from the experiments. The values predicted from the ANN models match the experimental values much better than those obtained from MRA models. Moreover, several performance indices such as determination coefficient (R 2 ), variance account for (VAF), mean absolute error (MAE), and root mean square error (RMSE) were calculated to check the prediction capacity of the ANN and MRA models developed. The obtained indices make it clear that the constructed ANN models have shown higher prediction performance than MRA models. It has been demonstrated that the ANN models can be used satisfactorily to predict swell percent and swell pressure as a rapid inexpensive substitute for laboratory techniques.