Modelling the mechanical behaviour of unsaturated soils has been the subject of many research works in the past few decades. A number of constitutive models have been developed to describe the complex behaviour of unsaturated soils. Despite the significant advances in the constitutive theories for unsaturated soils, none of the existing models can completely describe the various aspects of the real behaviour of unsaturated soils. In this paper, a new unified approach is presented, based on the integration of a neural network and a genetic algorithm, for the modelling of unsaturated soils. In the proposed approach, a genetic algorithm was used to optimise the weights of the neural network. A three-layer sequential architecture was chosen for the neural network. The network had eight input neurons, five neurons in the hidden layer and three neurons in the output layer. The eight input neurons represented the initial gravimetric water content, initial dry density, degree of saturation, net mean stress with respect to pore-air pressure, axial strain, deviatoric stress, soil suction and volumetric strain, and the three neurons in the output layer represented the deviatoric stress, suction and volumetric strain at the end of each increment. The network was trained and tested using a database that included results from a comprehensive set of triaxial tests on unsaturated soils from the literature. The predictions of the proposed model were compared with the experimental results. The comparison of the results indicates that the proposed approach was accurate and robust in representing the mechanical behaviour of unsaturated soils. 相似文献
Modeling of karstic basins can provide a better understanding of the interactions between surface water and groundwater, a more accurate estimation of infiltrated water amount, and a more reliable water balance calculation. In this study, the hydrological simulation of a karstic basin in a semiarid region in Iran was performed in three different stages. In the first stage, the original SWAT model was used to simulate surface-water flow. Then, the SWAT-MODFLOW conjunctive model was implemented according to the groundwater characteristics of the study area. Finally, due to the karstic characteristics of the region and using the CrackFlow (CF) package, the SWAT-MODFLOW-CF conjunctive model was developed to improve the simulation results. The coefficient of determination (R2) and the Nash-Sutcliffe efficiency coefficient (NSE) as error evaluation criteria were calculated for the models, and their average values were 0.63 and 0.57 for SWAT, 0.68 and 0.61 for SWAT-MODFLOW, 0.73 and 0.7 for SWAT-MODFLOW-CF, respectively. Moreover, the mean absolute error (MAE) and root mean squared error (RMSE) of the calibration for groundwater simulation using the SWAT-MODFLOW model were 1.23 and 1.77 m, respectively. These values were 1.01 and 1.33 m after the calibration of the SWAT-MODFLOW-CF model. After modifying the CF code and keeping the seams and cracks open in both dry and wet conditions, the amount of infiltrated water increased and the aquifer water level rose. Therefore, the SWAT-MODFLOW-CF conjunctive model can be proposed for use in karstic areas containing a considerable amount of both surface water and groundwater resources.
Rockfill is the most abundant building material. It is often used for water retention under different contexts, such as dams, embankments or drainage systems. Climate change may cause water levels to rise in reservoirs. As rockfill structures are not able to resist strong overtopping flow, rising water levels will constitute a danger for rockfill dam stability as well as for people living nearby. This work is aimed at the development of an empirical formula that enables calculation of the critical water level of overflow at the crest from the geometrical and physical parameters of a dam. To achieve these objectives, several experimental tests on a rockfill dam model with two different impervious cores, moraine with a sand filter and an empty wooden formwork, were conducted in a hydraulic channel at the hydro-environmental laboratory at École Polytechnique de Montréal. The purpose of these tests was to study the initiation of a riprap failure under the influence of different variables, such as rock size, riprap bank, downstream side slope and bed slope. Results showed linear trends between the critical water level and both the downstream side slope and bed slope. Also, a power trend was observed between the critical level and riprap grain size. A formula that gives the critical overtopping water level was developed from these results. 相似文献
AbstractThe effect of pH on the physical and mechanical properties of a sediment was investigated through a set of experimental tests. The sediment was formed from deposition of suspended particles in a fluid. Two different types of clay soil were suspended in fluids with different pH (2, 4, 7, 9 and 11) in cylindrical tubes with volume of 1?liter and also in special cylindrical reservoirs. The height of the sediment was measured in the cylindrical tube until equilibrium was achieved. The sediment deposited in the reservoirs was dried in air and then Atterberg limit, compaction and unconfined compressive strength (UCS) tests were conducted on samples prepared from each sediment. The results showed that the final height of the settled sediment is a function of pH; the height of sediment is increased with increasing the pH. Also, the Atterberg limits increased with increasing the pH. The maximum dry unit weight and optimum water content decreased and increased with increasing the pH. The final strength of the sediment decreased with increasing pH. Based on the SEM analysis, it was found that the values of pH influence the properties of the formed sediments. 相似文献
The main objective of this study is to integrate adaptive neuro-fuzzy inference system (ANFIS), support vector machine (SVM) and artificial neural network (ANN) to design an integrated supervised committee machine artificial intelligence (SCMAI) model to spatially predict the groundwater vulnerability to seawater intrusion in Gharesoo-Gorgan Rood coastal aquifer placed in the northern part of Iran. Six hydrological GALDIT parameters (i.e., G groundwater occurrence, A aquifer hydraulic conductivity, L level of groundwater above sea level, D distance from the shore, I impact of the existing status of seawater intrusion in the region, and T thickness of the aquifer) were considered as inputs for each model. In the training step, the values of GALDIT’s vulnerability index were conditioned by using the values of TDS concentration in order to obtain the conditioned vulnerability index (CVI). The CVI was considered as the target for each model. After training the models, each model was tested using a separate TDS dataset. The results indicated that the ANN and ANFIS algorithms performed better than the SVM algorithm. The values of correlation were obtained as 88, 87, and 80% for ANN, ANFIS, and SVM models, respectively. In the testing step of the SCMAI model, the values of RMSE, R2, and r were obtained as 6.4, 0.95, and 97%, respectively. Overall, SCMAI model outperformed other models to spatially predicting vulnerable zones. The result of the SCMAI model confirmed that the western zones along the shoreline had the highest vulnerability to seawater intrusion; therefore, it seems critical to consider emergency protection plans for study area.