Calculation of surface settlements caused by EPBM tunneling using artificial neural network, SVM, and Gaussian processes   总被引：2，自引：2，他引：0  

Ibrahim Ocak  Sadi Evren Seker 《Environmental Earth Sciences》2013,70(3):1263-1276

Increasing demand on infrastructures increases attention to shallow soft ground tunneling methods in urbanized areas. Especially in metro tunnel excavations, due to their large diameters, it is important to control the surface settlements observed before and after excavation, which may cause damage to surface structures. In order to solve this problem, earth pressure balance machines (EPBM) and slurry balance machines have been widely used throughout the world. There are numerous empirical, analytical, and numerical analysis methods that can be used to predict surface settlements. But substantially fewer approaches have been developed for artificial neural network-based prediction methods especially in EPBM tunneling. In this study, 18 different parameters have been collected by municipal authorities from field studies pertaining to EPBM operation factors, tunnel geometric properties, and ground properties. The data source has a preprocess phase for the selection of the most effective parameters for surface settlement prediction. This paper focuses on surface settlement prediction using three different methods: artificial neural network (ANN), support vector machines (SVM), and Gaussian processes (GP). The success of the study has decreased the error rate to 13, 12.8, and 9, respectively, which is relatively better than contemporary research.  相似文献   

Evaluation of hydrogeochemical parameters and quality assessment of the groundwater in Kottur blocks, Tiruvarur district, Tamilnadu, India     

Thirunavukkarasu Ramkumar  Senapathi Venkatramanan  Irudhayanathan Anithamary  Sheik Mohamed Syed Ibrahim 《Arabian Journal of Geosciences》2013,6(1):101-108

A study was conducted to evaluate the water quality of Kottur block, Thiruvarur district, Tamilnadu. Groundwater samples from hand pumps and tube wells of 16 stations were analyzed during postmonsoon and premonsoon (2008) with the help of standard methods of APHA (1995). Dominance of cations are in the following order Na>Ca>K>Mg and Cl>SO₄>HCO₃>NO₃ by anions in both the seasons. The analytical results shows higher concentration of total dissolved solids, electrical conductivity, sodium, chloride, and sulfate which indicate signs of deterioration but values of pH, calcium, magnesium, and nitrate are within permissible limit as per World Health Organization standards. From the Piper trilinear diagram, it is observed that the majority of groundwater samples are Na-Cl and Ca-Mg-SO₄ facies clearly indicates seawater incursion. In Wilcox diagram, most of the samples fall in low to very high sodium hazard and low to very high salinity hazard indicates moderately suitable for agricultural activities. Kelly’s ratio and magnesium ratio indicates most of the samples fall in suitable for irrigation purpose.  相似文献   

On the relation between excess hydraulic head and flow in a conduit imbedded within a porous matrix     

David E. Loper  Ibrahim A. Eltayeb 《地球物理与天体物理流体动力学》2013,107(3):281-297

This article provides a partial answer to the question “What is the relation between excess hydraulic head and volume flux of water in a conduit within a porous matrix?”, focusing on the case that the forcing is steady. The conduit is modelled as a horizontal circular cylinder, imbedded within a porous matrix of rectangular cross section, having constant head prescribed on the sidewalls and being confined top and bottom. Laminar flow in the matrix is assumed to obey Darcy's law, while turbulent flow in the conduit is quantified using the Darcy–Weisbach equation. Analysis of the latter equation shows that the length scale of variations in the direction of the conduit is large compared with the scale of lateral and vertical variations. This permits separation of the full three-dimensional non-linear problem into a two-dimensional linear problem for head within the matrix and a one-dimensional non-linear problem for head within the conduit. Analytic solutions are obtained for the distribution of head in the matrix and in a conduit of either infinite or finite length. In both cases, the volume flux of water is proportional to the excess head to the 2/3 power, the conduit radius to the 5/3 power, the matrix permeability to the 1/3 power and gravity to the 1/3 power. The scale of variation of head along the conduit is proportional to the excess head to the ?1/3 power, the conduit radius to the 5/3 power, the matrix permeability to the ?2/3 power and gravity to the 1/3 power.  相似文献   

On the potential of Bartlett Lewis rectangular pulse models for simulating rainfall in Peninsular Malaysia     

Ibrahim Suliman Hanaish  Kamarulzaman Ibrahim  Abdul Aziz Jemain 《水文科学杂志》2013,58(8):1690-1703

Abstract

The applicability of two versions of the Bartlett Lewis rectangular pulse model, the original and the modified model, is discussed for describing the temporal and spatial variation of rainfall patterns observed at 15 raingauge stations in Peninsular Malaysia over the period 1971–2008; 17 different sets of moment combinations are fitted to these models based on the generalized method of moments approach. The common statistics included in all sets are the mean, variance, lag-1 autocorrelation and the probability of dry based on the hourly rainfall data. The analysis was carried out on hourly rainfall data from all 15 stations for all months of the year. Two stations, Petaling Jaya and Kemaman, located on the west and east coasts of the Peninsula, respectively, are considered for illustration of the results, taking the months of July and November, which correspond to the driest and wettest months, corresponding to the southwest monsoon (May–August) and northeast monsoon (November–February), respectively. The best moment combination found for the illustrative results is based on the common statistics, as well as the mean and variance based on 24-h aggregated rainfall data, the inclusion of which successfully improved the model performance; the errors were significantly reduced. It was also found that the performance of the fitted models based on the mean absolute deviate error varies according to the type of Bartlett Lewis model applied: errors are much smaller for the fitted model based on the modified model as compared to the original model. In addition, the fitted statistics: mean, lag-1 autocorrelation and probability of dry are quite well fitted for several aggregated time scales; however, the variances are underestimated in both models for all aggregated time scales, particularly in the case of the original model. The results of extreme value analysis indicate that the modified model failed to reproduce the annual hourly and daily rainfall extremes satisfactorily.