Artificial Neural Networks in Geotechnical Engineering

Das, Sarat Kumar

doi:10.1016/b978-0-12-398296-4.00010-6

Cited by 71 publications

(41 citation statements)

References 107 publications

(93 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…NetFC therefore has a sample training size of 173, while NetLLPI has a sample training size of 138, giving them a training sample size to weights ratio of 13.3 and 8.1 respectively. While, these values are considerably less than the 30 recommended by Amari they compare favourably with other Geotechnical ANN studies whose ratios rarely exceed 5, a compilation of such studies can be found in Table 10.7 in [6].…”

Section: Artificial Neural Networkmentioning

confidence: 47%

“…The sigmoid function is the activation function most commonly used in feed forward neural networks and is shown in Eq. (6). The two layer feed forward neural network used in this study was trained with the Levenberg-Marquardt backpropagation algorithm [14].…”

Section: Artificial Neural Networkmentioning

confidence: 99%

“…However, it is important to note that the training process may lead to overfitting if there is insufficient training data and the network error has been reduced too much [6]. In general, to prevent this from occurring the number of weights in the network should be far less than the number of training samples.…”

Section: Artificial Neural Networkmentioning

confidence: 99%

See 2 more Smart Citations

Automatic classification of fine-grained soils using CPT measurements and Artificial Neural Networks

Reale

Gavin

Librić

et al. 2018

Advanced Engineering Informatics

View full text Add to dashboard Cite

Soil classification is a means of grouping soils into categories according to a shared set of properties or characteristics that will exhibit similar engineering behaviour under loading. Correctly classifying site conditions is an important, costly, and time-consuming process which needs to be carried out at every building site prior to the commencement of construction or the design of foundation systems. This paper presents a means of automating classification for fine-grained soils, using a feed-forward ANN (Artificial Neural Networks) and CPT (Cone Penetration Test) measurements. Thus representing a significant saving of both time and money streamlining the construction process. 216 pairs of laboratory results and CPT tests were gathered from five locations across Northern Croatia and were used to train, test, and validate the ANN models. The resultant Neural Networks were saved and were subjected to a further external verification using CPT data from the Veliki vrh landslide. A test site, which the model had not previously been exposed to. The neural network approach proved extremely adept at predicting both ESCS (European Soil Classification System) and USCS (Unified Soil Classification System) soil classifications, correctly classifying almost 90% of soils. While the soils that were incorrectly classified were only partially misclassified. The model was compared to a previously published model, which was compiled using accepted industry standard soil parameter correlations and was shown to be a substantial improvement, in terms of correlation coefficient, absolute average error, and the accuracy of soil classification according to both USCS and ESCS guidelines. The study confirms the functional link between CPT results, the percentage of fine particles FC, the liquid limit w L and the plasticity index I P. As the training database grows in size, the approach should make soil classification cheaper, faster and less labour intensive.

show abstract

Section: Artificial Neural Networkmentioning

confidence: 47%

Section: Artificial Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Automatic classification of fine-grained soils using CPT measurements and Artificial Neural Networks

Reale

Gavin

Librić

et al. 2018

Advanced Engineering Informatics

View full text Add to dashboard Cite

show abstract

“…It should be noted that, from the optimization point of view, "learning" is equivalent to minimize the global error function [58]. With that said, neural networks are interesting because they can handle some complex problems through a powerful and flexible framework [59].…”

Section: Fcm and Anfismentioning

confidence: 99%

Hybrid Adaptive Computational Intelligence-based Multisensor Data Fusion applied to real-time UAV autonomous navigation

Paulino

Guimarães

Shiguemori

2019

View full text Add to dashboard Cite

Nowadays, there is a remarkable world trend in employing UAVs and drones for diverse applications. The main reasons are that they may cost fractions of manned aircraft and avoid the exposure of human lives to risks. Nevertheless, they depend on positioning systems that may be vulnerable. Therefore, it is necessary to ensure that these systems are as accurate as possible, aiming to improve navigation. In pursuit of this end, conventional Data Fusion techniques can be employed. However, its computational cost may be prohibitive due to the low payload of some UAVs. This paper proposes a low-cost Multisensor Data Fusion application based on Hybrid Adaptive Computational Intelligence (HACI) -the cascaded use of Fuzzy C-Means Clustering (FCM) and Adaptive-Network-Based Fuzzy Inference System (ANFIS) algorithms -that have been shown able to improve considerably the accuracy of current positioning estimation systems for real-time UAV autonomous navigation, reducing the error in approximately 300cm 2 . In addition, the proposed methodology outperformed two other Computational Intelligence methodologies -Artificial Neural Networks and Regression Models, with 18 different tested approaches -in estimating an UAV position considering the Root-Mean-Square Error against the real trajectory. The generated Fuzzy Inference System has proved to be effective in providing an improved positioning estimation with a low computational burden, about 600 times faster than the fastest embedded sensor refresh rate.

show abstract

“…Similarly, Goh (2002) reported that for liquefaction susceptibility analysis of ground using in-situ data, ANN based model is more efficient compared to available empirical models. Das (2013) presented a comprehensive review of the successful application of ANN in different geotechnical engineering problems. Das and Basudhar (2008) used artificial neural network (ANN) modelling to predict the f r of clay, but their study was limited to tropical soil of a specific region only.…”

Section: Introductionmentioning

confidence: 99%

Prediction of the residual strength of clay using functional networks

Khan

Suman

Pavani

et al. 2016

Geoscience Frontiers

View full text Add to dashboard Cite

a b s t r a c tLandslides are common natural hazards occurring in most parts of the world and have considerable adverse economic effects. Residual shear strength of clay is one of the most important factors in the determination of stability of slopes or landslides. This effect is more pronounced in sensitive clays which show large changes in shear strength from peak to residual states. This study analyses the prediction of the residual strength of clay based on a new prediction model, functional networks (FN) using data available in the literature. The performance of FN was compared with support vector machine (SVM) and artificial neural network (ANN) based on statistical parameters like correlation coefficient (R), NashSutcliff coefficient of efficiency (E), absolute average error (AAE), maximum average error (MAE) and root mean square error (RMSE). Based on R and E parameters, FN is found to be a better prediction tool than ANN for the given data. However, the R and E values for FN are less than SVM. A prediction equation is presented that can be used by practicing geotechnical engineers. A sensitivity analysis is carried out to ascertain the importance of various inputs in the prediction of the output. Ó 2015, China University of Geosciences (Beijing) and Peking University. Production and hosting by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/ licenses/by-nc-nd/4.0/).

show abstract

Artificial Neural Networks in Geotechnical Engineering

Cited by 71 publications

References 107 publications

Automatic classification of fine-grained soils using CPT measurements and Artificial Neural Networks

Automatic classification of fine-grained soils using CPT measurements and Artificial Neural Networks

Hybrid Adaptive Computational Intelligence-based Multisensor Data Fusion applied to real-time UAV autonomous navigation

Prediction of the residual strength of clay using functional networks

Contact Info

Product

Resources

About