dc.contributor.author |
Jebur A. |
|
dc.contributor.author |
Al-Jumeily D. |
|
dc.contributor.author |
Aljaaf A. |
|
dc.contributor.author |
Aljanabi K. |
|
dc.contributor.author |
Khaddar R. |
|
dc.contributor.author |
Atherton W. |
|
dc.contributor.author |
Alattar Z. |
|
dc.contributor.author |
Majeed A. |
|
dc.contributor.author |
Mustafina J. |
|
dc.date.accessioned |
2020-01-15T22:11:13Z |
|
dc.date.available |
2020-01-15T22:11:13Z |
|
dc.date.issued |
2019 |
|
dc.identifier.uri |
https://dspace.kpfu.ru/xmlui/handle/net/156931 |
|
dc.description.abstract |
© 2018 IEEE. The principal aim of this study was to develop and verify a new Artificial Intelligence model to predict the hyperbolic soil stress-strain parameter, namely the modulus exponent (n). To achieve the planned aim, artificial neural network was developed and trained, additionally, it targeted to provide an appropriate empirical model to predict the parameter n with high efficiency. A database of laboratory measurements encompasses total of (83) case records for modulus exponent (n). Four input parameters namely: Dry unit weight, Plasticity index, Confining stress, and Water content, are considered to have the most substantial influence on the nonlinear soil stress-train relationship parameter, which are used as individual input parameters to the developed the proposed model. Multi-layer perceptron class trained using back propagation approach in this work. The effect of several issues in relation to the proposed model construction such as artificial neural network geometry and internal parameters on the performance of the model is investigated. Information on the relative importance of the factors affecting the (n), is presented, and practical equations for its prediction are proposed. |
|
dc.subject |
Artificial-Intelligence-(AI) |
|
dc.subject |
Soil-stress-strain-parameters |
|
dc.subject |
Unconsolidated-undrained-triaxial-test |
|
dc.title |
A novel artificial neural network scheme for modelling of nonlinear soil stress-strain modulus exponent |
|
dc.type |
Conference Paper |
|
dc.collection |
Публикации сотрудников КФУ |
|
dc.relation.startpage |
62 |
|
dc.source.id |
SCOPUS-2019-SID85063091265 |
|