Journal Published Online: 19 March 2014
Volume 37, Issue 3

Determination of Relationships Between Menard Pressuremeter Test and Standard Penetration Test Data by Using ANN model: a Case Study on the Clayey Soil in Sivas, Turkey

CODEN: GTJODJ

Abstract

An extensive site characterization study, consisting of 70 Menard pressuremeter tests and 77 standard penetration tests was undertaken on the clayey soil deposit of high plasticity prior to the construction of 4 Eylul Dam’s water treatment plant which provides water supply to Sivas (Turkey). It is aimed to investigate the possible relationships between standard penetration and pressuremeter parameters for the clayey soil in Sivas. To obtain the accurate empirical equations, nonlinear regression analysis between standard penetration blow counts, pressuremeter modulus, and limit pressure were undertaken. An artificial neural network model was developed for accurate estimation of corresponding pressuremeter data based on the measured values. The measured data and calculated performance values were used at the design of Levenberg–Marquardt based multi-layer perceptron artificial neural network model. Calculated values were compared to the predicted values using the statistical error analysis. The relationship between limit pressure and undrained shear strength obtained from the conducted unconfined compression strength tests is also presented. The proposed equations are compared with those obtained from previous studies. The paper concludes that the presented equations based on the nonlinear regression analysis may be valuable for similar soils and beneficial to the practitioners for comparing, evaluating and reciprocal conversion of the parameters obtained from standard penetration and pressuremeter tests.

Author Information

Agan, Celal
Deptartment Of Civil Engineering, Harran Univ., Sanliurfa, TR
Algin, Halil
Deptartment Of Civil Engineering, Harran Univ., Sanliurfa, TR
Pages: 13
Price: $25.00
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Stock #: GTJ20130123
ISSN: 0149-6115
DOI: 10.1520/GTJ20130123