Prediction of the Maximum Tensile Load in Reinforcement Layers of a MSE Wall Using ANN-Based Response Surface Method and Probabilistic Assessment of Internal Stability of the Wall

Pramanik, R and Babu, GLS (2022) Prediction of the Maximum Tensile Load in Reinforcement Layers of a MSE Wall Using ANN-Based Response Surface Method and Probabilistic Assessment of Internal Stability of the Wall. In: International Journal of Geomechanics, 22 (8).

PDF int_jou_geo_22-81_2022.pdf - Published Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

The design of mechanically stabilized earth (MSE) walls depends greatly on the maximum tensile loads developed in the reinforcement layers. In practice, it has been found that the measured tensile loads significantly differ from the predicted values, and it is quantified by a bias value named load bias. It is defined as the ratio of the measured to the predicted maximum tensile load. Further, to evaluate the load bias, prediction of the maximum tensile load in reinforcements needs to be assessed properly for wall safety. Therefore, in this paper, a new artificial neural network (ANN)-based response surface method has been proposed to predict the maximum tensile load developed in reinforcements of MSE walls reinforced with steel strips. Both tensile strength and pullout limit states have been considered in this study. The sensitivity of the proposed load model on the design outcome (reliability index or probability of failure) has been assessed and compared with the existing response surface-based load model. One practical example problem has been considered, and the feasibility of the proposed model in predicting the reliability index (or probability of failure) is examined for different values of coefficient of variation of the nominal load and resistance. Design charts in terms of the failure probability of the wall over depth are presented throughout this study for both tensile strength and pullout limit states, and results reveal that the satisfactory performance of the proposed load model is achieved in predicting the reliability of the wall. © 2022 American Society of Civil Engineers.

Item Type:	Journal Article
Publication:	International Journal of Geomechanics
Publisher:	American Society of Civil Engineers (ASCE)
Additional Information:	The copyright for this article belongs to American Society of Civil Engineers (ASCE).
Keywords:	Failure analysis; Forecasting; Reinforcement; Retaining walls; Surface properties; Tensile strength; Tensile stress, Coefficients of variations; Limit state; Load bias; Load modeling; Mechanically stabilized earth wall; Network-based; Reinforcement layers; Reliability Index; Response surfaces methods; Tensile loads, Neural networks
Department/Centre:	Division of Mechanical Sciences > Civil Engineering
Date Deposited:	16 Jun 2022 10:00
Last Modified:	16 Jun 2022 10:00
URI:	https://eprints.iisc.ac.in/id/eprint/73574

Actions (login required)

View Item