01661034

Component

In pavement mechanistic empirical (ME) design, the soil-water characteristic curve (SWCC) is used to estimate the resilient modulus of unbound materials at different saturation conditions. The Fredlund-Xing equation is employed in the pavement ME design to generate the SWCC and the fitting parameters of the equation are correlated to the soil physical properties using regression models. Most of the existing regression models do not have a high level of prediction accuracy. To overcome this issue, this study aims at improving the prediction accuracy of SWCC using an artificial neural network (ANN) approach. Two three-layer ANN models are constructed for plastic and non-plastic soils separately, which consist of one input layer, one hidden layer, and one output layer. The input variables include soil gradation indicators, particle diameter indicators, atterberg limits, saturated volumetric water content and climatic factors. The hidden layer has a total of 20 neurons and the output layer variables are the fitting parameters of the Fredlund-Xing equation. The SWCC database from the National Cooperative Highway Research Program (NCHRP) 9-23A project is used to develop ANN models with 80 percent of the dataset for training and 20 percent of the dataset for validation. The developed ANN models have R² values between 0.91 and 0.95 for predicting the SWCCs of unbound material, which are significantly higher than other regression models. Finally, the developed ANN models are validated by comparing a new dataset collected from both the NCHRP 9-23A project and other literature sources to the model predictions.

This paper was sponsored by TRB committee AFP60 Standing Committee on Engineering Behavior of Unsaturated Geomaterials.

18-02107

English

6p

2018

Transportation Research Board 97th Annual Meeting

Digital/other

Figures; References; Tables

Accuracy; Mathematical prediction; Moisture content; Neural networks; Regression analysis; Soil suction; Soil water

Soil water characteristic curve

Geotechnology; Highways; Hydraulics and Hydrology; Materials; Pavements

Transportation Research Board Annual Meeting 2018 Paper #18-02107

TRIS, TRB, ATRI

Jan 8 2018 10:31AM