Inverse engineering stress-dependent resilient moduli of cement-treated base materials in asphalt pavements using falling weight deflectometer

ABSTRACT

Cement-treated aggregates (CTA) and cement-treated soils (CTS) are prevalent materials for road base and subbase layers. These materials have stress-dependent resilient moduli. This study introduces a novel inverse engineering approach for determining the stress-dependent resilient moduli of cement-treated base materials. Unlike traditional methods requiring complex triaxial tests, this approach utilises falling weight deflectometer (FWD) data from asphalt pavements. It involves creating a pavement finite element method (FEM) model that combines the viscoelastic asphalt layers. An artificial neural network (ANN) model is trained using the FEM-generated dataset ($R^2$> 0.99). The nonlinear parameters are effectively determined by integrating the ANN model with a genetic algorithm (GA) for solution optimisation, thereby eliminating the need for triaxial tests. Results show that the inverse engineering approach accurately and efficiently predicts the stress-dependent resilient modulus of pavement base layers ($R^2$> 0.98). The nonlinear parameters obtained from the model are comparable to the laboratory values. Specifically, the CTA is more resilient and can withstand higher stresses without permanent deformation. Additionally, both CTA and CTS demonstrate similar sensitivity to bulk stresses, while the CTS exhibits a greater response to shear stresses compared to CTA.

KEYWORDS:

• Cement treated aggregates
• Resilient modulus
• Falling weight deflectometer
• Finite element method
• Artificial neural network

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was sponsored by Project 52108423 supported by National Natural Science Foundation of China, Start-up Research Fund of Southeast University under Grant No. RF1028623231, and Distinguished Youth Fund (Overseas) under Grant No. 1121002327 supported by National Natural Science Foundation of China.