Optimization of distributed fiber optic sensors for pavements by combining DEM-FDM coupled numerical simulation method and response surface method

Abstract

The cooperative deformability of sensors embedded into a host material and the measurement accuracy of these sensors affect pavement health monitoring. In consideration of these concerns, this study combined numerical simulation and the response surface method to optimize sensor design and standardize the distributed fiber optic sensors (DFOSs) used in pavement engineering. Numerical models were developed by coupling the discrete element method and the finite difference method. First, on the basis of the discrete-continuous coupled numerical simulation method, two working conditions of unembedded and embedded DFOSs in asphalt mixtures were set up. Virtual specimen models were established, and the numerical simulation method was verified by implementing the indoor four-point bending beam test. Second, the effects of DFOSs on the physical field of asphalt mixtures were analyzed by comparing the two conditions, and optimized evaluation indexes of DFOSs for road use were proposed based on the stress, displacement, and strain difference degrees in consideration of cooperative deformation and measurement accuracy. Last, the Box-Behnken response surface method was used to optimize the cross-section size and elastic modulus of the encapsulation material of the DFOSs, and the optimal design parameter range of the DFOSs was determined. In summary, this study revealed the cooperative deformation law between embedded sensors and asphalt mixtures to facilitate the development and design of DFOSs with high measurement accuracy for pavement applications.