Evaluation of void beneath the airport pavement slab corner based on the strain monitoring

Abstract

ABSTRACT Void beneath the slab corner as a significant problem of airport pavement maintenance is always paid more attention by airport operators. The objective of this paper is to establish neural network models to predict the void sizes and load position based on the monitored strain values. A finite element model was established and validated by the data collected from a full-scale test pavement. Pavement strain values of 3168 scenarios considering different pavement structure parameters, load positions and void sizes were calculated. The simulated bottom strains at concrete pavement slabs were analysed for different void sizes and loading conditions. A coefficient and its warning thresholds were proposed to predict the occurrence of voids. Then, two Genetic Algorithm Backpropagation (GA-BP) neural network models were developed to predict void sizes and load position using the numerical simulation results as the training dataset.