Studies on the validity of strain sensors for pavement monitoring: A case study for a fiber Bragg grating sensor and resistive sensor

To address the problems associated with the validity of pavement sensor measurement, a method of combining indoor experiments with finite element (FE) simulations for strain measurement in asphalt pavement is developed in this paper to analyze the validity of strain sensors for practical measurements. First, correlation analysis and one-way analysis of variance (ANOVA) between the simulated strain and the measured strain of the resistive (R) sensor and the fiber Bragg grating (FBG) sensor are developed, and the results show that the strain simulation of asphalt mixtures with sensors by means of FE simulation is feasible for short-term loading tests; therefore, the simulated strain without sensors is considered the true strain. The FBG sensor is more appropriate than the R sensor for use in the measurement of horizontal strain based on the stability of the regression model. Furthermore, creep experiments and FE simulations with a modified Burgers model are developed, and the results demonstrate that the FE simulation is also effective for long-term dynamic loading tests. Finally, the effect of the ratio of the modulus of the FBG sensor to that of the asphalt mixture on the stability of the regression model is analyzed, and the results suggest that the stability worsens as the modulus ratio increases at the same temperature, which could guide the selection of encapsulating materials. Moreover, the research method could also provide resources for studying the validity of sensors under more complex loading modes.