A novel method for evaluating asphalt pavement structural performance based on vehicle vibration

Abstract

ABSTRACT Fast, accurate, and high-frequency detection for evaluating pavement structural performance is vital for road maintenance decision-making. Traditional detection methods make it challenging to balance the detection speed, frequency, and cost. Pavement condition identification technology based on vehicle vibration has become increasingly mature and can overcome the above problems. This study aims to evaluate the structural performance of asphalt pavement. A vehicle-based test system based on two-axle vibration sensing was developed. The vibration signal from the vehicle driving in the asphalt pavement was generated. Further, the deep signal features was extracted based on the expression of the probability distribution. The evolutionary characteristics of probability density distribution and cumulative distribution were summarized. The main conclusions are as follows: The theoretical model analyzed the correlation between the probability density and the pavement modulus. Further, estimate the peak of the probability density (PKDE) and the cumulative distribution fitting value (CDF0). The estimated indexes’ correlation coefficients were calculated by matching the dynamic deflection index of 10 road sections. As a result, PKDE has a robust correlation with a correlation coefficient of 0.81. Therefore, the vehicle vibration probability distribution expression method proposed is effective for structural performance evaluation and has high application feasibility.