Damage Condition Assessment of Expressway Asphalt Pavement Based on RBF Neural Network

Abstract

In order to make the performance evaluation of highway asphalt pavement more scientific and reasonable, carrying out pavement maintenance management is more necessary. Taking advantage of excellent adaptability of neural network technology to deal with nonlinear mapping problem, a breakage condition evaluation model based on radial basis function (RBF) neural network is presented. This model considers four main affecting factors including pavement rut condition, crack condition, pit slot condition and repair condition. Certain number of sample data is chosen to train and simulate the RBF neural network model. The tests results, accordant with expectation, indicate that the model is qualified for practical engineering applications.