Quality Evaluation of Expressway Pavement Based on Fuzzy Neural Networks

In order to improve the precision of highway asphalt pavement quality evaluation, a comprehensive evaluation model of pavement quality was built using Takagi-Sugeno (T-S) fuzzy theory combined with back propagation (BP) neural network. In this model, 4 indexes including expressway asphalt pavement condition index, pavement structure strength index, road riding quality index, and skid resistance index are taken as input variables; a nonlinear mapping relationship of the pavement quality evaluation system is established by fuzzy inference rules; pavement detection indicators undergo the fuzzy neural network learning and training, until the error between network output and the expected output reach a minimum value; after defuzzification, quantitative quality evaluation result of each pavement is obtained. In addition, the proposed method was verified by an example using the real measured data. The results show that the method has the logical reasoning ability of fuzzy system and the quantitative data processing ability of neural network. Compared to the expected values, the pavement quality comprehensive evaluation results simulated by the proposed method have a relative error of less than 2.1%.