Performance Decay Model and Preventive Maintenance of Highway Asphalt Pavement

Abstract

This work aims to study the law of performance decay of highway asphalt pavement to facilitate preventive highway maintenance. By studying the performance decay law of asphalt pavement, this work proposes to optimize the Back Propagation Neural Network (BPNN) model with the Mind Evolutionary Algorithm, and apply the optimized model to the performance prediction of highway asphalt pavement. Some sections of Xi'an city are selected as sample data to train the neural network. A new type of asphalt material emulsified asphalt is proposed for pavement construction. Finally, the maintenance suggestions of asphalt pavement are given. The results show that the optimized BPNN model has a better fitting effect. The comprehensive evaluation indexes Root Mean Squared Error, Relative Root Mean Squared Error and Mean Absolute Error of the model prediction performance are lower than those of the model before optimization, and the decline rates are 62.46%, 62.46% and 62.71%, respectively. However, the values of Nash-Sutcliffe Efficiency and R2 increase by different degrees, with the increasing rates of 55.92% and 15.42%, respectively. It reveals that the optimized BPNN has more advantages in predicting road pavement performance, with higher accuracy and lower error rate. It provides a new idea for studying the maintenance of highway asphalt pavement. The performance of new asphalt materials is better than that of ordinary asphalt materials.