An Augmented Model of Rutting Data Based on Radial Basis Neural Network

The rutting depth is an important index to evaluate the damage degree of pavement. Therefore, establishing accurate prediction model can guide pavement design and provide necessary basis for maintenance. However, sample size data small, sampling not standardized, which makes it hard establish a with high accuracy. Based on RIOHTrack’s asphalt structure, this study builds reliable data-augmented model. In paper, different augmented models based Gaussian radial neural networks are constructed temperature loading pavements as main features. Experimental results show that method outperforms classical machine learning methods in augmentation, average root mean square error 3.95 R-square 0.957. Finally, training, multiple network used prediction. Compared unaugmented data, accuracy increased by 50%.