Permanent Deformation Process of Asphalt Concrete Pavement Based on Neural Network Modeling

Abstract:

This paper numerically simulates the deflection response of layers on the cross section of a medium-strength subgrade (MFC) flexible pavement under repeating load, by a radial basic function (RBF) neural network model. The RBF modeling focuses on the functional relationship between the local points in the top deflection curves of pavement layers. The input and output data of the RBF model utilizes the last deflection profiles on the tops of four layers in the test. The deflection curve of the pavement surface is set as the input data since its developing process can been watched and measured in the test. The deflection curves of the other three layers are as the output data, because their deflection process was invisible in the test. Thus, the deflection process of the pavement layers invisible in the test can be simulated by the trained RBF neural network model, which results in a further analysis based on the obtained simulation data.

• Periodical: Applied Mechanics and Materials (Volumes 341-342)
• Pages: 1486-1490
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.341-342.1486
• Online since: July 2013
• Authors: Fu Cheng Yin, Guang Chun Zhou
• Keywords: Asphalt Concrete, Permanent Deformation, RBF Neural Network Model, Simulation

References

[1] G. Navneet , et. al. Report: Posttraffic Testing at the National Airport Pavement Test Facility: Test Item MFC. U.S. Department of Transportation, Federal Aviation Administration , (2001).

[2] D. Donovan and E. Tutumluer. Analysis of NAPTF Trafficking Response Data for Pavement Foundation Deformation Behavior. In Proceedings of FAA Worldwide Airport Technology Transfer Conference, Atlantic City, NJ, (2007).

[3] J. Dona and S. Beena, Investigation of the Performance of Flexible Airport Pavements under Moving Aircraft Wheel Loads with Wander Using Finite Element Analysis. Proceedings of GeoHunan International Conference, Challenges and Recent Advances in Pavement Technologies and Transportation Geotechnics, Changsha, Hunan, China , 2009, 44~49.

DOI: 10.1061/41046(353)7

[4] S. J. Bredenhann, K. J. Jenkins , A. J. Bester, A Practical Approach to Non-Linear Structural Analysis of Pavements Using Finite Element Analysis Techniques. 3rd International Symposium on 3D Finite Element for Pavement Design and Research, Amsterdam , (2002).

[5] M. Sungho, Nonlinear finite element analysis of pavements and its application to performance evaluation. PhD Thesis, Department of Civil Engineering, North Caralina State University , (2003).

[6] P. Saad, H. Mitri, and H. Poorooshash, Three-Dimensional Dynamic Analysis of Flexible Conventional Pavement Foundation. Journal of Transportation Engineering, ASCE, 2005, 131(6): 460~467.

DOI: 10.1061/(asce)0733-947x(2005)131:6(460)

[7] Y. Hao, S. Mansour, K. Tanmay and S. Shelley, The effect of loading time on flexible pavement dynamic response: a finite element analysis. Mech Time-Depend Mater , 2007, 11: 265~288.

DOI: 10.1007/s11043-008-9047-8

[8] X. Ma, Z. Y. Guo, Q. Yang, 3D Finite Element Analysis on Asphalt Pavement Dynamic Response under Moving Load. Journal of Chongqing Jiaotong University (Natural Science), 28(4): pp.702-708, (2009).

[9] C. J. Darken, J. Moody, Fast learning in networks of locally-tunned processing units. Journal of Neural Computation , 1989, 1: 281~294.

DOI: 10.1162/neco.1989.1.2.281