A Neural Network-Based Rapid Backcalculation Model for Composite Pavement Systems

The study focuses on the development of artificial neural network (ANN)-based advanced backcalculation models as pavement structural analysis tools for the rapid and accurate analysis of asphalt concrete (AC) overlaid Portland cement concrete (PCC) composite pavement systems using the falling weight deflectometer (FWD) data. The DIPLOMAT model, specifically developed for the analysis of composite pavement systems, was used for obtaining deflection profiles under typical highway loadings. The DIPLOMAT solutions were then compared with the solutions of ISLAB2000 and ILLI-PAVE pavement analysis programs. ANN-based backcalculation models trained with the results from the DIPLOMAT solutions have been found to be practical alternatives for routine pavement evaluation using the FWD deflection data. These advanced models are capable of predicting AC and PCC layer moduli, along with the coefficient of subgrade reaction with very low average absolute errors. A dimensional analysis approach was also employed by introducing the dimensional terms of AC modulus over PCC modulus and PCC modulus over coefficient of subgrade reaction ratios. Both methods were verified by synthetically generated DIPLOMAT deflection profiles. ANN-based backcalculation models are capable of successfully and rapidly (analysis of 100,000 FWD deflection profiles in a single second) predicting the pavement layer moduli from the FWD deflection basins in real time during field testing. These models have successfully been validated by results from the Long-Term Pavement Performance (LTPP) FWD tests conducted on US29, Spartanburg County, South Carolina.

Language

  • English

Media Info

  • Media Type: DVD
  • Monograph Title: TRB 87th Annual Meeting Compendium of Papers DVD

Subject/Index Terms

Filing Info

  • Accession Number: 000000
  • Record Type: Publication
  • Report/Paper Numbers: 08-2416
  • Files: TRB
  • Created Date: Jan 29 2008 4:50PM