Presentation
Fatigue Endurance Limit Model Utilizing Artificial Neural Network for Asphalt Concrete Pavements
ASCE Airfield and Highway Pavements 2019 Conference
(2019)
Abstract
Asphalt healing is directly related to the endurance limit strain and if the pavement experiences this value of strain, or lower, no fatigue damage would develop. The purpose of this paper is to extract the beam fatigue tests data that was collected under the NCHRP Project 9-44A and utilize it to create an artificial neural network predictive model (ANN) to determine the endurance limit values for conventional asphalt concrete pavements. The paper demonstrates the created ANN model architecture as well as how to utilize it to predict the endurance limit. Also, a stand-alone equation that is capable to predict the strain values, separate from the ANN model environment, was derived. The model training and validation data included 934 beam fatigue laboratory data points conducted under NCHRP Project 9-44A. The developed model is able to determine the endurance limit (strain) as a function of the stiffness ratio, number of cycles to failure, initial stiffness, and rest period. The created model had a reasonable R2 and significance values indicating the reliability of both the developed ANN model and the stand-alone equation.
Keywords
- Asphalt pavements,
- Strain,
- Neural networks,
- Beams,
- Fatigue (material),
- Fatigue tests,
- Concrete pavements,
- Asphalt concrete
Disciplines
Publication Date
2019
Location
Chicago, IL: American Society of Civil Engineers
DOI
https://doi.org/10.1061/9780784482476.005
Citation Information
Mena I Souliman and M. Isied. "Fatigue Endurance Limit Model Utilizing Artificial Neural Network for Asphalt Concrete Pavements" ASCE Airfield and Highway Pavements 2019 Conference (2019) Available at: http://works.bepress.com/mena-souliman/93/