Development of an AI-Powered Dynamic Modulus Test with a Low-Cost Loading Frame

This project will develop an artificial intelligence (AI)-powered dynamic modulus (E*) test for measuring linear viscoelastic property of asphalt mixtures to help design durable asphalt pavements. Work in Stage 1 will focus on developing an AI-powered E* test and validating it by comparison with the standard E* test in the laboratory. At least 5 mixtures will be selected for laboratory and modeling work to validate 3-D LVE-FE model for the IDEA-CT under small displacement. The applicable range of displacement will also be determined. ABAQUS 3D-LVE-FE simulations will be employed to populate the E*-Displacement-Force-Time database for AI-Learning. AI-Learning Algorithms will be developed to Predict E* Master Curve and Shift. The database structure will be prepared and reorganized for machine learning to predict E*. About 80% of data in the database will be used for machine learning using the MATLAB toolbox while the remaining 20% data will be used to verify the machine learning algorithms. The validation of the AI-Powered E* Test will be conducted in two phases. Phase I will employ the laboratory test data collected earlier in Stage 1 while in Phase II the E* and modified IDEAL-CT tests will be performed with five new mixtures. The lab-measured E* and associated sigmoidal master curve and shift factor curve will be compared with those AI predicted values or curves. The focus of Stage 2 would be to develop a draft test procedure, standardize the hardware and software, and establish an implementation plan for the AI-powered E* test. A draft AASHTO test procedure for the AI E* test will be prepared based on the experience gained in Stage 1 and the existing AASHTO E* test procedures. Necessary adjustments to the existing loading frames will be made and data analysis algorithms finalized to calculate the E* values. The research team will work with four state DOTs (Texas, California, Iowa, and Missouri) to develop a plan for implementing the AI-powered E* test. The final report will provide all relevant data, methods, models, and conclusions along with guidance for state DOTs on how to use the developed test method in their asphalt mix design activities.

• Record URL:
  http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=5417

Language

• English

Project

• Status: Active
• Funding: $135000
• Contract Numbers:

  Project 20-30, IDEA 242

• Sponsor Organizations:

  National Cooperative Highway Research Program

  Transportation Research Board
  500 Fifth Street, NW
  Washington, DC  United States  20001

  American Association of State Highway and Transportation Officials (AASHTO)

  444 North Capitol Street, NW
  Washington, DC  United States  20001

  Federal Highway Administration

  1200 New Jersey Avenue, SE
  Washington, DC  United States  20590
• Managing Organizations:

  Transportation Research Board

  500 Fifth Street, NW
  Washington, DC  United States  20001
• Project Managers:

  Jawed, Inam

• Performing Organizations:

  Texas A&M Transportation Institute

  ,    
• Principal Investigators:

  Zhou, Fujie

• Start Date: 20230101
• Expected Completion Date: 0
• Actual Completion Date: 0

Subject/Index Terms

• TRT Terms: Artificial intelligence; Asphalt mixtures; Dynamic modulus of elasticity; Test procedures; Viscoelasticity
• Subject Areas: Design; Highways; Pavements;

Filing Info

• Accession Number: 01865730
• Record Type: Research project
• Source Agency: Transportation Research Board
• Contract Numbers: Project 20-30, IDEA 242
• Files: TRB, RIP
• Created Date: Nov 29 2022 8:53AM