You are here

Machine Learning Tools to Optimize Metal Additive Manufacturing Process Parameters to Enhance Fatigue Performance of Aircraft Components

Award Information
Agency: Department of Defense
Branch: Navy
Contract: N68335-20-C-0477
Agency Tracking Number: N20A-T002-0222
Amount: $140,000.00
Phase: Phase I
Program: STTR
Solicitation Topic Code: N20A-T002
Solicitation Number: 20.A
Timeline
Solicitation Year: 2020
Award Year: 2020
Award Start Date (Proposal Award Date): 2020-04-30
Award End Date (Contract End Date): 2020-11-02
Small Business Information
3190 Fairview Park Drive Suite 650
Falls Church, VA 22042-4549
United States
DUNS: 010983174
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Anahita Imanian
 (703) 226-4078
 aimanian@tda-i.com
Business Contact
 Scott Bradfield
Phone: (703) 226-4061
Email: sbradfield@tda-i.com
Research Institution
 Carnegie Mellon University
 Kristen Jackson
 
5000 Forbes Ave.
Pittsburgh, PA 15213-3815
United States

 (412) 268-8746
 Nonprofit College or University
Abstract

In this SBIR effort, TDA and its team partners propose to develop a comprehensive toolset based on an Integrated Computational Material Engineering (ICME) framework using Machine Learning (ML) and Artificial Intelligence (AI) algorithms to predict mechanical performance and fatigue life in additively manufactured (AM) metallic components. The toolset addresses fatigue contributing factors, including defect distribution, surface roughness, residual stress, and microstructure effects, including the influence of post-processing treatments. It provides suggested design, process, and post-process parameters to optimize fatigue and mechanical performances.   The key products from the proposed framework are: (1) combined in-situ process monitoring methods of acoustic monitoring of laser/melt pool interactions to identify flaw formation, high speed imaging to identify splat formation; and thermal imaging to identify hot spot formation, all enabled by machine vision and machine learning analyses; (2) ML and AI based algorithms to forecast fatigue contributing factors (e.g., defects distribution, surface roughness residual stress and microstructure) based on in-situ process monitoring, ex-situ process analysis and FEM models; (3) Fatigue prediction algorithms based on database built upon crystal plasticity and stored energy theories, and AI algorithms; (4) critical verification and validation experiments; and (5) surrogate based optimization toolsets.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government