SELF-LEARNING PROCESS MODELLING IN COMPOSITES MANUFACTURING

Abstract

Aerospace composite manufacturing faces challenges in high production cost and skilled labour shortage. Machine learning technology plays a critical part in next-generation automation for composite manufacturing. This project aims to establish a framework of machine learning applications in both autoclave and out-of-autoclave manufacturing processes for aerospace composites. The study focuses on machine learning for manufacturing induced defect control, through the big data generated from physical processes and virtual simulations. Coupled with the numerical process modelling, the data of pressure mapping, cure profile and laminate layup qualities will be acquired by using high temperature pressure matt, dielectric analyser and machine vision. By harnessing the big data, the machine learning algorithms will proactively adapt appropriate process parameters to minimise the predicted defect formations. The proposed study will advance the application of artificial intelligence in aerospace composite manufacturing.

Document Details

Document Type
DoD Grant Award
Publication Date
Aug 11, 2021
Source ID
FA23862014012

Entities

People

  • Peter Schubel

Organizations

  • Air Force Office of Scientific Research
  • United States Air Force
  • University of Southern Queensland

Tags

Fields of Study

  • Materials science

Readers

  • Industrial Economics
  • Neural Network Machine Learning.
  • Reinforced Composite Materials

Technology Areas

  • AI & ML
  • AI & ML - DoD AI Strategy
  • Space