Intelligent Welding: Real Time Monitoring, Diagnosis, Decision and Control Using Multi-Sensor and Machine Learning

Abstract

In this project, the industry problem of real-time weld quality assurance is studied. An automated weld quality assurance can increase the efficiency and the productivity of weld manufacturing. In order to ensure an adequate weld quality, the selection of proper evaluation approaches is critical. Currently, inspections are usually conducted either destructively or in the post-weld stage. Thus, if defects are found in welded product, few of them can be remedied. This may result in the disposal of expensive material, thus decreasing overall productivity. Therefore, an efficient nondestructive weld quality monitoring method is critically needed.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Dec 31, 2018
Accession Number
AD1090879

Entities

People

  • Didem Ozevin

Organizations

  • University of Illinois at Chicago

Tags

Communities of Interest

  • Autonomy
  • Energy and Power Technologies
  • Sensors

DTIC Thesaurus Topics

  • Acoustics
  • Arc Welding
  • Assembly
  • Data Acquisition
  • Detection
  • Elastic Properties
  • Electromagnetic Fields
  • Engineers
  • Fabrication
  • Geometry
  • Manufacturing
  • Mechanical Working
  • Mechanics
  • Microelectromechanical Systems
  • Modulus Of Elasticity
  • Supervised Machine Learning
  • Wave Propagation

Readers

  • Metallurgy
  • Organizational Process Management (OPM).
  • Systems Analysis and Design

Technology Areas

  • AI & ML