Session: DAC-02-02: Artificial Intelligence and Machine Learning for Challenging Real-World Problems in Design Automation

Paper Number: 114741

114741 - A Heuristic Approach to Classify Geometrically Defective Bead Segments Based on Range of Sound Power, Range of Curvature and Maximum Height 

The paper presents a classification methodology for determining geometrically good and defective bead segments for the Wire Arc Additive Manufacturing (WAAM) process based on optimally determined thresholds of three parameter metrics: namely, the range of sound power, range of curvature, and maximum height. We show how a heuristic search can be performed to systematically identify the suitable threshold for the various parameter metrics for bead segment classification. The heuristic search is performed by finding a threshold value that minimizes the overlapping area between the Kernel Density Estimation (KDE)s formed by the distribution of the good and defective bead segments. We combine these thresholds to yield a three-dimensional threshold window for use in identifying geometrically defective bead segments. Experiments were conducted on the Inconel 718 material, with $660$ bead segment data collected. We verify the results of our threshold selection using real bead scans and evaluate its mean squared error (MSE) with the parabolic bead model commonly used in literature to model bead profiles. We found that our proposed approach can separate the starting and ending segments of the beads well and identify geometrically defective bead segments between them. We also observe that the MSE of geometrically defective bead segments’ is much higher than the good segments, demonstrating the validity of our approach to classifying geometrically defective bead segments.

Presenting Author: Nowrin Akter Surovi Singapore University of Technology and Design (SUTD)

Presenting Author Biography: Graduate Student

Authors:

Nowrin Akter Surovi Singapore University of Technology and Design (SUTD)
Gim Song Soh Singapore University of Technology and Design