Skip to content

Improved Process Parameter Optimization Using Machine Learning

  • today
  • access_time 3:00 - 3:25 PM EDT
  • location_onRoom 142
  • blur_circularConference
  • monetization_onPaid Upgrade
  • schoolOptimization

This presentation will be a briefing on the technical results of an America Makes program that was recently completed. The program was focused on demonstrating a machine learning (ML)-enabled approach to process parameter development. The project team (Senvol, Northrop Grumman, WSU-NIAR, Stratasys Direct Manufacturing, and Pilgrim Consulting) demonstrated in a side-by-side comparison that a ML-enabled approach to process parameter development was substantially more accurate and sophisticated than a traditional approach. 

In the program, there were 219,856 different possible parameter combinations to choose from. The project team used an ML approach to optimally select which parameter sets to use to best achieve multiple different performance requirements. In parallel, Northrop Grumman’s design team attempted to select optimal parameters for each performance requirement without using an ML approach; i.e., Northrop Grumman was used as a control group to demonstrate a traditional parameter development and selection approach vs. the ML approach. 

The project culminated in a validation build, where it was demonstrated that the ML-enabled approach successfully selected parameter sets that met all performance requirements, whereas the traditional approach led to the selection of parameter sets that failed to meet the performance requirements. 

This presentation will explain how the machine learning approach is different from traditional approaches, detail the technical results of this program, and walk the audience through the steps that a user can take to rapidly find optimal parameters while minimizing data generation time and costs. 

Learning Objectives:

  • Understand how the machine learning approach can drastically cut down on time and cost for parameter optimization, and how the machine learning approach differs from a traditional approach.
  • Understand a framework with which to analyze the relationships between process parameters, material properties, and mechanical performance of AM components.
  • Evaluate whether or not the machine learning approach can be used by the participant to efficiently develop process parameters for their own applications.
  • Zach Simkin
    President
    Senvol
  • Tayelor McKay
    Principal Additive Manufacturing Engineer
    Northrop Grumman Corporation — Aeronautics Systems