Principal investigator: Guha Manogharan
University: The Pennsylvania State University
Industry partner: Verder Scientific, Inc.
Advanced Manufacturing through metal Additive Manufacturing (AM) systems is one of the most important sectors in the Pennsylvania Manufacturing and Energy economy. The goal of this project is to position PA AM industries as global and national leaders in rapid qualification of metal AM processing by: (1) developing a novel Artificial Intelligence (AI)/Machine Learning (ML) model to predict PSP relationship in L-PBF which will provide a unique capability to significantly, reduce lead time in qualification of metal AM, (2) furthering current-future workforce-pipeline between PSU and PA AM industry, (3) demonstrating technological viability and economic feasibility of university-developed AI/ML modeling into commercial products for the PA industries and 4) providing innovative design and manufacturing solutions with complementary tools to validate AI/ML models being developed at PSU.
The Commonwealth of Pennsylvania has many metal AM companies across small and medium enterprises (Verder Scientific, American Additive Manufacturing, Xact Metals, Barnes Group), AM powder suppliers (Advanced Powder Products Inc., Carpenter Technology, Alpha Precision Group) and AM end-users (Westinghouse, Wabtec, GE, Johnson & Johnson) which support the stable growth of the market for the metal AM. Particularly for the characterization, qualification and modeling of metal AM process, Verder Scientific in Newton, PA provides advanced characterization, elemental analysis, heat treatment and powder morphology analysis for metal AM.
Metals AM has been embraced by aerospace, defense, and medical industries for production and repair of high-value components and is estimated to constitute a market size of several billion dollars by 2025. Specifically, in Commonwealth of Pennsylvania, there are over 125 manufacturing companies that are either directly or indirectly involved in additive manufacturing.
There is a timely opportunity to develop a novel image-based AI/ML model that will be validated for metal AM part production which have the potential to revolutionize qualification and material response prediction in metal AM.