Case ID: M25-095P

Published: 2025-08-21 12:54:24

Last Updated: 1755780864


Inventor(s)

Shenghan Guo
Hasnaa Ouidadi

Technology categories

Applied TechnologiesArtificial Intelligence/Machine LearningPhysical ScienceWireless & Networking

Licensing Contacts

Physical Sciences Team

Multi-Parameter Simulation Generative Adversarial Network (MPS-GAN)

Background

Quality in manufacturing ensures the adherence of built products to the specifications set by the designer and their proper functioning when put into operation. An important factor influencing products’ quality is the contamination of process (or build) parameters used during the manufacturing process. The choice of these build conditions is crucial to the success or failure of manufacturing a particular product and thus needs to be planned both thoughtfully and efficiently. However, identifying the optimal build parameters requires considerate trial-and-error experiments, which is associated with high labor ad material costs. One practical way to leverage AI to automate the optimal designs of experiments and simulate the quality before the actual manufacturing process. .

Invention Description

Researchers at Arizona State University have developed the Multi-Conditional Generative Adversarial Network (MPS-GAN), a generative AI model designed to predict and optimize manufacturing outcomes through advanced simulation. This cutting-edge model simulates the impact of various input parameters on manufacturing processes. By synthesizing thermal and X-ray computed tomography images for the manufacturing outcome, this model provides a high-resolution, high-fidelity simulation for pre-process quality assessment. MPS-GAN’s simulation outcomes are conditional on the build parameters, thus enabling a precise evaluation of the optimality of user-selected parameter values.

Potential Applications:

  • Energy Storage & Conversion
  • Computational Materials Science & Machine Learning
  • Training & Educational Platforms

Benefits and Advantages:

  • Cost efficient – Reduces the need for costly and time-consuming experimental manufacturing trials
  • Effective – Overcomes limitations in microscopic imaging resolution for quality characterization
  • Economical – Reduces time and resource consumption in manufacturing process optimization