Fast Thermal Simulation for Metal AM

The influence of temperature plays a pivotal role in various aspects of metal additive manufacturing (AM), including melt-pool dimensions, defect distribution, and microstructure formation. Consequently, understanding temperature dynamics provides valuable insights for part design, process planning, and control. However, physics-based models are computationally inefficient for inverse design and real-time applications. To overcome this, we utilize machine learning techniques to create surrogate models capable of real-time simulations while maintaining high fidelity across diverse scenarios.

Related publications

  1. JCISE
    Accelerating Thermal Simulations in Additive Manufacturing by Traing Physics-Informed Neural Networks with Randomly-Synthesized Data
    Jiangce Chen, Justin Pierce, Glen Williams, Timothy W Simpson, Nicholas Meisel, Sneha Prabha Narra, and Christopher McComb
    Journal of Computing and Information Science in Engineering, 2024
  2. JMS
    Data-driven inpainting for full-part temperature monitoring in additive manufacturing
    Jiangce Chen, Mikhail Khrenov, Jiayi Jin, Sneha Prabha Narra, and Christopher McComb
    Journal of Manufacturing Systems, 2024
  3. JMSE
    Capturing Local Temperature Evolution During Additive Manufacturing Through Fourier Neural Operators
    Jiangce Chen, Wenzhuo Xu, Martha Baldwin, Björn Nijhuis, Ton van Boogaard, Noelia Grande Gutiérrez, Sneha Prabha Narra, and Christopher McComb
    Journal of Manufacturing Science and Engineering, 2024