Skip to main content

Context

  • Goal of Numerical Simulation and Machine Learning : Predict the behaviour/performance of a system/product/process
  • Data can be rather expensive 
  • Specific challenge of small but smart datasets
       → Frugal learning 
  • Combine simulation and ML towards hybrid modeling approaches & Digital Twins
  • Vast amount of domain prior knowledge is available  
       → Multi-source data  

What's the Objective ?

  • Develop hybrid modeling methods for more reliable and explainable models → Guide critical decisions and better anticipate
  • For manufacturing and process industries

What's at stake ?

  • AM has revolutionized the manufacturing sector - Freedom of design & potential of (nearly) zero-waste
  • Not yet widespread because numerous and complex parameters to be monitored and controlled  → Part defects are still a major issue
  • AI could help to better understand and master the AM process and tend towards in-process control, certified quality assurance and real-time closed-feedback loop 

What are the challenges ?

  • Complex and still young process
  • Improve the quality and reduce the cost of the simulations
  • Enhance industrial process mastering to avoid systematic post-production control
  • Take better decisions in engineering processes and reduce production cycles

    → Towards Process Quality Control & First Time Right 

How to use AI to better master and understand manufacturing processes by jointly exploiting numerical simulations, experimental data and additional prior knowledge ?

What are the possible solutions ?

  • "Real-time" physical fields prediction 
  • Merge heterogeneous data (experimental data, simulated data of different levels of fidelity)
  • Hybridize expert knowledge with ML via Informed Neural Networks
    → Better results and more physical solutions with less data
  • Use explainable AI to expose the correlations between process, properties and performances
  • Combine reinforcement learning with human feedback for process control 

→ Synergy / close cooperation needed between computer science, material science and process know-how

Key AI topics

  • Fields prediction
  • Physics-informed Neural Network
  • Fusion of data
  • Frugal Learning
  • Reinforcement learning

Responsible(s) for the challenge