Accelerating Multiphase Control: An AI Surrogate Framework

Real-time control of multiphase processes with learned operators

KVL Staff on Project

Didier Barradas Bautista
didier.barradasbautista@kaust.edu.sa
Building 1, Level 0, Office 0125

KAUST PI on Project

William Roberts
william.roberts@kaust.edu.sa

KAUST RS on Project

Paolo Guida
paolo.guida@kaust.edu.sa

Research Breakthrough

  • Bridging the Real-Time Gap: Developed an AI framework that replaces slow, traditional physics simulations with Fourier Neural Operators, enabling the first real-time control of complex multiphase flows.
  • Precision AI Decision-Making: Integrated Bayesian Optimization with neural surrogates to precisely track liquid levels in bubble columns, overcoming the mathematical “noise” that typically breaks standard control algorithms.
  • Accelerating Industrial Innovation: Reduced the computational cost of multiphase flow management by orders of magnitude, providing a scalable foundation for smarter, more efficient chemical reactors and energy systems.

🔬 Research Visualization This graph show the acceleration of the code . You can access the paperhere.

🎯 KVL's Contribution: KVL's visualization scientist Didier Barradas-Bautista contributed significantly to the implementation of algorithm through:

  • Updating the code to efficientiyl use the GPU memory
  • refactoring the code to output efficientyl the results
,