The National Science Foundation Division of Chemical, Bioengineering, Environmental, and Transport Systems awarded Purdue University $270,522 on February 15, 2026, under the Engineering program (CFDA 47.041) to develop methods for predicting and controlling cavitation in engineering systems through combined physics-based simulation and explainable deep learning.
The award funds development of a framework integrating first-principles thermodynamic models with artificial intelligence to understand and suppress cavitation in turbulent flows. High-fidelity simulations will model formation and evolution of vapor bubbles, generating detailed datasets linking fluid motion to cavitation events. Explainable deep-learning methods will identify flow patterns most strongly influencing cavitation and establish causal links between turbulence structures and cavitation onset. The research addresses cavitation's impact on ship propellers, water turbines, medical ultrasound, and drug delivery devices, where cavitation creates noise, vibration, and damage while remaining difficult to predict. Applications span marine, energy, and biomedical technologies.
Work is performed in West Lafayette, Indiana, with a period of performance through January 31, 2029. The award includes educational and outreach efforts to train the future science and engineering workforce.