Approximating Solutions to Fluid Dynamics Problems from Constrained Datasets & Anonymous Exposure Notification: A Mobile App Intervention for Protecting Privacy and Health During COVID-19

Cristina White, Stanford University

Photo of Cristina White

Incorporating computational fluid dynamics in the design process for jets, spacecraft or gas turbine engines often is challenged by the required computational resources and simulation time, which depend on the chosen physics-based computational models and grid resolution. An ongoing problem in the field is how to simulate these systems faster but with sufficient accuracy. While many approaches involve simplified models of the underlying physics, others are model-free and make predictions based only on existing simulation data. We present a novel model-free approach that reformulates the simulation problem to effectively increase the size of constrained pre-computed datasets and introduce a novel neural network architecture (called a cluster network) with a bias suited to highly nonlinear computational fluid dynamics solutions. Compared to the state-of-the-art in model-based approximations, we show that our approach is nearly as accurate, an order of magnitude faster, and easier to apply. Furthermore, we show that our method outperforms other model-free approaches.

Additionally, for pandemic mitigation, we show that mobile technology can provide instantaneous and high-accuracy exposure notification, even between strangers, at low social and economic cost. We propose a novel protocol called CEN/TCN that performs automatic, anonymous and decentralized exposure notification using Bluetooth proximity networks. We founded Covid Watch to implement this technology as an open source nonprofit in February 2020 with a mission to fight the pandemic while defending digital privacy. Our team now has more than 500 volunteers from around the world, including advisers in public health, epidemiology, privacy, policy and law from such universities as Stanford, Waterloo, Washington, UC San Francisco and UC Berkeley. The Covid Watch team was the first to publish a white paper and develop open-source, decentralized Bluetooth exposure notification technology in early March. This was followed by the development of very similar anonymous protocols like DP3T, PACT and Google/Apple exposure notifications. Covid Watch volunteers and nonprofit staff also built a fully open-source mobile app for sending anonymous exposure notifications.

Abstract Author(s): Cristina White, James Petrie