Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about innovative approaches to reduce computational costs in ocean modeling through a 55-minute technical talk presented by Professor Annalisa Quaini from the University of Houston. Explore the implementation of a data-driven reduced order model (ROM) and Large Eddy Simulation (LES) technique for simulating two-layer quasi-geostrophic equations (2QGE) in wind-driven ocean dynamics. Discover how proper orthogonal decomposition (POD) and long short-term memory (LSTM) recurrent neural networks combine to create an efficient framework for predicting ocean behavior. Follow along as the speaker demonstrates the effectiveness of this computational approach using the double-gyre wind forcing test, showcasing significant time savings while maintaining accuracy. Gain insights from Professor Quaini's extensive expertise in computational fluid dynamics and its applications in various fields, including her work as a William and Flora Hewlett Foundation Fellow at the Harvard Radcliffe Institute and her role as founding editor of Advances on Computational Science and Engineering.
