Machine learning makes building rocket engines easier | Science and Technology - Futurity: Research News

Methods from scientific machine learning could address the challenges of testing the stability of rocket engines, researchers report, says John Holden, Communications Strategist at Oden Institute for Computational Engineering and Sciences UT Austin.
 

Photo: laboratorio linux/Flickr

Time, cost, and safety prohibit testing the stability of a test rocket using a physical build “trial and error” approach. But even computational simulations are extremely time consuming.
A single analysis of an entire SpaceX Merlin rocket engine, for example, could take weeks, even months, for a supercomputer to provide satisfactory predictions.

Scientific machine learning is a relatively new field that blends scientific computing with machine learning. Through a combination of physics modeling and data-driven learning, it becomes possible to create reduced-order models—simulations that can run in a fraction of the time, making them particularly useful in the design setting...

But these models do more than just repeat the training simulation.

They also can simulate into the future, predicting the physical response of the combustor for operating conditions that were not part of the training data.

Although not perfect, the models do an excellent job of predicting overall dynamics. They are particularly effective at capturing the phase and amplitude of the pressure signals, key elements for making accurate engine stability predictions.
Read more...

Source: Futurity: Research News