Boosting US Fighter Jets – NASA Research Applies Artificial Intelligence to Hypersonic Engine Simulations

Researchers from the National Aeronautics and Space Administration (NASA) in collaboration with the US Department of Energy’s Argonne National Laboratory (ANL) have developed artificial intelligence (AI) to accelerate simulations to study the behavior of air around supersonic and hypersonic aircraft. Engines.

Fighter jets such as the F-15s regularly outperform the Mac 2 – twice the speed of sound – which is called the supersonic level. On a hypersonic aircraft with a Mac 5 and above, an aircraft can fly at more than 3,000 miles per hour.

Hypersonic speeds have been possible since the 1950s with propulsion systems used for rockets, but engineers and scientists have been working on advanced jet engine designs to make the hypersonic aircraft much less expensive than a rocket launch. Exploration, and national defense interests.

In a newly published paper, a team of NASA and ANL researchers outlined machine learning techniques to reduce the memory and cost required to perform computational fluid dynamics (CFD) simulations related to fuel combustion at supersonic and hypersonic speeds.

This paper was previously presented at the American Institute of Aeronautics and Astronautics Sitech Forum in January.

Before building and testing any aircraft, CFD simulators are used to determine how the various forces surrounding the aircraft in the aircraft interact with it. CFDs have numerical expressions that indicate the behavior of fluids such as air and water.

F-15 Breaks Sound Barrier (US Air Force)

When a plane breaks the sound barrier traveling at speeds exceeding the sound, it creates a ‘shock wave’ which causes the air around it to become hot, dense and highly compulsive, behaving very violently.

At hypersonic speeds, the generated air friction is so strong that it can melt parts of a conventional commercial aircraft.

Since air-breathing jet engines take in oxygen to burn fuel while flying, CFD simulations must cause large changes in the behavior of the air, not only around the aircraft but also when moving through the engine and interacting with the fuel.

New technologies

Although there are fan blades to push air into a conventional aircraft, on aircraft coming at speeds of Mach 3 and higher, their movement compresses the air itself. These aircraft models, known as scramjets, are important for achieving fuel efficiency levels that rocket propulsion cannot.

So, when it comes to CFD simulations on an aircraft capable of breaking the sound barrier, all of the above add a whole new level of complexity to an already computationally intense workout.

f-22
File Image: F-22 Raptor

“Because the chemistry and turbulence interactions in these engines are so complex, scientists need to develop sophisticated combustion models and CFD codes to accurately and efficiently describe the physics of combustion,” said Sibendu Som, co-author and director of the Provisional Center for Argon Studies. Center for Advanced Propulsion and Power Research.

NASA has a hypersonic CFD code called VULCAN-CFD, which is specifically intended to mimic combustion behavior in such a volatile environment.

This code uses so-called ‘flamelet tables’ where each ‘flamelet’ is a small unit of flame in the overall combustion pattern. This database contains various snapshots that burn fuel in a large collection, which takes up large amounts of computer memory to process.

Therefore, researchers at NASA and ANL are exploring the use of AI to speed up the development of aircraft to overcome obstacles, to simplify these CFD simulations by reducing intensive memory requirements and computing costs.

Hypersonic
File image: Hypersonic missile

Enumerators at ANL used a flamelight table created by argon-developed software to train an artificial neural network applicable to NASA’s VULCAN-CFD code. The supersonic engine used values ​​from the AI ​​flamelight table to learn shortcuts about determining combustion behavior in environments.

“The partnership has enhanced the efficiency of our internal VULCAN-CFD tool by influencing Argonne research efforts, enabling us to analyze fuel combustion properties at a much lower cost,” said Robert Bourle, a research scientist at NASA Longley Research. Center.

Countries around the world are competing to achieve hypersonic flight capability and an important part of this race is simulation experiments, where there is huge potential for the use of emerging technologies such as AI and machine learning (ML).

Last month, Chinese researchers led by a high-level adviser to the Chinese military on hypersonic weapon technology reported significant progress in the AI ​​system, which could autonomously design new hypersonic vehicles, according to a recent EurAsian Times report.

In addition, in February the Chinese space agency Space Transportation announced plans for tests on a hypersonic aircraft capable of traveling at 7,000 miles per hour from next year.

The company claims that their flight can fly from Beijing to New York in less than an hour.

Leave a Comment