by the High Performance Computing Modernization Office

San Diego, CALIF. — The V-22 crew has just completed its search and rescue mission and is now ready to return to the ship. As the V-22 approaches, the controller gives the pilot clearance to land. The pilot brings the rotorcraft to a hover and begins to descend. Suddenly, the V-22 encounters some unexpected turbulence. The pilot, having practiced this situation many times in a simulator, lands the rotorcraft safely.

Although not currently available, this simulation scenario could be played out in the future based on research conducted by Susan Polsky and Chris Bruner, Naval Air Warfare Center, Aircraft Division, Patuxent River, Maryland. Ms. Polsky and Mr. Bruner are using high performance computing resources to analyze the effects the air wake generated by a ship has on hovering and maneuvering rotorcraft. One of the objectives of their work is to develop a database of important air wake conditions so these conditions can be incorporated into a manned flight simulator. The simulation environment can then be used to train pilots and develop ship flight envelopes without the expensive at-sea trials with real aircraft, crew, and ships.

Another objective is to determine how the air wake affects the rotorcraft along the departure and approach paths. Because tiltrotor aircraft operate from smaller ships that tend to be closer to the action, the ships do not always have the luxury of moving in the same direction as the wind. Also, each ship has a number of different landing spots, and the safe conditions – the angle and speed of the wind – for each may be different.

When the V-22 lands on the bow of a ship, it will have to first descend through air that is flowing from bow to stern. As it continues down it must essentially descend through dead air, and finally as it approaches the deck, the air is flowing from stern to bow. To determine the air wake affects, they used the computational fluid dynamics (CFD) code Cobalt, which was developed as part of the High Performance Computing Modernization Program’s Common High Performance Computing Software Support Initiative. The nature of ship air wake flows requires that time- accurate modeling techniques be used in order to get a good representation of the flow field characteristics and dynamics. The CFD grids must be fairly dense so those important features such as vortices don’t prematurely fizzle out and this, along with time-accuracy, means that these calculations require a lot of memory and a lot of compute cycles. Therefore, high performance computing resources are used so that the problem can be computed in a reasonable amount of time.

The numerical analysis revealed features in the ship air wake that had not been observed before using conventional experimental methods. One of the features was that “bubbles” are periodically shed from the bow. The “bubbles” being shed appear to excite some sort of disturbance in the deck edge vortices that pulse down the side of the ship. This affects all the landing spots along the side. It was known that swirling air existed along the deck edge and pilots are trained to expect it. But the “bubbles” periodically change the nature of the air even as the pilot flies through it. This can increase pilot workload significantly because he must react to an unexpected disturbance. Worse yet, it may only affect one rotor – making the aircraft want to roll when the pilot least expects it.

Another feature the analysis revealed is that the surface flow patterns indicated several separation and reattachment lines. Along a separation line, the air lifts off the deck in an upflow. If you were standing on the deck at a separation line, you would feel air rushing up at you. Conversely, a reattachment line is where already separate flow “reattaches” to the deck. Here you would feel air hitting you on the head. Think of it as driving your car really fast over a speed bump. When your car hits the bump, it will become air born. The spot where it looses touch with the ground is the “separation” line. When the car finally comes down, the place where it hits the ground is the “reattachment” line. The separated flow regions indicate the V-22 will have to pass through a region where the air is swirling in different directions depending on where it is at any given time making for a higher workload for the pilot.

This work could significantly reduce development costs of shipboard flight envelopes by providing realistic, time-varying wake models to manned flight simulators. The simulators could then be used to develop the initial flight envelopes in a virtual environment thus reducing the need for expensive at-sea trials. Also, the analysis techniques developed could be used to aid in the design of the flight decks and islands on new ships and thereby address flight operations at the very initial stages of ship design.

An alternative use could be in the unmanned air vehicle (UAV) area. These vehicles may be launched from and, perhaps more importantly, recovered to ships. These vehicles are quite small so they will be influenced by the ship’s air wake to an even greater extent.

============================================================