This paper considers the application of conventional energy based topology optimization methods for design of aircraft wing box ribs. Compared to standard topology optimization work undertaken at Airbus, the topology optimization of wing box ribs posed several additional challenges, mainly due to the wing box ribs being embedded in a redundant wing box structure. Several approaches to solving this problem have been investigated and are being reported as part of this paper, including a global analysis/optimization approach and two local analysis/optimization approaches. The paper also deals with both the selection of a suitable objective/constraint function formulation for topology optimization and selection of a suitable formulation for handling multiple load cases in topology optimization, but does not deal with any detailed sizing optimization of topology optimized designs.

Institute of Aeronautics and Astronautics 2 λ = Crossflow wavelength parallel to leading edge (dimensional) λ 1 = Target crossflow wavelength parallel to leading edge (dimensional) λ 2 = Control crossflow wavelength parallel to leading edge (dimensional) ν = Freestream viscosity I. Introduction wept-wing laminar flow control (SWLFC) on transport aircraft has long been an area of active academic and commercial research interest. Mature SWLFC could result in drag and fuel savings of as much as 10%. In comparison with other technologies, SWLFC would be the single largest contributor to near-term efficiencies on transport aircraft (Collier, 2010). In order to realize SWLFC on production aircraft, additional technology maturation is needed. The NASA Environmentally Responsible Aviation (ERA) initiative is a national plan for maturation of near and medium term improvements in transport aircraft fleet fuel efficiency, emissions, and noise. The Subsonic Aircraft Roughness Glove Experiment (SARGE) is one such initiative intended to raise the Technical Readiness Level (TRL) of passive SWLFC using a spanwise-period array of micron-sized Discrete Roughness Elements (DREs). Saric, Carrillo, & Reibert (1998) demonstrated that passive laminar-turbulent transition delay is possible using a judiciously designed pressure distribution in conjunction with DREs placed near the attachment line on a swept wing. At chord Reynolds number Re c = 2.4 million (unit Reynolds number Re' =0.4 million/ft) on a swept wing with leading edge sweep Λ LE = 45°, results in the Klebanoff-Saric Wind Tunnel showed that, as long as transition is due to stationary crossflow, DREs have the capability to delay transition past the pressure minimum. The Flight Research Laboratory at Texas A&M University has completed a successful in-flight demonstration of DREs on a Λ LE = 30° swept wing at Re c = 8 million and Re' = 1.7 million/ft (Carpenter, Saric, & Reed 2010; Saric, Carpenter, & Reed 2011; Rhodes, Reed, Saric, & Carpenter 2010). Roughness receptivity studies are presently in progress under these conditions in order to quantify the role of roughness amplitude in generating crossflow waves. In addition, this technique has been demonstrated for supersonic flight (Saric, Reed, & Banks 2004). The overall goal of SARGE is execute a successful flight test that extends the demonstrated effectiveness of DREs to a transportrelevant Mach number, Re c , and C l in addition to Reʹ as demonstrated previously. Belisle, Neale, Reed, & Saric (2010) demonstrated the feasibility of a laminar-flow glove experiment at Re c = 15-20 million using a Gulfstream II as host aircraft. The logical extension of these efforts is to design a single experiment capable of covering this range as well as extending the test to higher Re c in the 22-30 million range. Demonstrating section lift coefficient (C l) and Re' representative of transport aircraft is an additional requirement. The preliminary design of an experiment under these conditions was presented in Belisle, Roberts, Tufts, Tuck...

EXECUTIVE SUMMARYThe present study aims to document the drag reduction for a two-vehicle platoon by operating two full-scale Ford Windstar vans in tandem on a desert lakebed. Drag forces are measured with the aid of a special tow bar force measuring system designed and manufactured at USC. The testing procedure consists of a smooth acceleration, followed by a smooth deceleration of the platoon. Data collected during acceleration allows the calculation of the drag force on the trailvehicle, while data collected during deceleration is used to calculate the drag on the lead vehicle.Results from the full-scale tests show that the drag behaviors for the two vans are in general agreement with the earlier conclusions drawn from the wind tunnel testsnamely, both vans experience substantial drag savings at spacings of a fraction of a car length.3

No abstract