Variable-fidelity multidisciplinary design optimization with uncertain design variables for control surface of tailless aircraft

Abstract

In aircraft design, tailless aircraft is an attractive concept for their aerodynamic efficiency. However, a tailless configuration has disadvantages in stability and control (S&C) compared to conventional aircraft. To overcome the weakness, many types of control surface have been researched with the Innovative Control Effector(ICE) configuration. In this paper, for the optimization of control surface with reduced cost of optimization, efficient global optimization (EGO) method is applied. For the EGO process, variable-fidelity design optimization framework is constructed with the concept of variable-fidelity(VF) Kriging analysis with sample point update, multi-point multi-objective infill sampling criteria(MPMO-ISC) and overlapping coefficient. The framework is validated with numerical test function, and all moving wingtip (AMT), the most effective control effector for ICE 101 configuration, is optimized. For variable-fidelity design optimization, three aerodynamic analysis methods with different fidelity, panel method, Euler method, and Navier-Stokes method, are used for the sampling. The optimization problem is formulated with standard requirements by Department of Defense.