Aero-structure-stealth Coupled Analysis And Quick Optimization Of High Aspect Ratio Wing

Due to the inherent features including high lift-to-drag ratio and large inside volume,high aspect ratio wing has been widely used in various aircrafts, such as civil airplane transport airplanes, long-endurance UAVs and so forth.Considering the high flexibility of the high-aspect ratio wing, an aero-structure-stealth coupled optimization of high-aspect ratio wings based on the high fidelity coupled analysis of aerodynamic, structure and stealth using adaptive metamodel method and decomposition strategy has been performed in this paper. The main contents are summarized as follow:(1) First, the aero-structure-stealth coupled analysis of high-aspect ratio wings based on high fidelity models is performed. The parametric geometry methods for airfoiland high aspect wing areemployed to achieve the automaticupdate of the wing. Computational fluid dynamic(CFD), structure finite element analysis(FEA) and computational electronic method(CEA)are adopted in aerodynamic,structural and stealthy disciplines respectively, as to improve the analysis accuracy. The aero-structure-stealth coupled analysis is also implemented through an interpolation approach.The coordinates of control points located at leading and ending edge of the wing are used to update the geometry model. After the aero-structure coupled analysis, the stealth analysis is executed in terms of the deformed wing to finally complete the aero-structure-stealth coupled analysis. Through an example of a high-aspect ratio wing, the coupled analysis method introduced in this paper is proved to be reasonable, and the neccesacity of developing aero-structure-stealth coupled analysis is illustrated simultaneously.(2) To improve the space-filling and projective properties of sequential sampling problems, a novel deterministic sequentialMaximin Latin Hypercube design methodbased on Maximin criterion(S-SLE) is introduced in this paper. First, a mesh-mapping algorithm is proposed to map the positions of existing points into the new hyper chessboard to ensure the projective property. According to Maximin distance criterion, new sequential samples are generatedthrough successive local enumeration iterations to improve the space-filling uniformity. Through a number of comparative studies with Quasi-LHD, MIPT and MCSR, several appealing merits of S-SLE are demonstrated: i) it is flexibleand robust to produce high quality multiple-stages sequential samples; ii) the proposed method can improve the overall performance of sequential metamodel-based optimization algorithms.(3) To reduce the computational cost of aero-structure-stealth coupled optimization, an optimization framework using adaptive metamodel and decomposition techniquesis researched in this paper. The optimization is performed in two-stage: the airfoil optimization and the wing optimization. The optimized result of airfoil optimization is used in the wing optimization. In the wing optimization, the whole design variables are divided into two levels: the system level and sub-system level. The adaptive metamodel-based optimization method is employed to optimize the system problem and the sequential quadraticprogram(SQP) in Nastran is applied to executing the sub-system optimization. First, the adaptive response surface method using intelligent space exploration strategy(ARSM-ISES) is studied. By testing numerical benchmark and engineering problems with ARSM, EGO, MPS etc., ARSM-ISES overweigh those algorithms in terms of convergence, efficiency and robustness. Then, a study about two-level optimization strategy with coupled analysis is performed, which is proved to have high efficiency through numerical examples and simple aero-structure coupled problem.(4) Finally, the aero-structure-stealth coupled optimization framework is employed to perform the design and optimization of a high aspect ratio wing. After airfoil optimizing, the life-to-drag ratio increases 18.6% and the RCS reduces 14.7% with the constraints satisfied. In the wing optimization, the lift-to-drag ratio reduced 1.5%, while the structural mass and RCS declined 51.8% and 68.6% respectively, so the synthetic performance of the wing is improved. Moreover, the number of function evaluations is only 51.That means that the high fidelity coupling analysis model and efficient optimization framework introduced in this paper can improve the synthetic performance of high aspect ratio wing and reduce the computational cost.