Search

Conventional approaches of design variable generation in nacelle design cases are usually proposed from either of two perspectives: (1) utilizing geometric resemblance to airfoil shape; or (2) selecting flexible geometric parametrization so that any potential geometry can be produced. However, the relationship between geometry and aerodynamic performance objectives is not emphasized in these approaches: design variables are likely not carrying the information that is most needed for aerodynamic optimization, as is the case in nacelle cowl design of drag reduction. In fact, an ideal aerodynamic information should be related with the move of samples toward better aerodynamic performance in design space. As an attempt to fix this issue, this paper proposes a new design variable generation method by extracting objective space (i.e., the aerodynamic performance of samples mapped from nacelle design space) information via linear discriminant analysis (LDA). By accomplishing this "objective space filtration," the aerodynamic performance improvement potential is injected into design variable generation. Proof-of-concept studies of nacelle cowl design are provided in the paper demonstrating a more direct relationship that links: (1) LDA-aided design variables versus cowl drag, and (2) LDA-aided design variables versus shock strength and laminar length, respectively. LDA is used in the studies as a classic technique of classification, so that the analysis of objective space information finds a way to have impact on design space. The new approach outperforms a controlled test realized by Proper Orthogonal Decomposition in terms of reflecting aerodynamic information in design space, in that a localized improvement of aerodynamic objectives can be more clearly seen in the newly generated design variables. The result shows that the new design variable generation helps improve aerodynamic design and that it can be further extended to other cases beyond the scope of nacelle cowl design. [ABSTRACT FROM AUTHOR]

In this paper, a new aerodynamic solution strategy for non-smooth configurations is proposed based on the wall modification model by machine learning to perform numerical simulations, rather than directly describing the global flow field with massive grids. The aerodynamic effect of non-smooth configurations in the presence of pressure gradients is investigated utilizing the proposed method. Flow features of non-smooth surface are provided by high-fidelity surface flow data acquired through lattice Boltzmann method simulation. The wall modification model is constructed by Fruit fly Optimization Algorithm-Generalized Regression Neural Network (FOA-GRNN) to reproduce the behavior of microflow near the non-smooth surface. Typical flow features, e.g., velocity corrections induced by surface texture as the output of the FOA-GRNN model, are imposed on configuration boundaries, improving computational efficiency and wall resolution. The novel aerodynamic solution strategy is validated by comparing the results of the experiment. In addition, the performance analysis of compressor cascade with micro riblet surface utilizing the above method is conducted. The results indicate that the non-smooth surface structure decreases skin friction and turbulent intensity in the flow channel compared with smooth cascade, thus significantly reducing the total pressure loss. The paper shows a positive prospect of the data-driven strategy in evaluating the aerodynamic performances of non-smooth configurations and provides a reliable solution method for the subsequent design of micro-nano surfaces. [ABSTRACT FROM AUTHOR]

In this paper, in order to overcome computational challenge in numerical simulations of airfoil covered with micro/nanopatterns for aerodynamic drag-reduction design, an ANN-Based boundary strategy is proposed to balance the accuracy and efficiency. Lattice Boltzmann Method (LBM) is utilized to extract the micro/nanoflow characteristics in the near-wall region considering the rarefied effect and the available microscopic data is used to train the surrogate boundary model by Generalized Regression Neural Network (GRNN). The modified boundary conditions obtained by ANN-Based model replacing the real complex and fine micro/nano structure are applied on the smooth configuration to perform macroscopic simulations. Rectangular riblets as micro/nano pattern are discussed for our study. Various arrangement strategies of rectangular riblets over airfoil are adopted by adjusting the width, height, and coverage area to investigate their effects on aerodynamic performances. The results indicate that for the riblet airfoil, the skin friction reduces and the transition position moves backward compared with those of the smooth airfoil. Furthermore, the lift-to-drag ratio also significantly increases and the rate of improvement is up to 13.8% at the Angle of Attack (AOA) of 1°. This paper shows a perspective in aerodynamic design with micro/nano pattern for drag reduction by providing an innovative multi-scale simplified simulation strategy. [ABSTRACT FROM AUTHOR]