Hypersonic starting flow at high angle of attack

Compressible starting flow at small angle of attack(Ao A) involves small amplitude waves and time-dependent lift coefficient and has been extensively studied before. In this paper we consider hypersonic starting flow of a two-dimensional flat wing or airfoil at large angle of attack involving strong shock waves. The flow field in some typical regions near the wing is solved analytically. Simple expressions of time-dependent lift evolutions at the initial and final stages are given. Numerical simulations by compuational fluid dynamics are used to verify and complement the theoretical results. It is shown that below the wing there is a straight oblique shock(OSW) wave,a curved shock wave(CSW) and an unsteady horizontal shock wave(USW), and the latter moves perpendicularlly to the wing. The length of these three parts of waves changes with time. The pressure above OSW is larger than that above USW, while across CSW there is a significant drop of the pressure, making the force nearly constant during the initial period of time. When, however, the Mach number is very large, the force coefficient tends to a time-independent constant, proportional to the square of the sine of the angle of attack.     

Subsonic impulsively starting flow at a high angle of attack with shock wave and vortex interaction

Impulsively starting flow, by a sudden attainment of a large angle of attack, has been well studied for incompressible and supersonic flows, but less studied for subsonic flow. Recently, a preliminary numerical study for subsonic starting flow at a high angle of attack displays an advance of stall around a Mach number of 0.5, when compared to other Mach numbers. To see what happens in this special case, we conduct here in this paper a further study for this case, to display and analyze the full flow structures. We find that for a Mach number around 0.5, a local supersonic flow region repeatedly splits and merges, and a pair of left-going and right-going unsteady shock waves are embedded inside the leading edge vortex once it is sufficiently grown up and detached from the leading edge. The flow evolution during the formation of shock waves is displayed in detail. The reason for the formation of these shock waves is explained here using the Laval nozzle flow theory. The existence of this shock pair inside the vortex, for a Mach number only close to 0.5, may help the growing of the trailing edge vortex responsible for the advance of stall observed previously.     

Low-frequency unsteadiness of vortex wakes over slender bodies at high angle of attack

A type of flow unsteadiness with low frequencies and large amplitude was investigated experimentally for vortex wakes around an ogive-tangent cylinder. The experiments were carried out at angles of attack of 60–80 and subcritical Reynolds numbers of 0.6–1.8×105. The reduced frequencies of the unsteadiness are between 0.038 and 0.072, much less than the frequency of Karman vortex shedding. The unsteady flow induces large fluctuations of sectional side forces. The results of pressure measurements and particle image velocimetry indicate that the flow unsteadiness comes from periodic oscillation of the vortex wakes over the slender body. The time-averaged vortex patterns over the slender body are asymmetric, whose orientation is dependent on azimuthal locations of tip perturbations. Therefore, the vortex oscillation is a type of unsteady oscillation around a time-averaged asymmetric vortex structure.     

三种用于加工特征识别的神经网络方法综述

加工特征自动识别技术是智能化设计与制造的关键支撑,已有的实用性算法普遍存在学习能力差、识别范围有限和识别速度慢等共性问题。神经网络方法在计算机视觉和模式识别领域获得了巨大成功,其自学习与自适应能力和高速计算等优势也已在加工特征识别中得到初步的展现。对加工特征识别中具有应用潜力的三种不同的神经网络方法进行了研究,剖析了神经网络识别加工特征中的预处理与编码和神经网络结构设计等关键性问题,分析了不同神经网络方法的异同点,总结了当前神经网络识别加工特征的发展方向,为相关领域的研究提供一定的理论指导与技术支持。     

模糊神经网络的侧滑角校正

叙述了一种利用模糊神经网络模型来校正侧滑角的方法。侧滑角的校正基于以下七个输入变量：动静压之比、动压、迎角、滚转速率、偏航速率、飞机重心处侧向过载、基准侧滑角。在模糊神经网络模型的结构辨识中，采用了前向搜寻算法。样本数据的一部分用于训练模糊神经网络，另一部分用于测试和评价所得模型的性能；参数辨识中，采用了反向传播学习算法（即BP算法）。某型飞机的六自由度仿真软件仿真验证表明，用这种方法校正后的侧滑角具有较高的精确度。     

Unsteady aerodynamic modeling at high angles of attack using support vector machines

Accurate aerodynamic models are the basis of flight simulation and control law design.Mathematically modeling unsteady aerodynamics at high angles of attack bears great difficulties in model structure determination and parameter estimation due to little understanding of the flow mechanism.Support vector machines(SVMs)based on statistical learning theory provide a novel tool for nonlinear system modeling.The work presented here examines the feasibility of applying SVMs to high angle-of-attack unsteady aerodynamic modeling field.Mainly,after a review of SVMs,several issues associated with unsteady aerodynamic modeling by use of SVMs are discussed in detail,such as selection of input variables,selection of output variables and determination of SVM parameters.The least squares SVM(LS-SVM)models are set up from certain dynamic wind tunnel test data of a delta wing and an aircraft configuration,and then used to predict the aerodynamic responses in other tests.The predictions are in good agreement with the test data,which indicates the satisfying learning and generalization performance of LS-SVMs.     

弹射救生系统大迎角大侧滑角气动特性工程计算方法研究

对弹射救生系统的气动特性进行了分析,发展了一套快速获取其气动特性的工程计算方法,并对某战斗机弹射救生系统的大迎角大侧滑角气动力特性进行了计算.结果表明,工程计算结果与风洞试验数据一致性良好.     

基于深度神经网络的客机总体设计参数敏感性分析

飞机总体主要设计参数敏感性分析揭示了总体主要设计参数对飞机特性指标的影响,有助于总体设计方案的决策。针对宽体客机总体主要设计参数敏感性问题,根据其总体主要设计参数和特性指标的特点,以及多学科间的耦合关系,建立了深度神经网络模型。该深度神经网络模型以客机总体主要设计参数为输入,对特性指标进行预测。在深度神经网络模型中,设置了多个输入层、多个输出层以及多个分块的隐藏层,从而模拟客机总体主要设计参数对特性指标的影响以及不同特性指标之间的相互作用。测试结果表明,与传统代理模型相比,深度神经网络模型对客机特性指标的预测精度更高,多参数适应性更好。利用该深度神经网络模型对客机总体主要设计参数进行敏感性分析。分析结果表明,机翼1/4弦线后掠角在30°~31.5°时,有利于减少最大起飞重量和起飞平衡场长;发动机海平面最大静推力和机翼面积对客机直接使用成本、最大起飞重量等特性指标的影响最为显著。