An inverse design method for supercritical airfoil based on conditional generative models

Inverse design has long been an efficient and powerful design tool in the aircraft industry.In this paper, a novel inverse design method for supercritical airfoils is proposed based on generative models in deep learning. A Conditional Variational Auto Encoder(CVAE) and an integrated generative network CVAE-GAN that combines the CVAE with the Wasserstein Generative Adversarial Networks(WGAN), are conducted as generative models. They are used to generate target wall Mach distributions for the inve...     

零角动量欠驱动航天器逆最优稳定控制律设计

研究轮控式零角动量欠驱动航天器姿态最优稳定控制问题。考虑到该类型航天器不存在定常光滑稳定控制律的特点,通过Lyapunov直接法和Backstepping方法设计了一种非线性不连续反馈控制律,同时得到控制Lyapunov函数（CLF）,并由此得到逆最优稳定控制律。该控制律可以避开求解Hamilton-Jacobi方程,最小化某一代价函数,同时具有扇形稳定裕度,对输入不确定性具有一定的鲁棒性。数学仿真结果表明,所设计的非线性不连续反馈控制律能够使姿态渐近稳定至平衡点,并具有最优性,以及在转动惯量存在不确定性时,扇形稳定裕度使系统具有一定的鲁棒性。     

Simultaneous ISAR imaging of group targets flying in formation

This paper proposes a novel inverse synthetic aperture radar(ISAR) imaging method based on second-order keystone transform(KT) and Sandglass transform for group targets flying in a formation with constant accelerated rectilinear motion in the same radar beam. First, range curvature and range walk of each sub-target among group targets are corrected by the second-order KT combined with the quadratic phase term compensation. After range alignment, the signals in each range frequency cell can be modelled as multiple chirp signals and then the Sandglass transform is utilized to cross-range imaging, which transforms the time–frequency distribution of the signals in each range frequency cell into beelines parallel to the slow time axis simultaneously. Finally, cross-range profiles of group targets in each range frequency cell are obtained via a projection of the perk of every scatterer in the two-dimensional accumulation plane onto the frequency axis. The advantage of the proposed method is that it can align range profiles of each sub-target simultaneously and image cross-range profiles directly without separating the returned signals, which simplifies the operation procedure. Simulation results are used to demonstrate the effectiveness of the proposed method.     

基于逆动态和变结构理论的空间飞行器大角度机动控制

针对飞行器大角度机动的控制问题具有强耦合、高度非线性、参数不确定性等特点,本文首先利用逆动态控制的非线性解耦能力对系统进行非线性解耦,然后基于变结构控制的鲁棒性进行控制器的设计,并在MATLAB环境下进行了仿真研究。仿真结果表明,该控制方法在有外界干扰和系统参数不确定的条件下具有较快的收敛性和一定的鲁棒性。     

考虑移动目标不确定行为方式的轨迹预测方法

针对现有方法难以预测出符合飞行移动目标不确定行为方式轨迹的问题,提出基于逆强化学习的飞行移动目标轨迹预测方法,通过学习目标行为偏好以及模拟目标行为决策过程的方式预测目标的移动轨迹。首先基于深度神经网络建立目标的行为决策模型与行为偏好模型,然后通过最大熵逆强化学习方法交替地学习模型参数。为了有效地学习目标的不确定行为特征,采用监督学习的方法学习出目标示例轨迹概率分布模型,用于指导目标行为偏好模型的训练以及初始化目标行为决策模型,同时通过对目标行为偏好模型进行预训练的方式提高其训练质量。仿真结果表明,提出的飞行移动目标轨迹预测方法可通过学习到的目标行为决策模型较为准确地模拟目标的行为方式,预测的目标轨迹分布与真实的目标轨迹分布在Kullback Leibler(KL)散度下的相似度可达0.24。     

Biaxial tensile behavior of Ti-6Al-4V under proportional loading

The biaxial tensile behavior of isotropic Ti-6Al-4 V is characterized in this paper. A novel cruciform specimen was designed and optimized to achieve uniform stress and strain distribution within the gauge area. Biaxial tensile tests were conducted at three different loading ratios by the biaxial testing machine. The Digital Image Correlation (DIC) technique was applied to determine strain distribution, and a high-speed camera was employed to record the fracture process. An Inverse Analysis (IA) approach with a combined experimental and numerical method was proposed to determine the true stresses at the gauge section of the specimen during biaxial tensile tests. The results indicate that the initial yield locus can be described by the Cazacu criterion accurately, whereas the Mises criterion can predict better the strengthening behavior of Ti-6Al-4 V in the first quadrant in the principal stress space.