基于数据包络分析的民航发动机维修效果评价

对民航发动机维修效果进行评价可以为发动机承修厂的选择提供决策支持,同时有助于发现发动机承修厂的潜在问题。为了实现发动机维修效果的评价,首先提出了评价指标选择的原则,在此基础上给出了发动机维修效果评价指标体系,建立了基于数据包络分析的C2WH模型的发动机维修效果评价模型,并对模型的构建过程进行了描述。针对决策单元输入输出...     

TIG补焊对2219铝合金搅拌摩擦焊接接头微观组织及性能影响

通过对原始搅拌摩擦焊（FSW）接头中间位置进行缺口挖补以模拟焊缝缺陷,采用钨极氩弧焊（TIG）补焊对其进行焊缝缺陷修复试验,研究该补焊工艺对接头微观组织及力学性能的影响规律,为贮箱的FSW焊缝性能评价提供技术支撑和后续工程指导。结果表明,相较于FSW接头,补焊接头TIG焊缝区主要由较大的等轴树枝晶组成,且与FSW焊缝交界处相互掺杂着粗大等轴晶和细小等轴晶;补焊接头固溶区受热循环影响较大,晶粒较为粗大,而过时效区受热循环影响较小,会发生过时效化并形成软化区。补焊接头抗拉强度与延伸率相较于原始FSW接头有所降低,硬度分布大致呈“W”形,WNZ区为硬度值最低处;补焊接头断裂位置始于焊趾处,断口存在大量韧窝,同时韧窝内存在第二相粒子,呈现韧性断裂机制。     

深空环境下卫星编队飞行队形重构实时重规划

针对深空环境中卫星编队自主队形重构的机动问题,设计了队形重构过程中的实时重规划方案。在每次重规划过程中首先用Legendre伪谱法将当前时刻到终端时刻的重构问题离散化为非线性规划问题;其次,根据卫星编队飞行队形重构的特点,使用协同进化粒子群(CPSO)方法对每次重规划进行求解,该方法既避免了传统优化方法对复杂问题梯度的求解,又能在整个优化过程中保证约束条件的满足,并且能够事先为粒子群的进化提供大致方向,极大地提高了每次重规划的计算速度;最后,提出了重构过程中具体的重规划策略,以保证队形重构的顺利进行。仿真结果表明,与传统的控制方法相比,在重构过程中,该方法能够在初始给定轨迹的基础上进一步优化,可以有效降低重构过程的能量消耗,并能够避免重构过程中碰撞的发生。     

Aircraft engine fault detection based on grouped convolutional denoising autoencoders

Many existing aircraft engine fault detection methods are highly dependent on performance deviation data that are provided by the original equipment manufacturer. To improve the independent engine fault detection ability, Aircraft Communications Addressing and Reporting System (ACARS) data can be used. However, owing to the characteristics of high dimension, complex correlations between parameters, and large noise content, it is difficult for existing methods to detect faults effectively by using ACARS data. To solve this problem, a novel engine fault detection method based on original ACARS data is proposed. First, inspired by computer vision methods, all variables were divided into separated groups according to their correlations. Then, an improved convolutional denoising autoencoder was used to extract the features of each group. Finally, all of the extracted features were fused to form feature vectors. Thereby, fault samples could be identified based on these feature vectors. Experiments were conducted to validate the effectiveness and efficiency of our method and other competing methods by considering real ACARS data as the data source. The results reveal the good performance of our method with regard to comprehensive fault detection and robustness. Additionally, the computational and time costs of our method are shown to be relatively low.     

Stacked-GRU网络在动力随动陀螺零部件选配上的应用

动力随动陀螺是红外导引头位标器的核心部件之一，其装配质量对于导弹制导精度而言有着决定性的作用。针对动力随动陀螺零部件一次选配成功率低的问题，提出了一种基于数据驱动的动力随动陀螺零部件选配方法。对动力随动陀螺零部件装配现状进行了分析，考虑到零部件装配参数间复杂的非线性映射关系，建立了堆叠门控循环单元（Stacked-GRU）网络数据驱动模型，对陀螺待装配零部件参数属性值进行了预测。根据预测结果指导了陀螺零部件的选配，并以陀螺典型零部件装配为例进行了验证。实验结果表明，所提方法能够实现动力随动陀螺零部件的选配，且预测精度优于传统神经网络。     

Regression model for civil aero-engine gas path parameter deviation based on deep domain-adaptation with Res-BP neural network

The variations in gas path parameter deviations can fully reflect the healthy state of aero-engine gas path components and units; therefore, airlines usually take them as key parameters for monitoring the aero-engine gas path performance state and conducting fault diagnosis. In the past, the airlines could not obtain deviations autonomously. At present, a data-driven method based on an aero-engine dataset with a large sample size can be utilized to obtain the deviations. However, it is still difficult to utilize aero-engine datasets with small sample sizes to establish regression models for deviations based on deep neural networks. To obtain monitoring autonomy of each aero-engine model, it is crucial to transfer and reuse the relevant knowledge of deviation modelling learned from different aero-engine models. This paper adopts the Residual-Back Propagation Neural Network (Res-BPNN) to deeply extract high-level features and stacks multi-layer Multi-Kernel Maximum Mean Discrepancy (MK-MMD) adaptation layers to map the extracted high-level features to the Reproduce Kernel Hilbert Space (RKHS) for discrepancy measurement. To further reduce the distribution discrepancy of each aero-engine model, the method of maximizing domain-confusion loss based on an adversarial mechanism is introduced to make the features learned from different domains as close as possible, and then the learned features can be confused. Through the above methods, domain-invariant features can be extracted, and the optimal adaptation effect can be achieved. Finally, the effectiveness of the proposed method is verified by using cruise data from different civil aero-engine models and compared with other transfer learning algorithms.     

Wetting behavior of AgCu-4.5Ti filler reinforced by carbon nanotubes on C/C composite

The effect of Carbon Nanotubes (CNTs) content on the wettability of AgCu-4.5Ti + x CNTs (wt%) composite filler alloys on C/C composite was investigated. The results show that the added CNTs reacted with element Ti in the filler and produced the dispersed fine in situ synthesized TiC particles, which increased the consumption of element Ti and provided the nucleus for the growth of Ti-Cu compounds simultaneously. The above effects of introducing CNTs, inhibited the formation of Ti-Cu compounds, also changed the distribution of compounds, which dramatically influenced the interfacial microstructure and characteristics of wetting behavior. The increase of CNTs content refined and dispersed coarse Ti-Cu compounds, decreased the initial spreading temperature, and improved the wettability, but high content of CNTs (more than 0.3wt%) decreased the wettability of the filler alloy. The wetting interfacial microstructure of corresponding composite filler alloys were analyzed by Scanning Electron Microscope (SEM), Energy Dispersive X-ray Spectrometer (EDS) and Transmission Electron Microscope (TEM), which consisted of TiC, TiCu, TiCu₂ and TiCu₄ compound. The typical wetting behavior of AgCu-4.5Ti + 0.3wt% CNTs composite filler on C/C composite was divided into four stages. The effect mechanism of CNTs content on the wetting behavior was proposed.     

Laser joining of CFRTP to titanium alloy via laser surface texturing

Grid pattern was textured on Ti-6Al-4V alloy (TC4) substrate surface by nanosecond laser system. Laser joining of carbon fiber reinforced thermoplastic composite (CFRTP) to TC4 joints were performed, and the effect of texture grid depth was investigated. The contact angle of molten CFRTP on textured TC4 surface was measured and the tensile-shear force was tested. The fracture surface and interface morphology were observed. The results indicated that the wettability of molten CFRTP on TC4 surface improved remarkably after laser textured TC4. Shear force of CFRTP/TC4 joints was increased by 156% after laser textured TC4 surface. When the depth of grid was deeper than 100 μm, contact angle increased and incomplete filling of molten CFRTP in grid occurred, the shear force thus decreased gradually. Resin-carbon fibers mixture was adhered on the fracture surface of TC4, and the variation tendency of adhesion ratio was consistent with that of shear force. TC4 matrix was exfoliated from substrate and adhered at the fracture surface of CFRTP, indicating stronger mechanical interlocking occurred at the joining interface after laser textured TC4 surface. Beside mechanical interlocking, compound layer consisted of CTi_0.42V_1.58 carburization phase was also confirmed at interface, suggesting that chemical bonding also occurred at the joining interface.     

基于微分平滑的欠驱动刚性航天器时间最优轨迹规划研究(英文)

针对非轴对称欠驱动航天器提出了一种新的基于微分平滑特性的时间最优轨迹规划算法。首先,在控制输入受限的约束条件下,对给定的初末姿态,选取航天器机动时间最短为待优化的性能指标。其次,在欠驱动航天器的非受控轴上引入虚拟控制输入,扩展系统的平滑特性,从而使得系统的姿态和控制输入变量均可由平滑输出函数及其各阶导数代数表征;同时为保证优化问题的等价性,引入虚拟控制输入始终为零的附加等式约束。进一步,利用伪谱法离散系统的平滑输出函数,将优化问题最终转化为一个低维的,仅含有代数约束的非线性规划问题,且无需积分求解系统的动态力学方程。仿真结果表明该算法不仅能够有效地减少优化所需的时间,而且运算精度较高。     

Time-optimal trajectory planning for underactuated spacecraft using a hybrid particle swarm optimization algorithm

A hybrid algorithm combining particle swarm optimization (PSO) algorithm with the Legendre pseudospectral method (LPM) is proposed for solving time-optimal trajectory planning problem of underactuated spacecrafts. At the beginning phase of the searching process, an initialization generator is constructed by the PSO algorithm due to its strong global searching ability and robustness to random initial values, however, PSO algorithm has a disadvantage that its convergence rate around the global optimum is slow. Then, when the change in fitness function is smaller than a predefined value, the searching algorithm is switched to the LPM to accelerate the searching process. Thus, with the obtained solutions by the PSO algorithm as a set of proper initial guesses, the hybrid algorithm can find a global optimum more quickly and accurately. 200 Monte Carlo simulations results demonstrate that the proposed hybrid PSO–LPM algorithm has greater advantages in terms of global searching capability and convergence rate than both single PSO algorithm and LPM algorithm. Moreover, the PSO–LPM algorithm is also robust to random initial values.