基于混合神经网络的无线电信号自动调制识别

随着无线电信号数据海量增加，复杂电磁环境下面临着未知威胁和目标侦察识别复杂度高的问题，本文针对未知无线电信号的特征提取任务，设计了一种混合神经网络以提高目标无线电信号的识别能力。先通过胶囊神经网络对未知信号的空间信息进行提取，再进一步运用门控循环单元提取信号在时间上的特征信息。设计混合网络模型将信号的时间和空间特征相结合，提高对目标信号的分类精度。通过RML2016.04C调制信号数据集，验证了混合神经网络的识别性能。结果表明:当信噪比为6 dB时，混合网络模型对多种不同调制信号的分类精度大于95%。因此，本文所设计的混合神经网络能够有效对不同调制信号进行准确分类。     

Ti/CFRP超混杂层板的研究与应用展望

Ti/CFRP超混杂层板具有高比强度、高损伤容限及优异的疲劳性能,是航空航天领域的重要选材对象之一。界面性能决定了层板综合力学性能的好坏,是Ti/CFRP层板目前的研究热点。首先综述了钛和钛合金不同表面处理工艺;然后介绍了Ti/CFRP层板力学性能试验及理论预报的研究现状;最后通过分析Ti/CFRP层板航空航天领域的现有应用情况,展望了Ti/CFRP层板未来潜在的应用方向。     

复合材料双钉胶螺混合连接接头多参数优化设计北大核心CSCD

为提高复合材料双钉胶螺混合连接接头的承载能力,本文结合实验和有限元方法,利用ABAQUS建立了有限元模型并验证了其可行性,基于代理模型和遗传算法,对碳纤维复合材料板与钛合金板双钉混合连接结构的板宽、端距和孔距等参数开展了多参数优化设计。结果表明:二次多项式代理模型的预测结果精度最高;与优化前试验得到的常规混合连接接头的载荷性能相比,优化后的结构承载能力提升了44.35%。     

Dynamic thermal coupling modeling and analysis of wet electro-hydrostatic actuator

The electro-hydrostatic actuator(EHA) used in more electric aircraft(MEA) has been extensively studied due to its advantages of high reliability and high integration. However, this high integration results in a small heat dissipation area, leading to high-temperature problems. Generally,to reduce the temperature, a wet cooling method of using the pump leakage oil to cool the motor is adopted, which can also increase the difficulty of accurately predicting the system temperature in the early desi...     

Source free unsupervised domain adaptation for electro-mechanical actuator fault diagnosis

A common necessity for prior unsupervised domain adaptation methods that can improve the domain adaptation in unlabeled target domain dataset is access to source domain dataset and target domain dataset simultaneously. However, data privacy makes it not always possible to access source domain dataset and target domain dataset in actual industrial equipment simultaneously, especially for aviation component like Electro-Mechanical Actuator (EMA) whose dataset are often not shareable due to the data copyright and confidentiality. To address this problem, this paper proposes a source free unsupervised domain adaptation framework for EMA fault diagnosis. The proposed framework is a combination of feature network and classifier. Firstly, source domain datasets are only applied to train a source model. Secondly, the well-trained source model is transferred to target domain and classifier is frozen based on source domain hypothesis. Thirdly, nearest centroid filtering is introduced to filter the reliable pseudo labels for unlabeled target domain dataset, and finally, supervised learning and pseudo label clustering are applied to fine-tune the transferred model. In comparison with several traditional unsupervised domain adaptation methods, case studies based on low- and high-frequency monitoring signals on EMA indicate the effectiveness of the proposed method.     

Lift enhancement based on virtual aerodynamic shape using a dual synthetic jet actuator

The Dual Synthetic Jet Actuator (DSJA) is used to develop a new type of lift enhancement device based on circulation control, and to control the flow over the two-dimensional (2D) NACA0015 airfoil. The lift enhancement device is composed of a DSJA and a rounded trailing edge (Coanda surface). The two outlets of the DSJA eject two jets (Jet 1 and Jet 2). Jet 1 ejects from the upper trailing edge, which increases the circulation of airfoil with the help of the Coanda surface. Jet 2 ejects from the lower trailing edge, which acts as a virtual flap. The Reynolds number based on the airfoil chord length and free flow velocity is 250000. The results indicate that the circulation control method based on Dual Synthetic Jet (DSJ) has good performance in lift enhancement, whose control effect is closely related to momentum coefficient and reduced frequency. With the increase of the reduced frequency, the control effect of the lift enhancement is slightly reduced. As the momentum coefficient increases, the control effect becomes better. When the angle of attack is greater than 4°, the increments of lift coefficients under the control of DSJ are larger than those under the control of the steady blowing at a same momentum coefficient. The maximum lift augmentation efficiency can reach 47 when the momentum coefficient is 0.02, which is higher than the value in the case with steady blowing jet circulation control.     

具有欺骗攻击的主从神经网络混合 事件触发同步控制

文章提出了一种针对受到欺骗攻击的主从神经网络的混合事件触发同步控制方法。所提出的方法采用以伯努利分布为特征的混合事件触发方案，通过仅在必要时传输数据来减轻通信负担并处理不可预测的网络环境。记忆事件触发机制可以在系统响应状态的峰值和低谷期间传输更多的数据。通过考虑实际情况中的不确定性，将网络诱导时滞和随机发生的欺骗攻击纳入一个统一的框架，并构建了相应的李雅普诺夫泛函。其次，使用李雅普诺夫理论，推导出一个稳定性判据。最后，通过数值算例验证了所提出方法的有效性。     

执行器故障下挠性航天器自适应边界容错控制

针对执行器故障下的挠性航天器的姿态控制及振动抑制问题,提出了基于模型变换的自适应边界容错控制方案,实现了挠性航天器的姿态跟踪及振动抑制。首先,为了准确刻画航天器刚体与挠性振动之间的强耦合特性,采用互联偏微分方程-常微分方程描述挠性航天器的混合动力学;其次,设计辅助信号建立边界条件与内部动力学之间的关系;进而,采用直接自适应控制技术设计边界容错控制律及参数更新律,保证挠性航天器闭环姿态控制系统的渐近跟踪性能;最后,仿真结果验证了所提出的边界容错控制算法的有效性。     

High dynamic output feedback robust control of cascaded extended state observer

High dynamic tracking performance is a key technical index of hydraulic flight motion simulator(HFMS). However, the strong nonlinearities, various model uncertainties and measurement noise in hydraulic actuation systems limit the high dynamic performance improvement. In this paper, the outer axis frame of a HFMS is taken as a case study and its nonlinear dynamic model with consideration of strong nonlinearities, matched and mismatched uncertainties is established.A novel cascaded extended state ...     

零电压型混合式限流装置换流后介质恢复过程研究

通过设计的试验电路，研究了零电压型混合式限流装置换流过程中的燃弧能量及开距长度对于介质恢复时间的影响，对临界击穿电压值函数内的参数进行了拟合，得到了一些规律性结论。