利用时延神经网络的动载荷倒序识别

将时延神经网络引入动载荷识别研究中,结合时延神经网络的"记忆"特性、因果有限长冲激响应（FIR）系统理论与振动响应的求解原理,提出一种利用时延神经网络的时域动载荷倒序识别方法。对一个受两点随机动载荷作用的舵面模型结构进行载荷识别验证实验,结果表明,用本文方法识别的两个激励点上识别载荷样本的时间序列与真实载荷样本的时间序列之间的均方根误差分别为0.635 4和2.543 7,识别载荷样本时间序列与真实载荷样本时间序列的相关系数分别为0.965 7和0.826 2,功率谱密度曲线也能够较好吻合。本文提出的方法具有不需要结构动力学模型、识别精度高的优点。     

变截面悬臂梁结构动载荷辨识方法

在航空航天领域，作用在结构上动载荷的确定对结构健康监测是非常必要和重要的。为此，本文以类似机翼结构的变截面悬臂梁结构为研究对象，提出了一种基于光纤光栅传感器与卡尔曼滤波器的动载荷识别方法。首先，根据变截面梁单元形式，推导出变截面梁的质量矩阵与刚度矩阵，建立动力学运动方程。然后，以光纤光栅传感器测得的应变信息作为观测信号，通过卡尔曼滤波器生成的增益矩阵、新息序列矩阵以及协方差矩阵，得到灵敏度矩阵和估计力的增益矩阵。在此基础上，利用广义回归模型及其最小二乘算法，估算出动载荷大小、判断出动载荷激励位置。借助数值仿真与实验手段，分别验证了该方法对于单点正弦激励、方波激励、锯齿波激励以及多点同时激励等工况下的动载荷识别效果。结果表明，本文所提算法具有较好的动载荷识别效果和噪声抑制能力，能够为未来风洞试验和真实飞行试验环境中诸如大展弦比机翼表面气动压力等载荷实时辨识、气动外形自适应控制以及结构健康监测提供技术支撑。     

基于模态滤波器改进的动载荷识别模型

 基于模态滤波器概念提出了对载荷识别模态模型法的原理性校正方案,构造了动载荷识别的离散模态滤波器模型。以其他点激励代替常常不可达的实际工作输入点,通过建立系统的模态参数并相继在模态和物理坐标下的求解对动态载荷作出估算,可以避免经典模态模型法通常对结构或边界条件所作的修改和由于模态参数自身误差而导致的较大识别误差,实例分析表明该模型识别精度较高并具有普遍的工程适用性。     

平稳随机载荷的信号特征提取与深度神经网络识别

针对线性时不变结构的平稳随机载荷识别问题,从结构的动力学响应求解原理出发,利用小波变换对于信号特征的提取能力与长短期记忆神经网络（LSTM）对于序列问题的强大建模与映射能力,提出了一种针对平稳随机载荷的特征信号识别方法,通过对作用于三自由度振动系统数值模型上的平稳随机动载荷识别,证明了方法的可行性。对一个受2点平稳随机载荷作用的加筋壁板结构模型进行动载荷识别实验,结果表明,用提出的方法识别的动载荷均方根相对误差均小于5%,该动载荷识别方法具有良好的识别能力。     

复合材料板壳任意位置处载荷识别

提出了一种受载单元判别法,根据遗传算法生成的染色体所对应的载荷作用点的横、纵坐标和每个单元四条边间的几何关系,判断出集中载荷作用的单元,再根据力的等效分配原则,把载荷等效分配到所在单元的节点上,进而基于有限元分析和遗传算法自身优越的全局搜索能力实现作用于复合材料结构网格节点或单元内部任意位置处的载荷识别.与现有的神经网络识别法和有限元反分析法相比,该方法具有能够识别非网格节点处的载荷的优点,且仿真算例表明其识别精度更高.      

基于数值模拟的气动载荷计算方法研究

飞机气动载荷的获取一般要根据多项实验结果,经过飞行参数的仿真模拟,实验结果的处理等过程才能获得,而在飞机的初步设计阶段不可能进行大量的实验,但需要计算飞机的载荷和寿命.本文提出了一种利用数值模拟来确定基本气动载荷的方法,对于偏副翼的情况,用工程计算方法获得偏转引起的附加量,加到基本气动载荷上得到偏副翼时的气动载荷,该方法已用于某飞机型号设计.     

随机噪声载荷作用下薄壳结构的随机振动实验研究

为研究薄壁壳结构在随机噪声激励下的随机振动响应特征,本文设计了一套针对薄壁壳在内部随机噪声作用下的振动问题进行研究的实验方案,同时对噪声载荷和薄壳振动加速度响应进行测量,并对实验结果进行初步的定性分析,了解噪声载荷在频率域上的分布特征和薄壁壳在内部随机噪声作用下的振动响应特征,获得在噪声激励的频率范围内能激起薄壳该频率范围内所有模态共振等结论,为薄壳结构在内部噪声作用下随机振动响应的计算方法的研究提供依据。      

加筋复合材料结构的冲击载荷识别

 提出了一种冲击载荷识别方法,同时识别冲击位置并重建冲击载荷时间历程。该方法采用一组适当的参数来表示冲击载荷,将时域内的载荷识别问题转换为参数识别问题;通过最小化冲击响应模型计算结果与实际量测信息之间的差别,智能优化方法自适应地识别出描述冲击位置和载荷时间历程的参数。该方法算法明确、过程简单、通用性强。将此方法应用于复合材料加筋结构的冲击载荷识别,加筋结构等效为材料性质不均匀分布的层板结构,采用假设模态法建立正向冲击响应模型。数值仿真结果表明了本方法的有效性和可应用性。     

飞机飞行载荷的简化计算方法研究

飞行载荷分析是一项复杂而繁重的工作,研究一种适用于方案设计阶段飞行载荷的快速分析方法,对于提高飞行载荷的计算效率具有重要意义。基于小扰动线性分析理论,归纳翼身气动载荷、平尾气动载荷、垂尾气动载荷、舵面铰链力矩的理论计算方法,以及升力面的气动载荷分布、惯性载荷分布、剪力和弯矩的工程计算方法。针对某型单座竞技飞机的飞行载荷,以外形尺寸、质量特性和气动导数作为输入,通过Matlab 仿真分析,得到各个部件的气动载荷、惯性载荷、舵面铰链力矩、剪力和弯矩等参数响应。结果表明：该简化方法能够根据较少的输入数据快速求解出各个动力学参数,计算结果可以作为方案设计阶段结构设计的载荷输入。     

飞行试验抖振载荷统计技术

随着现代飞机使用要求的不断提高,飞机抖振问题越发突出.在借鉴F/A-18和F/A-22等飞机抖振载荷试飞的基础上,分析、归纳了抖振载荷试飞方法和测量方法,并利用功率谱密度法、统计学方法对某型运输机抖振载荷数据进行了分析研究.结果表明,这些方法在飞机抖振载荷研究方面具有较好的应用价值.     

基于旁路溢流原理的流体脉动主动控制

 随着现代飞机液压系统向高压、大功率方向发展,周期性流体脉动更容易在液压能源系统中产生流固耦合振动,导致液压管路疲劳破坏。针对被动式流体脉动抑制方法不具备自适应性、对低频脉动抑制效果差等缺点,提出了基于旁路溢流原理的流体脉动主动控制方法,即在与能源主管路连通的分叉管路上安装压电陶瓷驱动伺服阀主动消振器,通过主动消振器的溢流产生次级脉动波与主管路中原有压力波相互抵消,使管路中流体脉动减小。并针对流体脉动周期性特点,提出一种在线频率估计的自适应前馈控制方法,利用参考传感器测量泵源脉动信号,在线获取脉动频率,并以此频率构造控制器的参考输入信号,实现对流体脉动的主动控制。为验证所提方法的有效性,设计了主动消振实验平台。实验结果表明,所提出的脉动主动控制方法具备很强的自适应性,能对流体脉动进行很好的抑制。     

Efficient power supply for the microwave electrothermal thruster(MET)

A resonant switch-mode power supply for the microwave electrothermal thruster (MET) is presented in this paper. The converter is operated with soft-switching at high frequency and exhibits a high efficiency. The soft switching technique used in this converter and the current-source inductor at the input minimize the EMI noise. Electric isolation between input and output is achieved with a center-tap transformer, whose magnetizing inductance is used as the resonant inductance of its resonant tank. The resulting high power density and increased reliability of the converter make it very suitable for aerospace applications. Simulation and experimental results of a 28 V/4.5 kV example are also presented     

Computer-Aided Design and Weight Estimation of High-Power High-Voltage Power Conditioners

Aspects of high-power high-voltage power conditioner design and weight estimation relevant to space subsystems are discussed. Weight has become an increasingly important parameter with the advent of larger and more sophisticated spacecraft, especially those for high-power communication. A computer program for estimating the weight of a high-power dc-to-dc power conditioner as functions of output power, operating frequency, input voltage range, maximum input voltage, and efficiency, respectively, is described, including computer-aided design of inductors and transformers. Curves of typical power conditioner weight as functions of the preceding parameters, derived from the power conditioner weight program, are presented.     

Analysis and design of (LC)(LC)-type series-parallel resonantconverter

A series-parallel resonant converter employing (LC)(LC)-type tank circuit operating in lagging power factor (PF) mode is presented and analyzed using complex ac circuit analysis. Design curves are obtained and the converter is optimized under certain constraints. Detailed Space Integrated Control Experiment (SPICE) simulation results are presented to evaluate the performance of the designed converter under varying load conditions. Results obtained from an experimental converter are also presented. The results obtained from the theory, SPICE simulation, and the experimental converter are compared. The proposed converter has high efficiency from full load to very light load (<10%). Switching frequency variation required for a wide change in the load (near load open circuit to full load) is narrow compared with the series resonant converter (SRC)     

Frequency-controlled series-resonant converter with synchronous rectifier

This paper presents an analysis and experimental results for a frequency-controlled series-resonant dc-dc converter that consists of a Class-D zero-voltage-switching (ZVS) series-resonant inverter and a center-tapped synchronous rectifier. If the dc output voltage is low, the efficiency of the converter is dominated by the efficiency of the rectifier. Low on-resistance metal-oxide-semiconductor field-effect transistors (MOSFETs) are used in the rectifier instead of diodes because the forward voltage drop across the rectifying device is low, resulting in a high efficiency. The dc output voltage is regulated against variations in the load resistance and the dc input voltage by varying the operating frequency. Experimental results are presented for a converter with a dc input voltage of 150 V, an output voltage of 5 V, and a dc load resistance ranging from 0.5 to 5.5 R. The measured efficiency was 86% for a 50 W output and 89% for a 25 W output. The theoretical results were in good agreement with the measured results.     

基于频率雨流计数法的发动机振动疲劳载荷谱编制

传统雨流计数法适用于常规疲劳载荷计算,不能保留振动疲劳的频率信息,但是频率对于振动疲劳十分重要,因此在传统雨流计数法的基础上提出一种频率雨流计数法。以某型号发动机试车振动加速信号为例,采用振动疲劳载荷谱的编制方法,编制振动信号功率谱、载荷 - 频次载荷谱、振动疲劳载荷谱,对频率雨流计数法进行验证。结果表明:采用本文方法编制的载荷谱能够反映发动机实际工作载荷,满足实验室疲劳试验载荷谱要求。     

采用频域多参考点模态识别法进行工作状态下的模态识别

 针对工程实际情况,从实测响应间的互相关函数入手,推导出适用于工作状态下的频域多参考点模态识别法,然后采用一飞机模型对该方法进行了试验验证。     

一种基于短时机动试飞数据的动力学辨识优化方法

飞行动力学辨识算法的一个关键问题是,如何通过简单的机动获取所关心频率范围的响应特性。短时倍脉冲是一种易于实施的激励信号,兼顾试飞安全性与经济性,但与频域辨识法通常使用的扫频输入激励相比,短时机动频谱范围窄、信噪比低,一般难以得到准确的辨识结果。对如何基于短时机动飞行试验数据,提高辨识结果准确性的问题进行了研究。首先分析了经典Welch谱估计进行时域-频域转换过程中,影响非参数模型辨识精度的主要因素,提出了削减窗函数边缘缩减效应的数据预处理方法,并结合多窗口综合技术,提高频域特性辨识结果的精度。在参数化模型辨识过程中,针对有限频谱范围,提出了利用相干函数和功率谱密度加权综合,确定等效拟配的频率范围和频率节点的自适应方法,使得低阶等效拟配与输入激励信号高度相关,提高参数化模型辨识的精度、一致性和适应性。通过不同类型飞机的大量短时机动和少量扫频飞行试验数据模型辨识的工程应用示例,验证了动力学辨识优化方法算法稳定、结果准确,可满足飞行品质模态特性评价等应用需求。     

Single-Stage Single-Switch Switched-Capacitor Buck/Buck-Boost-Type Converter

A new dc-dc converter featuring a steep step-down of the input voltage is presented. It answers a typical need for on-board aeronautics modern power architectures: power supplies with a large conversion ratio able to deliver an output voltage of 1–1.2 V. The proposed structure is derived from a switched-capacitor circuit integrated with a buck converter; they share the same active switch. The proposed solution removes the electromagnetic interference (EMI) emission due to the large di/dt in the input current of the switched-capacitor power supplies. Compared with a quadratic buck converter, it presents a similar complexity, a smaller reduction in the line voltage at full load (but less conduction losses due to smaller input inductor current and capacitor voltage), lower voltage stresses on the transistor and diodes, lower current stresses in the diodes, and smaller size inductors. A similar structure using a buck-boost converter as the second stage is also presented. The experimental results confirm the theoretical developments.     

Distributed event-triggered adaptive control for second-order nonlinear uncertain multi-agent systems

In this paper, the event-triggered consensus control problem for nonlinear uncertain multi-agent systems subject to unknown parameters and external disturbances is considered. The dynamics of subsystems are second-order with similar structures, and the nodes are connected by undirected graphs. The event-triggered mechanisms are not only utilized in the transmission of information from the controllers to the actuators, and from the sensors to the controllers within each agent, but also in the communication between agents. Based on the adaptive backstepping method, extra estimators are introduced to handle the unknown parameters, and the measurement errors that occur during the event-triggered communication are well handled by designing compensating terms for the control signals. The presented distributed event-triggered adaptive control laws can guarantee the boundness of the consensus tracking errors and the Zeno behavior is avoided. Meanwhile, the update frequency of the controllers and the load of communication burden are vastly reduced. The obtained control protocol is further applied to a multi-input multi-output second-order nonlinear multi-agent system, and the simulation results show the effectiveness and advantages of our proposed method.