涡扇发动机高压涡轮前温度组合估计方法

考虑到目前暂无法实现机载条件下高压涡轮前温度直接、可靠的测量,提出一种用于涡扇发动机高压涡轮前温度估计的方法.基于涡扇发动机的能量守恒原理,建立高压涡轮前温度与气路参数的热力学关系,进而推导出高压涡轮前温度的6个估计模型.将各温度模型中不易测量的参数以整体的形式作为温度模型系数,并利用某涡扇发动机性能仿真模型建立温度模型系数与可测状态参数的多项式关系,最终确立高压涡轮前温度的组合估计模型.验证结果表明:组合估计方法在发动机健康及性能衰退状态下都具有较高的精度,其性能最好模型的方均根误差不超过1%.与已有线性拟合、神经网络等方法的对比也表明组合估计方法不论在精度还是性能稳定性方面都具有明显优势.      

基于互联网看国内无人直升机发展

介绍近年来国内无人直升机在航空监测、电力放线施工、农田飞行施药、电力巡线和科学考察等民用领域的应用实例,分析国内无人直升机的基本机型及其优缺点,指出国内无人直升机起飞重量低,续航时间短,飞行速度慢的应用现状,在行业应用方面,无人直升机在电力行业有广阔的发展前景。     

Beacon Monitoring: Reducing the Cost of Nominal Spacecraft Operations

Beacon monitoring is an architecture for augmenting on-board health assessment software with the following elements: a transmitter that periodically broadcasts this health assessment to the ground, a network of globally distributed low-cost monitoring stations that relays the health assessment to mission control, and an automated mission control system for notifying on-call operators and initiating appropriate response actions. While beacon monitoring is often cited as a means of lowering nominal monitoring costs for particular missions, these claims have typically been qualitative and undiscriminating in nature. This study introduces modeling and experimentation as means of providing a more fundamental validation of the cost-effectiveness of beacon monitoring. Results include simple quantitative estimates of first-order performance metrics, experimental data that verifies predicted performance and validates the use of beacon monitoring for a test mission, and a method for assessing the value of beacon monitoring for general satellites missions.     

滑油系统全流量在线磨粒静电监测技术研究

 针对滑油系统非金属材料难以实现在线监测的问题,基于静电感应原理展开全流量磨粒监测技术研究。研究润滑条件下荷电磨粒的产生机理和全流量磨粒静电监测原理,着重进行了静电信号特征提取方法研究。为验证磨粒静电监测技术的可行性,设计并搭建了模拟实验平台,开展了故障颗粒注入实验和循环润滑条件下销盘滑动摩擦磨损实验,使用自制的全流量在线磨粒静电传感器对油路中的磨粒进行监测。结果表明:静电传感器能够监测到金属、非金属等不同材料荷电磨粒;感应电压幅值与磨粒大小具有相关性;感应电压波形与荷电磨粒特性有关;静电传感器可以在线监测滑油回路中非金属摩擦副的摩擦磨损状态。     

航空发动机状态监视、故障诊断研究及验证

提出了发动机状态监视、故障诊断的理论方法并搭建了该系统的软硬件平台,为建立机载发动机健康管理系统奠定了坚实的基础。首先,建立并验证了含有健康参数的发动机线性化模型,在模型的基础上设计了用于故障诊断的卡尔曼滤波器;其次,用设计好的滤波器可以准确估计出反应发动机运行状态的不可测参数;随后又用了一组卡尔曼滤波器诊断、隔离了传感器故障;最后,介绍了该部分机载系统原理样机的软硬件配置,并进行了算法平台验证,从操作和实现方式上验证了软硬件平台。该设计满足算法需求且界面人性化、易于操作。     

基于数据挖掘的航空发动机状态参数离群点分析

利用机载数据采集系统和传感器获得的发动机状态参数往往存在粗大误差,影响故障诊断的准确性。采用数据挖掘算法中基于统计学的离群点分析,并与厂家系统的分析结果相比较,结果表明,该方法能够有效地剔除粗大误差且不会误删有用数据。     

基于LSTM-ResNet模型的时变结构损伤检测

面向火箭结构健康监测,提出了一种基于深度学习的损伤检测方法,直接将多个通道的振动数据作为输入,并基于由长短时记忆网络LSTM(Long Short-Term Memory Networks)和残差卷积神经网络ResNet(Residual Convolutional Neural Networks)组合而成的LSTM-ResNet网络进行损伤识别。其优点在于,首先利用LSTM提取信号的时间依赖特征,减轻了由某些通道信号缺失带来的影响,再利用ResNet在不损耗特征的情况下进一步提取空间特征,提高了训练效率和损伤辨识准确性。通过充液圆筒振动放水实验模拟火箭飞行状态下的燃料消耗,并基于自主构建的数据集和公用数据集对LSTM-ResNet、LSTM、ResNet以及ResNet-LSTM网络进行了训练,训练结果表明,LSTM-ResNet组合网络无论在传感器是否存在故障的情况下都具有更好的性能,损伤检测精度更高。     

地面站设备监控软件框架通用性研究

针对目前卫星地面应用系统中设备监控软件可重用性差的缺点,提出一种可扩展的通用设备监控软件框架设计方案.结合模块化和抽象接口设计思想,简化了组件开发,提高了软件的复用性.通过控制反转技术,降低了组件之间的耦合,使框架更加易于扩展,在实际应用中取得了良好的效果.     

Some additional results in the regular balloon monitoring of cosmic rays in the Earth’s atmosphere

The different types of the data recorded in the experiment of the regular balloon monitoring of cosmic rays (carried out since 1957 by Lebedev Physical Institute, Moscow, Russia, in several locations) are described. So called detailed information (the form of each pulse detected by the ground-based receiver) recorded during the last 12 years is discussed in more details. The use of these data both for getting and correcting the standard results of the experiment and for obtaining some additional information on the cosmic rays in the Earth’s atmosphere is considered.     

Time and spatial variations of the foF2(night)/foF2(day) values

The ratio of daytime and nighttime values of the foF2 critical frequency is analyzed on the basis of the data of 35 ionospheric stations of the eastern hemisphere and 7 stations of the western hemisphere. The analysis confirms the results of Danilov [Danilov, A.D. Variations in the ratio of nighttime and daytime critical frequencies of the F2 layer. Geomagn. Aeronomy, 47 (6), 551–561 (in Russian) 2007; Danilov, A.D. Time and spatial variations in the ratio of nighttime and daytime critical frequencies of the F2 layer, J. Atmos. Sol-Terr. Phys. 70, 1201–1212, 2008.] that after about 1980, a systematic behavior of this ratio with time is observed: an increase with time (a positive trend) or a decrease with time (a negative trend). The sign of this trend is shown to be governed by the signs of the magnetic declination D and magnetic inclination I at the given ionospheric station. This fact makes it possible to assume that the above indicated trend is caused by long-term variations in the zonal component V_ny of the horizontal wind in the thermosphere, the latter component contributing into the vertical drift velocity W.