基于视觉识别的电解加工多孔表面评定方法研究

针对电解加工多孔表面特点,提出表面形貌评价指标,获取多孔表面图像并对其进行预处理,提取腐蚀率、不均匀度、灰度等特征参数,揭示各特征参数与电解加工多孔表面形貌的相关性,并通过实例证明该方法适用于电解加工多孔表面的测量。     

基于聚偏二氟乙烯(PVDF)压电薄膜的密封舱壁穿孔损伤识别技术

日益恶化的空间碎片环境将严重威胁空间站及航天员的在轨安全。文章基于聚偏二氟乙烯（polyvinylidene fluoride,PVDF）压电薄膜设计了一种密封舱壁穿孔损伤识别技术方案,用于识别空间碎片超高速撞击对航天器密封舱壁造成的损伤模式,可为航天员合理选择应急措施提供依据。首先,发射超高速弹丸穿透铝合金靶板以模拟密封舱壁被击穿损伤的情况,形成的碎片云撞击PVDF压电薄膜探头,利用信号采集设备获取高速撞击引起的去极化效应信号,识别该信号的频率特征;其次,对PVDF压电薄膜探头及其支撑结构进行敲击试验,模拟在轨运行时因碰撞产生的干扰信号,掌握此类干扰信号的频率特征。试验结果表明,当系统采样频率为20 MHz时：1）所获得的探头去极化效应信号具有极为陡峭的上升沿,且上升沿的时长为亚μs量级;2）去极化效应信号主要由1 MHz以下的信号组成,但也包含少量的3～10 MHz高频成分;3）敲击探头及其支撑结构所产生的干扰信号频率在20 kHz以下。可根据频率差异进行两种信号的识别。     

An unsupervised pattern recognition methodology based on factor analysis and a genetic-DBSCAN algorithm to infer operational conditions from strain measurements in structural applications

Structural Health Monitoring (SHM) suggests the use of machine learning algorithms with the aim of understanding specific behaviors in a structural system. This work introduces a pattern recognition methodology for operational condition clustering in a structure sample using the well known Density Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm. The methodology was validated using a data set from an experiment with 32 Fiber Bragg Gratings bonded to an aluminum beam placed in cantilever and submitted to cyclic bending loads under 13 different operational conditions (pitch angles). Further, the computational cost and precision of the machine learning pipeline called FA + GA-DBSCAN (which employs a combination of machine learning techniques including factor analysis for dimensionality reduction and a genetic algorithm for the automatic selection of initial parameters of DBSCAN) was measured. The obtained results have shown a good performance, detecting 12 of 13 operational conditions, with an overall precision over 90%.     

北京上空高分辨率气球探空观测的温度垂直波数谱

使用分辨率为10 m的温度剖面检查2.90～8.01 km和14.65～19.76 km高度范围内归一化温度扰动谱的谱特性,并且将它们与模式谱比较.结果表明,在波数9.8×10-4～2.5×10-2m-1范围内,对流层的垂直波数谱有约-1.9的平均谱斜率;低平流层的垂直波数谱有约-2.2的平均谱斜率.这两个平均谱斜率大大偏离了目前饱和谱模式和普适大气谱模式表示的-3.0和-2.4谱斜率,并且认为是目前气球测量中曾经观测到的最平谱斜率.在波数9.8×10-4～2.5×10-2m-1范围内,对流层和低平流层的平均谱振幅比饱和谱模式的谱振幅分别大24倍和5倍,这也与饱和谱模式以及目前文献中的观测结果有很大不同.这些较大平均谱振幅与最平的平均谱斜率一道共同表明,观测的温度谱并不遵循目前的饱和谱模式和普适大气谱模式.      

基于模糊理论的复合材料超声检测缺陷类型识别

本文以复合材料常见缺陷分层、孔隙、疏松的超声波检测缺陷信号为研究对象,选取缺陷波的时域及频域特征,考虑到各特征量之间的不可避免地存在耦合关系,而且对缺陷区分的重要性、敏感度也有很大的差别,采用加权多面体隶属度概念构造隶属函数,依据最大隶属原则进行模式识别,从而区分复合材料缺陷类型。     

数字卫星电视信号模型及性能分析

在基于数字卫星电视信号的无源探测系统中,分析其波形特点,探讨其用于雷达照射源信号的性能,对于整个系统研究来说,具有重要的意义。文中以“鑫诺一号”卫星为例,给出了基于“鑫诺一号”卫星参数为数字卫星电视信号的模型,并对该模型进行了仿真。仿真结果表明,数字卫星直播电视信号作为无源探测雷达的照射源信号具有较强的适应性,是一种优良的照射源信号。     

基于局部平稳AR模型的信号结构变化点精确估计方法

研究了局部平稳AR模型,提出了基于递归辅助变量（RIV）双格型滤波算法,通过统计分析信息量AIC的变化精确估计信号结构变化点。对遥测速变参数的分析表明,该方法能够精确定位信号结构的变化点,实现局部平稳信号的精确分割。     

基于模糊粗集的航空发动机特征参数提取算法

提取发动机性能状态的特征参数,是提高发动机故障识别正确率和可靠性的必要条件.针对实际航空发动机故障参数所具有的模糊和连续性特点,提出了一种基于模糊粗糙集的特征参数提取算法,并应用到某型航空发动机故障识别.研究结果表明:属性约简的核则为导致发动机故障的特征参数,以此特征参数进行故障诊断,可保证较高的诊断精度;同时,该算法的抗干扰性提高了整个系统故障识别的正确率.该算法可用于航空发动机故障分类、故障诊断以及状态监控.      

考虑动态控制分配的空天飞行器再入姿态复合控制设计

针对空天飞行器再入大气层阶段的直接力/气动力复合控制分配问题,设计了一种基于改进指标函数的动态控制分配律,兼顾了反作用控制系统燃料消耗和闭环控制系统响应速度需求。此外,针对空天飞行器面对称大攻角再入引发的横侧向气动强耦合问题,在预测控制律中引入了通道间交叉耦合反馈项,对姿控过程进行增稳。对比仿真结果表明,动态分配算法在提高了再入姿态控制精度的同时降低了反作用控制系统的燃料消耗;交叉耦合反馈项的引入,缓解了倾侧角指令突变对偏航通道稳定性的干扰影响。     

Understanding the mineralogy of a compositional anomaly north of the Korolev basin using Moon mineralogy mapper

The work attempts to understand the mineralogy of the reported geochemical anomaly located in the north – northeast region of the Korolev basin using Moon Mineralogy Mapper (M³) onboard Chandrayaan ?1 and other lunar datasets. To understand the mineralogy, colour composite images using integrated band depth parameters and mineral indices were prepared, and the M³ spectral signatures corresponding to the unique colours of these colours composites were investigated. Further, Modified Gaussian Model (MGM) deconvolution was applied to these spectra. The results of spectra studies reveal that the area is made up of heterogeneous lithologies, predominantly composed of anorthosite along with minor occurrences of pyroxene-bearing (both low-calcium and high-calcium variety) and spinel-bearing lithologies. Correlation of spectral studies with the morphology revealed that pyroxene was typically associated with fresh craters and their ejecta. Spinel was found to be ubiquitous and is well-dispersed, possibly distributed along with the ejecta blanket of large impact craters. This association hints that the pyroxene-hosted layer most likely occurs beneath the spinel-bearing layer. Such observed assemblages may have resulted from physical mixing during the impact cratering events. This mixed lithology could arise due to the mafic mineral-bearing ejecta of the South-Pole Aitken (SPA) basin and spinel-bearing Orientale ejecta. Korolev most likely impacted on a thick layer of SPA ejecta, and impact basins such as Hertzsprung and Orientale have occurred post-Korolev formation. Orientale being the youngest of the large impact basins, its ejecta carrying the light plain material would have overprinted the older SPA ejecta. Smaller impact craters would have churned the ejecta so that presently we observe a composite lithology with a variable abundance of pyroxene and spinel.