复杂时变激励下失谐叶盘瞬态强迫响应分析

提出一种高效的失谐叶盘瞬态强迫响应分析方法，不同于传统的数值积分方法，该方法推导出瞬态强迫响应的解析表达式，能更为高效地预测失谐叶盘的瞬态强迫响应。首先，对叶盘的高保真有限元模型进行减缩建模，在精确地描述叶盘结构的动力学特性的前提下，极大的减少了模型的自由度数目。其次，模拟加速旋转的涡轮叶盘经过复杂流场时叶片表面上的气动载荷，并建立叶盘固有频率和振型随转速变化的数学函数；通过共振分析确定叶盘共振的转速区间并分析引起共振的激励阶次成分。最后，计算了不同旋转加速度和阻尼下叶盘的瞬态强迫响应，并对叶盘的失谐幅值放大因子进行研究。应用本办法对某86个叶片的涡轮叶盘进行了数值分析，结果表明，相同阻尼水平下，叶盘的瞬态强迫响应幅值随旋转加速度增加而降低，失谐幅值放大因子在瞬态条件下大于稳态条件下，最高可达30%。     

飞机高升力系统扭力管旋转角度测试方法

在飞机地面模拟试验环境下需要测试高升力系统襟翼扭力管旋转角度。由于扭力管属于系统试验件,在不改变试验环境构型的情况下不可截断或更换安装用于测试扭力管旋转角度的旋转编码器。本文利用同步轮法实现了在不破坏扭力管前提下安装旋转编码器测试扭力管旋转角度,方法简单、可靠,且容易实现,很好的满足了试验大纲要求。     

一种圆锥运动条件下的等效旋转矢量算法研究

传统旋转矢量姿态算法一般采用陀螺的角增量信号来构造积分算法,当应用于输出为角速率的光纤陀螺捷联系统时,通过角速率提取角增量,算法会损失一定精度。提出了一种以陀螺角速率信号与角增量信号同时作为输入的改进旋转矢量姿态算法,进一步补偿了圆锥误差,提高了计算精度。仿真结果表明,该算法与传统二子样算法相比较,计算量相当,姿态精度、速度精度、位置精度有大幅度的提升。     

一种小天体旋转参数估计方法

小天体旋转参数是科学数据，也是小天体测绘，着陆导航的基础数据.研究一种在小天体探测接近段过程中使用的基于图像上特征点跟踪和扩展卡尔曼滤波器的小天体旋转参数估计方法.该方法首先建立小天体旋转关系模型，表示小天体在相机坐标系中的姿态变化；然后定义小天体旋转的状态方程，推到了扩展卡尔曼滤波器的计算过程.通过对观测图像序列上的特征点跟踪，利用扩展卡尔曼滤波器方法得到小天体旋转轴指向和自转角速度估计.实验中，在仿真相机与小天体相机100 km距离上，分析了相机坐标系小天体坐标系之间的四元数初值，图像上特征点跟踪精度，相机的观测指向等因素对小天体旋转参数估计精度的影响.实验结果表明，提出的基于图像特征点跟踪和扩展卡尔曼滤波器的小天体旋转参数估计方法能够得到具有较高精度的估计结果.     

基于图像识别的机械振动信号特征提取与寿命预测方法研究

滚动轴承作为许多机械设备的关键组件,被广泛应用于机械制造、航空航天等领域,其健康状态直接影响了相应设备的剩余寿命,因此在设备故障预测与健康管理(Prognostics and Health Management, PHM)领域,滚动轴承寿命预测具有很高的研究价值。目前基于数据驱动的轴承寿命预测方法主要利用特征提取并构造健康因子(Health Indicator, HI),然而在这一过程中特征的选择与融合依然依赖于专家先验知识,并且健康因子也很难从复杂的时序数据中进行提取。因此,提出了一种新型的数据驱动寿命预测算法,在特征提取方面,通过连续小波变换(Continuous Wavelet Transform,CWT)将传感器振动信号转换为时频谱图,再通过深度残差网络(Deep residual network, ResNet)结合时空卷积网络(Temporal Convolutional Network, TCN)将时频谱图中的时域频域特征构造成为健康因子,最后完成剩余寿命预测。本研究在PRONOSTIA数据集上与现有的数据驱动算法进行了对比,证明了该算法可以更准确地完成剩余寿命预测。     

The role of motor processes in mental rotation: selective shaping of cognitive processing via specific sensorimotor experience

The involvement of motor processes in mental rotation is experience-dependent: different levels of expertise in sensorimotor interactions lead to different strategies in mental rotation. In the present study, wrestlers, gymnasts, and nonathletes physically rotated objects that were either light (wooden) or heavy (lead) but otherwise having the same sizes and shapes. They then performed a mental rotation task using photographs of these objects in which the material and therefore the weight was visible. I hypothesized that wrestlers would rely more heavily on experience-based sensorimotor strategies in performing mental rotation because during their athletic practice they not only manipulate external “objects” (i.e., their opponent) but also have to plan future actions taking into account past experience of these “objects” (for example their weight). All participants reported that lead objects were harder to physically rotate than wooden ones. However, only wrestlers mentally rotated lead objects more slowly than wooden ones—as they would if they were physically rotating them—suggesting the involvement of motor processes. These findings show that the involvement of motor processes in mental rotation depends on specific rather than mere sensorimotor experience.     

Mechanism of error propagation from the subdaily Universal Time model into the celestial pole offsets estimated by VLBI

Within the analysis of space geodetic observations, errors of the applied subdaily Earth rotation model can induce systematic effects in different estimated parameters. In this paper, we focus on the impact of the subdaily Universal Time (UT1) model on the celestial pole offsets (CPO) estimated from very long baseline interferometry (VLBI) observations. We provide a mechanism that describes the error propagation from the subdaily UT1 into the daily CPO.In typical 24-h VLBI sessions the observed quasars are well distributed over the sky. But the observations, if looked at from the Earth-fixed frame, are not homogeneously distributed. The amount of observations performed in different terrestrial directions shows an irregularity which can be roughly compared to the case where the observations are collected in only one Earth-fixed direction. This peculiarity leads to artefacts in VLBI solutions, producing a correlation between the subdaily variations in UT1 and the position of the celestial pole. As a result errors in diurnal terms of the subdaily UT1 model are partly compensated by the estimated CPO. We compute for each 24-h VLBI session from 1990 until 2011 the theoretical response of the CPO to an error in the subdaily UT1 by setting up a least-squares adjustment model and using as input the coordinates of the observed quasars and observation epochs. Then real observed response of the estimated CPO derived from the VLBI session solutions is compared to the predicted one. A very good agreement between the CPO values estimated from VLBI and the predicted values was achieved. The presented model of error propagation from the subdaily UT1 into the daily CPO allows to predict and explain the behaviour of CPO estimates of VLBI solutions computed with different subdaily Earth rotation models, what can be helpful for testing the accuracy of different subdaily tidal models.     

基于室内无线模型的穿墙损耗校正

K-M（Keenan-Motley）模型将单墙固定损耗值相加来计算室内无线信号穿透多墙的总损耗值，存在较大误差。针对该问题选取多种室内场景分别进行连续波（Continuous Wave，CW）测试，对无线信号穿墙损耗的影响因素和变化规律进行分析，提出了一种基于人工神经网络的室内无线模型穿墙损耗校正方法，对预处理后的测试数据进行训练并建立预测模型。经验证该预测模型符合校正判别准则，在实际场景下具有良好的预测准确度。     

锚固站数量及布局对导航星座自主定轨影响分析

针对导航星座自主定轨中的星座整体旋转问题,采用增设少量地面锚固站的方法可有效解决该问题。通过推导星地距离对卫星轨道升交点赤经的偏导数,证明了星地距离对卫星轨道升交点赤经可观。仅考虑在我国大陆范围内布设锚固站的条件下,仿真分析了锚固站数量以及布局对导航星座自主定轨精度的影响。仿真实验结果表明:采用3个以上的锚固站,即可有效控制星座整体旋转,在14d的仿真时段内卫星自主定轨精度保持4m以内;锚固站数量越多,自主定轨精度越高,但随着锚固站数量的增加,自主定轨精度改善程度越来越小;在保持4个锚固站的情形下,采用不同的锚固站布局方案,自主定轨精度并无明显差别。