Development of a thin section device for space exploration: Overview and system performance estimates

In this paper we present a conceptual design of a spaceborne instrument for the in situ production of rock thin sections on planetary surfaces. The in situ Automated Rock Thin Section Instrument (IS-ARTS) conceptual design demonstrates that the in situ production of thin sections on a planetary body is a plausible new instrument capability for future planetary exploration. Thin section analysis would reduce much ambiguity in the geological history of a sampled site that is present with instruments currently flown. The technical challenge of producing a thin section device compatible with the spacecraft environment is formidable and has been thought too technically difficult to be practical. Terrestrial thin section preparation requires a skilled petrographist, several preparation instruments that individually exceed typical spacecraft mass and power limits, and consumable materials that are not easily compatible with spaceflight. In two companion papers we present research and development work used to constrain the capabilities of IS-ARTS in the technical space compatible with the spacecraft environment. For the design configuration shown we conclude that a device can be constructed that is capable of 50 sample preparations over a 2 year lifespan with mass, power, and volume constraints compatible with current landed Mars mission configurations. The technical requirements of IS-ARTS (mass, power and number of samples produced) depend strongly on the sample mechanical properties, sample processing rate, the sample size and number of samples to be produced.     

Development of a thin section device for space exploration: Rock cutting mechanism

We have developed a rock cutting mechanism for in situ planetary exploration based on abrasive diamond impregnated wire. Performance characteristics of the rock cutter, including cutting rate on several rock types, cutting surface lifetime, and cut rock surface finish are presented. The rock cutter was developed as part of a broader effort to develop an in situ automated rock thin section (IS-ARTS) instrument. The objective of IS-ARTS was to develop an instrument capable of producing petrographic rock thin sections on a planetary science spacecraft. The rock cutting mechanism may also be useful to other planetary science missions with in situ instruments in which sub-sampling and rock surface preparation are necessary.     

自助法中若干问题研究及其在命中精度评估中的应用

在小样本的试验评估中,自助法是一种较优的小子样方法。本文在简要阐述自助法的基础上,深入分析了自助法应用中存在的若干问题,如再抽样样本的获取方式、再抽样样本量、各样本数据容量和适用范围,并给出了自助法的实施步骤,最后比较了自助法和落入概率点小样本方法在命中精度评估中的应用效果,指出了自助法的优点。     

Fault diagnosis of bearings based on deep separable convolutional neural network and spatial dropout

Bearing pitting, one of the common faults in mechanical systems, is a research hotspot in both academia and industry. Traditional fault diagnosis methods for bearings are based on manual experience with low diagnostic efficiency. This study proposes a novel bearing fault diagnosis method based on deep separable convolution and spatial dropout regularization. Deep separable convolution extracts features from the raw bearing vibration signals, during which a 3 × 1 convolutional kernel with a one-step size selects effective features by adjusting its weights. The similarity pruning process of the channel convolution and point convolution can reduce the number of parameters and calculation quantities by evaluating the size of the weights and removing the feature maps of smaller weights. The spatial dropout regularization method focuses on bearing signal fault features, improving the independence between the bearing signal features and enhancing the robustness of the model. A batch normalization algorithm is added to the convolutional layer for gradient explosion control and network stability improvement. To validate the effectiveness of the proposed method, we collect raw vibration signals from bearings in eight different health states. The experimental results show that the proposed method can effectively distinguish different pitting faults in the bearings with a better accuracy than that of other typical deep learning methods.     

Rapid impactor sample return (RISR) mission scenario

Due to the long lead time and great expense of traditional sample return mission plans to Mars or other astronomical bodies, there is a need for a new and innovative way to return materials, potentially at a lower cost. The Rapid Impactor Sample Return (RISR) mission is one such proposal. The general mission scenario involves a single pass of Mars, a Martian moon or an asteroid at high speeds (7 km/s), with the sample return vehicle skimming just 1 or 2 m above a high point (such as a top ridge on Olympus Mons on Mars) and releasing an impactor. The impactor strikes the ground, throwing up debris. The debris with roughly the same forward velocity will be captured by the sample return vehicle and returned to Earth. There is no delay or orbit in the vicinity of Mars or the asteroid: RISR is a one-pass mission. This paper discusses some of the details of the proposal. Calculations are presented that address the question of how much material can be recovered with this technique. There are concerns about the effect of Mars tenuous atmosphere. However, it will be noted that such issues do not occur for RISR style missions to Phobos, Deimos, or asteroids and Near Earth Objects (NEOs). Recent test results in the missile defense community (IFTs 6–8 in 2001, 2002) have scored direct hits at better than 1 m accuracy with closing velocities of 7.6 km/s, giving the belief that accuracy and sensing issues are developed to a point that the RISR mission scenario is feasible.     

直升机动部件小子样全尺寸疲劳试验件的x~2检验

本文将X~2检验——“检验一个小子样是否来自已知标准差的母体”方法应用于直升机动部件的疲劳试验数据统计处理中,并与传统方法获得的结论进行了对比。结果表明:同样的小子样试验数据,若运用检验合格的结论,在同样的可靠度下,能较大幅度地提高零件的疲劳寿命。     

基于模糊距离变换的骨架剪枝算法

针对传统骨架提取算法结果无法保证骨架单像素性,并且伴有毛刺产生的现象,提出了一种能够有效去除骨架中毛刺分枝,充分体现物体形态特征的骨架剪枝算法。该算法以模糊距离变换为理论基础,在物体粗骨架图像中计算每条尾枝的模糊距离变换值,使用根据骨架特点确定的动态阈值判断骨架分枝是否被去除,并利用分级剪枝方法实现骨架剪枝操作。实验结果表明,该算法在进行剪枝操作中可有效地保证骨架的准确性和连续性。     

基于大规模训练集SVM的发动机故障诊断

提出了一种新的学习策略,用于解决发动机故障诊断中大规模支持向量机(SVM)的训练问题.通过保留初始SVM分类器支持向量超平面附近的样本以及错分样本,使最终得到的约减集规模明显缩小,从而可在保持较高分类精度的前提下使训练时间明显缩短;同时,由于支持向量的数量减小,分类时间也相应缩短.探讨了序贯最小优化(SMO)算法的参数选择和实现过程中的关键问题,为这种极具潜力的算法在发动机故障诊断中的实际应用奠定了坚实的基础.仿真实例表明,这种基于大规模训练集SVM的发动机故障诊断方法有效、可靠,容易实现,可以作为工程应用的基础.      

Human spaceflight and an asteroid redirect mission: Why?

The planning of human spaceflight programmes is an exercise in careful rationing of a scarce and expensive resource. Current NASA plans are to develop the new capability for human-rated launch into space to replace the Space Transportation System (STS), more commonly known as the Space Shuttle, combined with a heavy lift capability, and followed by an eventual Mars mission. As an intermediate step towards Mars, NASA proposes to venture beyond Low Earth Orbit to cis-lunar space to visit a small asteroid which will be captured and moved to lunar orbit by a separate robotic mission. The rationale for this and how to garner support from the scientific community for such an asteroid mission are discussed. Key points that emerge are that a programme usually has greater legitimacy when it emerges from public debate, mostly via a Presidential Commission, a report by the National Research Council or a Decadal Review of science goals etc. Also, human spaceflight missions need to have support from a wide range of interested communities. Accordingly, an outline scientific case for a human visit to an asteroid is made. Further, it is argued here that the scientific interest in an asteroid mission needs to be included early in the planning stages, so that the appropriate capabilities (here the need for drilling cores and carrying equipment to, and returning samples from, the asteroid) can be included.     

火炮首发命中率试验脱靶弹数据处理方法

通过对出现脱靶弹情况下的散布特征参数估计方法进行研究,建立正态双边截尾样本均值和方差估计数学模型,并给出了数值计算方法,为火炮武器系统首发命中概率试验脱靶提供了理论依据。