Semi-supervised gear fault diagnosis using raw vibration signal based on deep learning

In aerospace industry, gears are the most common parts of a mechanical transmission system. Gear pitting faults could cause the transmission system to crash and give rise to safety disaster. It is always a challenging problem to diagnose the gear pitting condition directly through the raw signal of vibration. In this paper, a novel method named augmented deep sparse autoencoder (ADSAE) is proposed. The method can be used to diagnose the gear pitting fault with relatively few raw vibration signal data. This method is mainly based on the theory of pitting fault diagnosis and creatively combines with both data augmentation ideology and the deep sparse autoencoder algorithm for the fault diagnosis of gear wear. The effectiveness of the proposed method is validated by experiments of six types of gear pitting conditions. The results show that the ADSAE method can effectively increase the network generalization ability and robustness with very high accuracy. This method can effectively diagnose different gear pitting conditions and show the obvious trend according to the severity of gear wear faults. The results obtained by the ADSAE method proposed in this paper are compared with those obtained by other common deep learning methods. This paper provides an important insight into the field of gear fault diagnosis based on deep learning and has a potential practical application value.     

Deorbiter CubeSat mission design

This paper presents the mission design for a CubeSat-based active debris removal approach intended for transferring sizable debris objects from low-Earth orbit to a deorbit altitude of 100 km. The mission consists of a mothership spacecraft that carries and deploys several debris-removing nanosatellites, called Deorbiter CubeSats. Each Deorbiter is designed based on the utilization of an eight-unit CubeSat form factor and commercially-available components with significant flight heritage. The mothership spacecraft delivers Deorbiter CubeSats to the vicinity of a predetermined target debris, through performing a long-range rendezvous maneuver. Through a formation flying maneuver, the mothership then performs in-situ measurements of debris shape and orbital state. Upon release from the mothership, each Deorbiter CubeSat proceeds to performing a rendezvous and attachment maneuver with a debris object. Once attached to the debris, the CubeSat performs a detumbling maneuver, by which the residual angular momentum of the CubeSat-debris system is dumped using Deorbiter’s onboard reaction wheels. After stabilizing the attitude motion of the combined Deorbiter-debris system, the CubeSat proceeds to performing a deorbiting maneuver, i.e., reducing system’s altitude so much so that the bodies disintegrate and burn up due to atmospheric drag, typically at around 100 km above the Earth surface. The attitude and orbital maneuvers that are planned for the mission are described, both for the mothership and Deorbiter CubeSat. The performance of each spacecraft during their operations is investigated, using the actual performance specifications of the onboard components. The viability of the proposed debris removal approach is discussed in light of the results.     

基于梯度提升决策树的量子科学实验卫星光学实验预测

量子科学实验卫星在轨运行期间完成4种光学实验，地面监测人员通过遥测参数阈值判断卫星是否进行光学实验、实验类型及实验结果.这种方法需要预先设定大量阈值，并且这些阈值需要根据在轨卫星重新设定，可扩展性较差.针对以上问题，提出一种基于机器学习的光学实验判别方法，将量子科学实验卫星的光学实验监测任务抽象为机器学习中的多元分类问题，构建分类模型，利用量子科学实验卫星的真实历史遥测数据对模型进行训练，并通过真实实验计划对训练得到的模型进行验证.实验结果表明，本文提出的方法在没有专家先验知识的前提下，判别准确率达到99%，可用于量子科学实验卫星光学实验的实时监测任务.提出的基于机器学习的判别方法具有较强的可扩展性，可应用于卫星在轨运行的其他监测任务.      

河北省高校农村生源学生思想现状调查研究

河北省农村人口占总人口的79％，是一个比较典型的农业大省。农村生源大学生，是河北省在校大学生的重要组成部分，占生源的最大比例。大学生是宝贵的人才资源，他们的素质尤其是思想政治素质如何，他们能否成为社会主义事业的合格建设者和可靠接班人，是一个非常重要的问题。因而，科学分析和把握他们的思想政治状况，将会增强我们把握大学生思想脉搏的准确性和说服力。为此笔者选取了河北省八个地市的不同高校，进行了问卷调查、座谈和访谈。对河北省农村生源大学生的思想现状进行了调查。通过统计软件的录入和整理，统计结果比较真实地显示了当前河北省农村生源大学生的思想现状。     

HML系统在商店偏好决策模型中的应用研究

对混合式机器学习系统（HML）进行了全面系统的介绍，并将其运用于服装消费商店偏好决策行为的研究中。采用上海统计局家庭调查网络，对300户家庭进行抽样分析后发现，服装消费商店偏好的主要影响因素是地区、季节、丈夫和妻子的学历、职业以及子女性别。在此基础上，将HML分析所得的结论与传统的研究方法和结果进行了系统比较：从方法上来看，因为属性变量包含间断变量和连续变量两种，因此传统统计分析要运用两种不同的检测方法对影响因素的相关性作出判断，结果需要经过统计学分析，才能得到结论；而HML分析结果比较直观和简单，便于理解。     

以培养适用型人才为宗旨建立新型的校外学习中心管理体系

校外学习中心是远程教学和远程学习的活动基地。校外学习中心必须能适应市场经济形势需求,以培养适用型人才为宗旨,不断完善自身机制,实行规范化管理,不断提高教学管理工作水平,建立起新型的管理体系。     

Astrophysics in 2006

The fastest pulsar and the slowest nova; the oldest galaxies and the youngest stars; the weirdest life forms and the commonest dwarfs; the highest energy particles and the lowest energy photons. These were some of the extremes of Astrophysics 2006. We attempt also to bring you updates on things of which there is currently only one (habitable planets, the Sun, and the Universe) and others of which there are always many, like meteors and molecules, black holes and binaries.     

基于特征模型的挠性悬臂板控制

针对航天器上太阳帆板这种悬臂外伸薄板结构的挠性附件 ,存在环境扰动下所引起的振动 ,本文采用压电智能结构作为执行器对悬臂板进行主动振动控制。基于板系统的特征建模 ,并结合自适应控制对挠性板的主动控制进行了研究。通过仿真研究结果与应变律反馈控制比较 ,表明该方法的有效性     

改进的P型迭代学习控制算法

针对多输入多输出的定常线性系统的迭代学习控制问题, 给出改进的P型迭代学习控制算法, 该算法中利用最新算出的控制分量代替旧的控制分量, 这样可以加快控制输入的收敛速度, 利用该算法进行学习控制, 使系统的实际输出以更快的速度收敛于系统的理想输出.