Subsonic flutter suppression using self-straining actuators

Analytical (rather than purely computational) study, using full continuum models of bending-torsion wing flutter in inviscid subsonic aerodynamics. Performance evaluation of self-straining actuators for flutter suppression. Time domain analysis — unsteady aerodynamics — using the theory of semigroups of operators. Root locus of aeroelastic modes and precise operational definition of flutter speed. Analytical relationship of flutter speed to structure parameters too complex. Numerical results for the Goland wing model with torsion mode flutter. Finding: while effective in increasing structure damping prior to flutter, self-straining controllers have little or no effect on the flutter speed.     

Bending-torsional flutter of a cantilevered wing containing a tip mass and subjected to a transverse follower force

This paper is a study of the bending-torsional flutter of a cantilevered wing subjected to a follower force, and containing a lumped mass, at the free end. In addition, a distributed aerodynamic loading is introduced along the wing. This results in a set of nonself-adjoint differential equations with variable, complex coefficients whose solutions are obtainable only in series form. Using the Frobenius method, a direct procedure is employed which retains the exact expression of the Theodorsen function and the unknown coefficients are evaluated on a computer which numerically converge to any prescribed accuracy. It is found that, as a result of the interaction of the two sources of non-conservative loadings, the follower force reduces considerably the critical speed of flows in all cases studied. An increase in the tip mass, however, has a stabilizing influence. The effect of structural damping is also examined and it is shown that internal damping forces may have pronounced influences on the flutter speed of the system.     

The transonic dip in aeroelastic divergence speed—An explicit formula

Using continuum models we develop a closed form analytical formula for the divergence speed as a function of the Mach number and in particular its dependence on wing camber. There is a transonic dip even when the angle of attack is zero and it depends on the wing thickness ratio. For nonzero camber the angle of attack in fact plays a lesser role. The main analytical tool is the Possio integral equation which is shown to have an explicit solution. Neither CFD nor FEM is employed.     

The design of networked distributed sliding mode controllers with non-ideal links

In this paper, a methodology is proposed for the design of Networked Distributed Sliding Mode Controllers (NDSMC) for large-scale systems consisting of interconnected discrete-time subsystems. The state of each of these subsystems has time varying delays and there is also delay and packet loss in their interconnection links. The Networked Distributed Control System (NDCS), which is subject to the non-idealities of these interconnection links, is modeled and a Linear Matrix Inequality (LMI) based sufficient condition for designing asymptotically stable sliding surfaces, as well as the stabilizing NDSMC, is proposed. Finally, the proposed NDSMC structure is employed for the formation control of Unmanned Aerial Vehicles (UAVs); in which five vehicles are flying in a leader–follower architecture, in two platoons, whilst attempting to maintain the desired speed with a constant desired Euclidean distance between successive vehicles despite the presence of external perturbations such as wind disturbances.     

未来交通飞行车设计

为了解决不断增长的汽车数量导致道路拥挤的问题,本文开发设计一种新型陆空两用的飞行车,实现陆地上能像汽车一样行驶,当道路拥堵时,可以垂直起降到畅通路段的基本功能,有效解决因道路堵塞给人们带来的困扰。这是一种基于自主巡航系统开发的三轴飞行车,本文主要论述了飞行车的初步设计与建模,包括对飞行车各部件进行初步设计、飞行车各部件的建模。     

鱼雷静不稳定状态下纵向三回路过载控制与仿真研究

静不稳定控制技术作为现代航空航天、水中兵器主动控制技术,在近些年逐渐成为研究热点。本文针对某静不稳定鱼雷,建立了纵向运动模型,运用最优控制理论设计了俯仰通道三回路过载控制系统,最后对设计出的控制器进行了特征点控制系统仿真。结果说明,三回路过载控制方法能较好地控制静不稳定鱼雷。     

“狗咬邮递员”查询研究

"狗咬邮递员"查询是移动点最近邻查询问题的派生问题,是要找到当狗和邮递员以某种恒定速度运动时狗能最快到达的邮递员集合。与一般的移动点最近邻查询不同,"狗咬邮递员"查询中,移动点以恒定速度运动。针对"狗咬邮递员"问题的特殊性,提出在Voronoi图基础上引入三维坐标系并构建移动Voronoi图,利用反向最近邻查询思想,通过检验内部区域和外部区域是否被破坏从而确定查询点能最快到达的点是否因为点的移动而发生改变,最终找到每一时刻移动查询点所能最快到达的点。     

CFM56-3发动机N2转速管理系统原理

CFM56-3系列发动机是由美国通用电气公司(GE)和法国国营航空发动机研制公司(SNECMA)组成的国际公司(CFMI)研制生产的为波音B737-300/400/500型航空器翼吊发动机;其N2转速管理系统为飞机提供精确的燃油控制功能以防止发动机超速及失速情况的发生。N2转速管理系统主要通过接受发动机的核心转子转速(N2)和动力角(PLA)来控制燃油流量使发动机实际转速与目标转速相符。     

后扰流板角度对汽车尾部受力影响的分析

以捷达为例通过分析在不同速度下,汽车后扰流板起升不同角度时,汽车所受到的气动升力和气动阻力的关系,以及由扰流板本身带来的形状阻力对汽车受力的综合影响,得出在满足最低的气动力系数条件下的最优角度。     

Exploring the effect of overlapping institutional applications on panel decision-making

The assessing of a pool of competing projects is a challenging task for scientific panels at funding agencies. Using large-scale and rich data from a broad set of panels from the UK's largest research council, we explore whether scientific panels treat overlapping applications from the same institution in a selection round differently. Building on previous research that suggested that panels may favour institutional diversity in funding outcomes, we find that applications at the margin of the funding decision are less likely to be funded when an application from the same institution is also funded. However, we find little evidence of welfare losses associated with this outcome. The implications of these findings for understanding the role of panels in shaping of scientific funding are also examined.     

上市公司自由现金流与资本结构动态调整研究

本文采用我国沪深两市A股非金融上市公司2004–2012年的财务数据,首先通过动态面板系统GMM模型,估算出我国上市公司的最优资本结构,并计算出资本结构调整速度为0.186;然后建立了反映自由现金流盈亏程度和资本结构偏离程度如何影响资本结构调整的固定效应模型,分析得出与单独调整资本结构的情况相比,企业同时调整资本结构和自由现金流时,调整成本降低,调整速度加快;当资本结构调整至(或接近)最优状态后,企业会保持资本结构在最优水平,不受自由现金流调整的影响。上述结论对企业是否存在融资约束具有稳健性,但受融资约束的企业调整成本相对较高,调整速度相对较慢。     

基于特征矩阵的高效数字识别算法

传统的数字识别算法存在识别速度、识别准确率和识别方法复杂度三者无法兼顾的问题,为解决该问题,提出了基于特征矩阵的高效数字识别算法。该算法首先在预处理的基础上获取字符的特征矩阵,然后用特征矩阵对字符的特征横线、竖线等特征进行提取,最后利用结构语句识别的方法实现数字识别。实验结果表明,基于特征矩阵的高效数字识别算法思路简单、速度快,且识别率达97％以上。     

Catching up with the market leader: Does it pay to rapidly imitate its innovations?

In this study we attempt to shed more light on the relationship between speed of new technology imitation and the sales performance of the imitator compared to the innovator, with a particular focus on the performance outcomes resulting from the rapid imitation of technologies introduced by the market leader. Using data on handset technologies mounted on more than 600 devices introduced to the UK market by 14 mobile phone vendors operating from 1997 to 2008, we study hundreds of imitative actions to test hypotheses on the extent to which an imitator can catch up (i.e., reduce the market share gap) with the market leader by rapidly imitating its innovations. First, we show that gaining advantage by rapidly imitating a technology pioneer is contingent on whether the pioneer is the market leader or a non-leader rival. Second, we find that the risks of rapid imitation of the market leader's technologies are mitigated when industry clockspeed is high, i.e., during a period of fast innovation and imitation cycles in an industry, resulting in rapid variations in product design. Third, we observe that the degree of competitive responsiveness of the technology pioneer when its innovations are imitated represents an important mechanism that can explain why speed of imitation may affect how an imitator can improve its market share gains relative to the pioneer. This paper advances competitive dynamics and imitation as predictive theories of how rapid imitators might catch up with market leaders in technology-intensive industries.     

价值性差异、知识域结构与知识计量研究——基于知识的一般性经济原理

 知识分类的混沌现象,不能较好解释知识的动态转化与经济之间的关系。因此,根据知识拥有的主体不同,从异质动态多样性、层次性、结构性并反映知识价值差异性特征角度,对具有整合、交易价值属性的不同层次、不同形态的同质知识和异质知识,按照时间序列和空间结构关系,从静态到动态螺旋上升整合过程,提出了可用于规范经济学理论阐述的静态知识、静态点知识、个体动态点知识、个体结构知识、个体边际结构知识、组织结构知识、组织边际结构知识等知识域纳入到经济学原理框架下分析、评价和计量,并揭示了知识租金在相对溢出和绝对溢出效应影响下,知识经济战略选择是专业化技术跳跃的创新产品知识速度经济→持续改进与学习的知识固本规模经济→研发专业化更新产品的速度经济交替循环战略发展组合过程,是知识经济规模总量与速度经济净增加量交替转换竞争战略关系.     

超强台风“桑美”(0608)近海观测特征研究

利用气象现代化手段观测的数据进行分析。从风云2C红外云图可以发现,在近海"桑美"台风强度加强过程中,台风外围云系被台风外核区吸收,风暴直径明显减小,结构更为密实,台风眼由模糊发展为清晰;在迅速加强阶段,台风迅速向内收缩,呈现眼壁上螺旋雨带的发展特征。根据温州新一代天气雷达CINRAD/SA的组合反射率和径向速度产品揭示了直到登陆前1 h,桑美台风的强度仍在加强的事实,从组合反射率可见,台风从单眼壁发展成双眼壁,之后发展成具备更加深厚眼壁的台风结构,此过程对应于雷达径向速度图表现为,从一个大速度区发展为二个大速度区,直至更大的速度区。从登陆前后几小时地面测站的变压计算气压梯度可以判断强风区大小,结合变压及小时极大风风向变化可判断登陆点,以及作出登陆后移动路径的短时预报;小时累计降水分布表明,其具有明显不对称分布特征,前进方向右侧大于左侧,强降水主要集中在台风环流主体区,外围降水不明显。     

Coarse synchronization of UWB signals in the dense multipath channel based on the golden section

In this paper we address the problem of coarse synchronization of ultra wideband signals in the dense multipath channel, following the widely used hypothesis testing approach in which the uncertainty region for the timing offset is divided into a certain number of cells. We propose a novel strategy for the search of the correct cells, within the constant increment perspective. We show that under a simple model the proposed strategy achieves practically the minimum possible mean acquisition time, independently of the length of the channel impulse response, as opposed to previous constant increment approaches. A comparative analysis of the proposed strategy and another convenient one reveals that, strictly speaking, none of them can be considered optimum from the point of view of the mean acquisition time. The proposed method is advantageous from the point of view of computational complexity in the general case. Simulations with a realistic channel model confirm that the proposed strategy is very appropriate for the problem of coarse synchronization of UWB signals.     

基于双层哈希表的中文分词算法优化

采用基于词典的正向增字最大匹配算法,分词词典采用改进的双层哈希表加动态数组的数据结构。在不提升已有典型词典机制空间复杂度与维护复杂度的情况下,一定程度上提高了中文分词的速度和效率。     

Effect of pulse direct current signals on electrotactic movement of nematodes Caenorhabditis elegans and Caenorhabditis briggsae

The nematodes (worms) Caenorhabditiselegans and Caenorhabditisbriggsae are well-known model organisms to study the basis of animal development and behaviour. Their sinusoidal pattern of movement is highly stereotypic and serves as a tool to monitor defects in neurons and muscles that control movement. Until recently, a simple yet robust method to initiate movement response on-demand did not exist. We have found that the electrical stimulation in a microfluidic channel, using constant DC electric field, induces movement (termed electrotaxis) that is instantaneous, precise, sensitive, and fully penetrant. We have further characterized this behaviour and, in this paper, demonstrate that electrotaxis can also be induced using a pulse DC electric signal. Worms responded to pulse DC signals with as low as 30% duty cycle by moving towards the negative electrode at the same speed as constant DC fields (average speed of C. elegans?=?296?±?43?μm/s and C. briggsae?=?356?±?20?μm/s, for both constant and pulse DC electric fields with various frequencies). C. briggsae was found to be more sensitive to electric signals compared to C. elegans. We also investigated the turning response of worms to a change in the direction of constant and pulse DC signals. The response for constant DC signal was found to be instantaneous and similar for most worms. However, in the case of pulse DC signal, alterations in duty cycle affected the turning response time as well as the number of responding worms. Our findings show that pulse DC method allows quantitative measurement of response behaviour of worms and suggest that it could be used as a tool to study the neuronal basis of such a behaviour that is not observed under constant DC conditions.     

Distributed cruise control of high-speed trains

In this work, the cruise control problem of high-speed trains’ movements is investigated. Both cases of a single high-speed train and multiple high-speed trains are under consideration. Different with most existing studies where the centralized control or the decentralized control methods are adopted based on a single point mass model of the train, in this paper, a distributed control mechanism is proposed by virtue of the graph theory, and the high-speed train’s model is built as a cascade of point masses connected by flexible couplers. For a single high-speed train, the neighboring cars interact through the coupling force with each other, which can be described by a connected topological graph by regarding each car as a node. Besides, the speed information communication among the cars is considered to be described by another directed topological graph. A distributed control strategy is then developed, with which all the cars of a train track a desired speed asymptotically and the neighboring cars keep a safety distance from each other. For the multiple high-speed trains running on a railway line, the in-train force interaction topology and the speed information communication topology of all the trains are more complex than those of a single train. A new cluster consensus technique is developed, by which a distributed control law is designed. Under the control law, the trains can track the desired speeds asymptotically, the headway distance between adjacent trains and the distance between the neighboring cars of a train can be kept in appropriate ranges. Finally, simulations are provided to illustrate the effectiveness of the obtained theoretical results.