摘要
电力工业的发展关乎国计民生，与之相配套的输变电设备则为现代坚强电网之骨骼。开关电器作为重要的输配电设备之一，在开断过程中不可避免地会产生电弧，其在毫秒级时间、毫米级空间内快速动态变化，从而影响开关开断性能。电弧作为物质形态的第四态，其衍生机理和发展规律是国内外研究者们普遍关注的研究热点与难点。本文通过引入混沌理论，从混沌时空域探究不同介质开关电弧之共性机理，以发现电弧之内在本质规律，实现电弧调控。
基于黑盒电弧模型理论分析与实验的低压空气电弧混沌特性研究。自主搭建220V空气电弧发生装置，以实际负载为研究对象，采用混沌相空间重构理论、C-C方法、Wolf算法和Cao方法，对纯阻性、阻感性负载下空气电弧电流动态变化进行定量分析，找出其混沌吸引子及通往混沌的道路；引入Cassie电弧模型，采用理论分析与实测数据比照，发现并描述其混沌特征。
基于三维耦合场域分析与合成回路试验线路模拟的中压真空电弧混沌特性研究。以40.5kV/25kA真空断路器为研究对象，采用有限元法进行数值模拟，得到不同触头结构型面纵向磁场时间序列，找到变触头结构与混沌特征之间的联系；引入DL标准四参数和两参数合成回路试验线路模型，将真空断路器置于试验模拟系统中，得到元件与系统中混沌行为演变；引入真空连续过渡模型，得到真空断路器开断过程中电弧行为定量描述，发现其线路中混沌特性与真空电弧吸引子形态。
基于冷气流开断的高压SF6断路器气流参数混沌变化追踪与分析。以252kV SF6断路器为研究对象，采用有限体积法与因子分析法相结合，得到开断全程中气流参数主要影响因素，分析其主元成分，得出评估气流参数性能数学描述；以550kV SF6断路器为对象，采用有限体积法对气流湍动特性进行分析，以动静触刚分时刻为时间节点，分析湍流参数在超程前后两时间段内湍流参数混沌特性，发现湍流自身所存在的间歇性，以及湍流区与非湍流区边界时空的不确定性。
基于短路大电流开断的高压SF6断路器电弧混沌行为研究。以550kV SF6断路器为研究对象，引入等效单元动态电弧模型，采用有限体积法对短路开断情况下流场进行数值模拟。通过对不同气流参数观测点时间序列提取，与冷气流开断相对比，得出高压SF6电弧混沌行为具有时空混沌特性，且为阵发混沌态。通过追踪电弧与气流运动形态，发现开断过程中，可通过增加电弧湍流区域，扩大电弧与气流相互作用区域，实现快速熄弧。通过改变喷口结构型面，得到不同喷口结构下气流参数混沌特性，得出变喷口结构是调整混沌强弱的有效途径，优化喷口结构可降低混沌效应，提高断路器开断能力，实现混沌调控与利用。
基于不同介质开关电弧时间域、相空间域特性分析的网络空间域共性研究。通过对空气电弧、真空金属蒸气电弧和SF6气体电弧参数时间序列复杂网络重构，将时间域数据以相空间理论拓展至网络空间域中，并对其所构建网络特征进行分析，以此进行不同介质形态电弧共性问题研究，发现不同介质开关时间序列网络域结构均具有无标度特性，均具有复杂网络特征。“混沌”和“复杂网络”为不同介质开关电弧理论与实验研究在时间域、相空间域与网络域之间架起一座崭新的桥梁。
Power development related to people's livelihood, and the power transmissionequipments are the bones of the modern strong power grid. Switches as one of the mostimportant electrical transmission and distribution equipments, arc is inevitably generated in themillisecond time-millimeter space during the breaking process, affecting the breakingperformance. Arc as the fourth state of physical form, the law of complicated mechanism anddevelopment is the hot spot and difficult spot that the domestic and foreign researchersconcerning. The chaos theory is introduced in order to research the common mechanism of thedifferent dielectric switches arc and discover the inherent laws of arc, and achieve the goal ofthe arc control.
Chaos characteristics research of the low voltage air arc based on the black arcmodel analysis and experiments. Using220V air arc generator device self-built, the actualloads are taken as the research objects, based on the chaotic phase space reconstruction theory,C-C method, Wolf algorithms and Cao method, the dynamic chaotic behavior of the air arccurrent is quantitatively researched under the resistive, inductive and resistive load conditions,and also the chaotic attractor and the road leading to chaos are identified. While Cassie arcmodel is introduced, compared with the measured data, the essential characteristics of the arcare revealed and described.
Chaos characteristics research by the three-dimensional field analysis and thesynthetic circuit experiment simulation of the medium voltage vacuum arc.40.5kV/25kAvacuum circuit breaker (VCB) is taken as the research object, based on the finite elementmethod (FEM) to simulate, the time series of longitudinal magnetic field with different typecontacts structures are obtained, the relationship between the structural changes and chaoscharacteristics is researched; and the introduction of two parameters and four parameters typesynthetic circuit experiment model that based on the DL standard to analyze the chaosevolution behavior of the VCB as a unit in the system. Based on the continuous transitionmodel to quantitatively describe the arc behavioral characteristics of the VCB in the process ofbreaking, and also the arc chaotic characteristics and attractor morphology are identified.
The chaos variation of gas flow parameters traced and analyzed in the high voltage SF6CB under the cold gas breaking condition. The252kV SF6CB is taken as the researchobject, and the finite volume method (FVM) and factor analysis methods are coupled used toobtain the main effects of gas flow parameters in the whole breaking process, the principalcomponent is analyzed, the mathematical description of gas flow parameters to judge the gasparameters performance is obtained; the550kV SF6CB is taken as the object, the gas flowturbulence characteristics are analyzed by use of FVM, taken the static and dynamic contactseparation time as the separation point, analyze the turbulence parameters the time before andafter super path, the turbulent parameters chaos characteristic are researched during the twoperiods, it is shown that turbulence itself is intermittent that behaving the uncertainty time andspace with turbulent and non-turbulent boundary zone.
The arc chaos behavior research of the high voltage SF6CB under the short currentbreaking condition.550kV SF6CB is taken as the object, the equivalent unit dynamic arcmodel is introduced, the gas flow field distribution under short-circuit breaking process aresimulated by use of FVM. Different observation points time series are extracted, comparedwith no load breaking condition that the high voltage SF6CB is a time-space chaos and is theintermittent chaotic state. By tracking the flow state of arc and the gas flow that it can increasethe arc turbulent region, expand the area between arc and gas flow, to achieve the rapidextinction purposes in the breaking. By changing the nozzle structures to adjust the chaos isthe effective way, and the nozzle structures are optimized which can reduce the chaotic effect,improve the CB breaking performance, and achieve the purpose of the chaos use and control.
The common research of switches arc characteristics of different dielectric in thetime domain, space domain and the net domain. By reconstructing the complex networksthrough the time series of the low voltage air arc, metal vapor vacuum arc and high voltageSF6gas arc, expand the time data to the network by use of the phase space reconstruction, alsothe complex network theory is used to analyze the network characteristic, and research thecommon problems of different dielectric morphology arc, it is shown that the networks ofdifferent switches time series have the scale-free characteristics, belong to the complexnetwork features."Chaos" and "Complex networks" as the means of different dielectric arctheory and experiment research among the time domain, space domain and net domain toestablish a new analysis and communication bridge.
引文
[1]董世勇．电力系统混沌振荡分析与抑制策略研究[D]．北京：华北电力大学，2008．
[2] Li T Y, Yorke J A. Period three implies chaos[J]. The American Mathematical Monthly,1975,82(10):985-992.
[3] Robert L Devaney. An Introduction to Chaotic Dynamical Systems[M]. New Jersey: Addison-Wesley,1989.
[4]李后强，汪富泉．分形理论及其在分子科学中的应用[M]．北京：科学出版社，1993．
[5]邹恩，李祥飞，陈建国．混沌控制及其优化应用[M]．湖南：国防科技大学出版社，2002．
[6] John A. An exploration of chaos[M]. Amsterdam: North-Holland,1994.
[7]曹志彤，郑中胜．电机运动系统的混沌特性[J]．中国电机工程学报，1998，18(5)：319-322．
[8]于东升，陈昊，卢胜利等．开关磁阻调速电机混沌特征分析与混沌扩频策略[J]．电工技术学报，2010，25(10)：66-72，80．
[9]周力行，李毅，窦鹏等．混沌—分形理论在变压器油热老化故障诊断中的应用[J]．电工技术学报，2008，23(12)：28-34．
[10]窦鹏，杨道武，李毅等．应用红外光谱和混沌学分析变压器油的老化特性[J]．高电压技术，2009，35(4)：849-854．
[11]李继胜，陈立彬，赵学风等．混沌支持向量机在变压器局部放电检测中的应用[J]．高压电器，2009，45(5)：104-111．
[12]贾宏杰，余贻鑫，王成山．电力系统混沌现象及相关研究[J]．中国电机工程学报，2001，21(7)：26-30．
[13]王宝华，杨成梧，张强．电力系统分岔与混沌研究综述[J]．电工技术学报，2005，20(7)：1-10．
[14]董世勇，鲍海，魏哲．双机电力系统中混沌振荡阀值的计算与仿真[J]．中国电机工程学报，2010，30(19)：58-63．
[15]刘劲，孙扬声，陈德树．周期参数扰动电力系统中的倍周期分叉、混沌和奇怪吸引子[J]．电网技术，1996，20(8)：1-3，8．
[16]王东风，韩璞，于朝辉．电力系统中的混沌研究与混沌应用[J]．电力科学与工程，2003，(3)：74-78．
[17]雷绍兰．基于电力负荷时间序列混沌特性的短期负荷预测方法研究[D].重庆：重庆大学，2005．
[18]刘东晖，夏承铨，史可琴等．混沌振荡与真空断路器动作特性之间相互作用的研究[J]．电工电能新技术，2007，26(3)：15-19．
[19]Ghorui S, Das A K. Theory of dynamic behavior in atmospheric pressure arc plasma devices: Part I: Theoryand system behavior[J]. IEEE Transactions on Plasma Science,2004,32(1):296-307.
[20]仲启树．基于弧声的故障电弧预测报警保护技术研究[D]．泉州：华侨大学，2008．
[21]Ghorui S, Sahasrabudhe S N, Murthy P S S, et al. Experimental evidence of chaotic behavior in atmosphericpressure arc discharge[J]. IEEE Transactions on Plasma Science,2000,28(1):253-260.
[22]施卒铭，何延庆．输配电设备发展现状和发展趋势简要分析(上)[J]．上海电气技术，2008，1(1)：49-55．
[23]施卒铭，何延庆．输配电设备发展现状和发展趋势简要分析(下)[J]．上海电气技术，2008，1(2)：57-62．
[24]李建基．高压断路器及其应用[M]．北京：中国电力出版社，2004．
[25]刘绍峻．高压电器[M]．北京：机械工业出版社，1988．
[26]李建基，彭勤荣，张汉才．国外中压真空灭弧室及其真空断路器的最新发展[J]．电气制造，2008(5)：22-29．
[27]黄润生，黄浩．混沌及其应用[M]．武汉：武汉大学出版社，2005．
[28]沈小峰．混沌初开[M]．北京：北京师范大学出版社，1993．
[29]李建基．高压开关设备的最新发展[J]．电气制造，2012(1)：27-32．
[30]郑军．国内外高压电器水平差距[J]．中国机电工业，2004(3)：32-33．
[31]施卫华，阮宁晖，王进．高压开关电器的发展及应用[J]．昆明冶金高等专科学校学报，2005，5(21)：45-51．
[32]王季梅，刘志远，修士新．国内外开发研究高压真空断路器和向超高压发展的概况[J]．电气技术，2006，(12)：5-9．
[33]王季梅，刘志远．国内外电压等级真空断路器发展动态[N]．国家电网报，2008-06-12(005)．
[34]刘晓明，王尔智，曹云东．高压SF6断路器电弧动态模型研究[J]．中国电机工程学报，2004，24(2)：102-106．
[35]刘晓明，曹云东，王尔智．高压SF6断路器电弧与气流相互作用研究[J]．中国电机工程学报，2005，25(7)：151-155．
[36]王尔智．高压SF6断路器介质恢复特性的数值模拟[M]．北京：科学出版社，2012．
[37]Lebouvier, Delalondre, Fresnet, et al. Three-dimensional unsteady MHD modeling of a low-currenthigh-voltage nontransferred DC plasma torch operating with air[J]. IEEE Transactions on Plasma Science,2011,39(9):1889-1899.
[38]Cassia A M. Theorie novelle des arc de rupture et de lagigidite des circuit[C]. CIGRE,1939:102.
[39]Mayr O. Theorie des statischen und des Dynamischen lichtbogens[J]. Arehiv fur Elektrotechnik,1943(12):588.
[40]Mirsad Kapetanovic. High voltage circuit breakers[M]. Sarajevo: ETF-Faculty of Electrorechnical Engineering,2011.
[41]Tseng K J, Wang Y. Dynamic electric arc model for electronic circuit simulation[J]. Electronics Lerters,1996,32(8):750-707.
[42]L van der Sluis，Rutgers W R，Akoreman C G. A physical arc model for the simulation of current zero behaviorof high voltage circuit breakers[J]. IEEE Transactions on Power Delivery, l992,7(2):1016-1022.
[43]L vand er Sluis, Rutgers W R. Comparison of test circuits for high voltage circuit breakers by numericalcalculations with arc models[J]. IEEE Transactions on Power Delivery,1992,7(4):2037-2045.
[44]Toshio Kamiya. Study on the simple identification and the variation of the parameters discharge in a circuitbreaker. IEEE Japan.,2001,21-B(11):1545-1552.
[45]殷凤良，胡绳荪，高忠林等．等离子体电弧数值模拟的研究进展[J]．兵器材料科学与工程，2007，30(6)：59-63．
[46]Schade E, Ragaller K. Dielectric recovery of an Axially Blown SF6arc after Current zero: PartⅠ: experimentalinvestigation[J]. IEEE Transactions on Plasma Science,1982,10(3):141-147.
[47]Schade E, Ragaller K. Dielectric Recovery of an axially blown SF6arc after current zero: PartⅡ: theoreticalinvestigation[J]. IEEE Transactions on Plasma Science,1982,10(3):148-154.
[48]Horinouchi K, Nakayama Y, Kohyama H, et al. Practical simulation method for magnetically driven arc ingas[C].11th International Conference on Gas Discharges and Their Applications,1995:102-105.
[49]刘亚芳，王其平．电弧二维数学模型及其在SF6/N2混合气体开断特性分析中的应用[J]．西安交通大学学报，1993，27(3)：33-38．
[50]李一滨，王其平．喷口电弧二维磁流体动力学(MHD)数学模型[J]．西安交通大学学报，1998，32(6)：1-4．
[51]刘晓明，袁大勇，曹云东等．SF6断路器磁流体二维电弧模型化与仿真[J]．沈阳工业大学学报，2009，31(3)：257-261．
[52]曹云东，刘晓明，王尔智．能量流电弧模型及SF6断路器在短路开断下的应用研究[J]．中国电机工程学报，2005，25(2)：93-97．
[53]王毅，王季梅．真空电弧等离子体弧柱现象模型分析[J]．中国电机工程学报，1992，12(5)：53-57．
[54]尚文凯，王季梅．大电流真空电弧收缩现象的研究[J]．中国电机工程学报，1989，9(2)：34-41．
[55]程礼椿，邹积岩．纵磁式真空开关电弧基础研究[J]．华通技术，1998，(4)：3-9．
[56]王立军，贾申利，史宗谦等．真空电弧磁流体动力学模型与仿真研究[J]．中国电机工程学报，2005，25(4)：113-118．
[57]Manfred Lindmayer, Matthias Springstubbe. Three-Dimensional-simulation of arc motion between arc runnersincluding the influence of ferromagnetic material[J]. IEEE Transactions on Components and PackingTechnologies,2002,25(3):409-414.
[58]Karetta F, Lindmayer M. Simulation of the gas dynamic and electromagnetic processes in low voltageswitching arcs[J]. IEEE Transactions on Components and Packing Manufacturing Technologies A,1998,21:96-103.
[59]Schlitz L Z, Garimella S V, Chan S H. Gas dynamics and electromagnetic processes in high-current arcplasmas[J]. Journal of Applied Physics,1999,85(5):2547-2555.
[60]杨茜，荣命哲，吴翊．低压断路器中空气电弧运动的仿真及实验研究[J]．中国电机工程学报，2006，26(15)：89-94．
[61]王艳辉．均匀大气介质阻挡放电特性及模式研究[D]．大连：大连理工大学，2006．
[62]吴彤．非线性动力学混沌理论方法及其意义[J]．清华大学学报(哲学社会科学版)，2000，15(3)：72-79．
[63]杨维明．时空混沌和耦合映像格子[M]．上海：上海科技教育出版社，1994．
[64](美) Lorenz E N著，刘式达，刘式适，严中伟译．混沌的本质[M]．北京：气象出版社，1997．
[65]程极泰．混沌的理论与应用[M]．上海：上海科学技术文献出版社，1992．
[66] Per Bak, Chao Tang, Kurt Wiesenfeld. Self-organized criticality[J]. Physical Review A,1998,38(1):364-374.
[67]吕小青，曹彪，曾敏等．短路过渡电弧的关联维数分析[J]．焊接学报，2005，26(6)：65-68．
[68]曹彪，吕小青，曾敏等．短路过渡电弧的Lyapunov指数分析[J]．材料科学与工艺，2007，15(3)：301-304．
[69]杨建红，张认成，房怀英．Lyapunov指数法在故障电弧早期探测中的应用[J]．电工电能新技术，2008，27(2)：55-58．
[70]Packard N H, Crutchfield J P, Farmer J D, et al. Geometry from a time series[J]. Physical Review Letters,1980,45(9):712-716.
[71]Takens F. Detecting strange attractor in turbulence[J]. Lecture Notes in Mathematics,1981,898(2):361-381.
[72]张筑生．微分动力系统原理[M]．北京：科学出版社，1987．
[73]陈士华，陆君安．混沌动力学初步[M]．武汉：武汉水利电力大学出版社，1998．
[74]张锁春．现代振荡反应的数学理论和数值方法[M]．郑州：河南科学技术出版社，1991．
[75]秦奕青，蔡卫东，杨炳儒．非线性时间序列的相空间重构技术研究[J]．系统仿真学报：2008，20(11)：2969-2973．
[76]Grassberger P, Procaccia I. Measuring the strangeness of strange attractors[J]. Physica D,1983,9:189-208.
[77]Kugiumtzis D. State space reconstruction parameters in the analysis of chaotic time series-the role of the timewindow length[J]. Physica D,1996,95:13-28.
[78]H S Kim, R Eykholt, J D Salas. Nonliner dynamics, delay times, and embedding windows[J]. Physica D,1999,127:48-60.
[79]吕金虎．混沌时间序列分析及其应用[M]．武汉：武汉大学出版社，2002．
[80]Wolf A, Swift J B, Swinney H L, et al. Determining Lyapunov exponents from a time series[J]. Physical D,1985:285-317.
[81]Cao Liangyue. Practical method for determining the minimum embedding dimensions of a scalar time series[J].Physical D,1997,110(1-2):43-50.
[82]Kennel M B, Brown R, Abarbanel H D. Determining embedding dimension for phase-space reconstructionusing geometrical construction[J]. Phy Rev A,1992,45(6):3403-3411.
[83]马骏，张印硕，曹占辉．基于Cao方法的空间目标RCS混沌分析[J]．现代雷达，2009，31(5)：61-64．
[84]Ruelle D, Takens F. On the nature of turbulence[J]. Commu. Math. Phys.,1971,20:167-192.
[85]沦兹．真空灭弧室纵向和横向磁场触头优化应用[J]．东北电力技术，2002，(1)：23-25．
[86]苏育生．真空灭弧室纵向磁场特性分析[D]．沈阳：沈阳工业大学，2011．
[87]杨姝．激光混沌的相空间重构研究[D]．长春：长春理工大学，2008．
[88]Andrews J G. The sheath criterion for a growing sheath[J]. Plasma Physics,1970,4(3):603-606.
[89]Varey R H, Sander K F. Dynamic sheath growth in mercury plasma[J]. Applied Physics,1969,2(2):541-550.
[90]王楠，苗迪．SPSS因子分析在企业社会责任评价中的应用[J]．价值工程，2012，(3)：112-113．
[91]贾东风，任长安．SPSS统计软件中的因子分析法在医学科研中的应用[J]．办公自动化，2009，(150)：53-55．
[92]刘菊香，沈霄凤．用SPSS统计软件对学生综合成绩的因子分析[J]．统计应用，2006，(1)：53-56．
[93]Potter M C, Wiggert D C. Mechanics of fluid [M].北京：机械工业出版社，2008．
[94]关盛楠．激波和湍流对SF6断路器介质恢复特性影响的研究[D]．沈阳：沈阳工业大学，2007．
[95]Signe Hammer，李裕清．湍流之谜[J]．世界科学，1986，(06)：8-9．
[96]Bohr T, Jensen M H, Paladin G, et al. Dynamical systems approach to turbulence[M]. London: CambridgeUniversity Press,1998.
[97]胡隐樵，陈晋北，左洪超．湍流强度定理和湍流发展的宏观机制[J]．中国科学，2007，37(2)：272-281．
[98]刘式达．孤波和湍流[M]．上海：上海科技教育出版社，1994．
[99]关盛楠，王尔智，刘海峰等．SF6高压断路器喷口中的湍流及其对介质恢复特性的影响[J]．中国电机工程学报，2007，27(36)：31-35．
[100]张建，张猛，王军．压气式SF6断路器中激波与湍流的相互作用对电弧特性影响的研究[C]．输变电年会，2010：10-15．
[101]Wissink J G.DNS of separating low Reynolds number flow in a turbine cascade with incoming wave[J].International Journal of Heat and Fluid Flow,2003,24(4):626-635.
[102]苏铭德，Friedrich R．用大涡模拟检验湍流模型[J]．应用数学和力学，1994，15(11)：991-997．
[103]李丞，李晓锋，苏雅璇等．用于建筑绕流模拟的零方程模型[J]．清华大学学报(自然科学版)，2013，53(5)：589-594．
[104]陶文铨．数值传热学[M]．西安：西安交通大学出版社，2001．
[105]Launder B E, Spalding D B. Mathematical models of turbulence[M]. USA: Academic Press,1972.
[106]Alan Wolf, Jack B Swift, Harry L Swinney, et al. Determining Lyapunov exponents from a time series[J].Physical D: Nonlinear Phenomena,1985,16(3):285-317.
[107]刘晓明．高压SF6断路器电弧动态数学模型及喷口结构优化设计[D]．沈阳：沈阳工业大学，2003．
[108]Frost L S, Liebermann R W. Composition and transport properties of SF6and their use in a simplified enthalpyflow arc model[J]. Proceedings of the IEEE,1971,59(4):474-485.
[109]曹云东．高压SF6断路器介质恢复特性的计算与分析[D]．沈阳：沈阳工业大学，2001．
[110]樊丁，陈剑虹，牛尾诚夫．TIG电弧传热传质过程的数值分析[J]．机械工程学报：1998，2(2)：39-45．
[111]Guimera R, Amaral L A N. Modeling the word-wide airport network[J]. The European Physics Journal B,2004,38:381-385.
[112]方小玲，姜宗来．基于脑电图的大脑功能性网络分析[J]．物理学报，2007，56(12)：7330-7339．
[113]郝崇清．基于时间序列的复杂网络构建与分析[D]．天津：天津大学，2012．
[114]Faloutsos M, Faloutsos P, Faloutsos C. On power-law relationships of the Internet topology[J]. ComputerCommunications Review,1999,29:251-262.
[115]汪小帆，李翔，陈关荣．复杂网络理论及其应用[M]．北京：清华大学出版社，2006．
[116]张欢欢．基于复杂网络的非线性时间序列的统计表征[D]．哈尔滨：哈尔滨工业大学，2011．
[117]胡海波，王林．幂律分布研究简史[J]．物理．2005，12：889-896．
[118]Zhang J, Small M. Complex network from pseudoperiodic time series: topology versus dynamics[J]. PhysicalReview Letters,2006,96(23):238701.
[119]Zhang J, Sun J F, Luo X D, et al. Characterizing pseudoperiodic time series through the complex networkapproach[J]. Physica D,2008,237:2856-2865.
[120]Gao Z K, Jin N D. Flow-pattern identification and nonlinear dynamics of gas-liquid two-phase flow incomplex networks[J]. Physical Review E,2009,79(6):066303.
[121]Gao Z K, Jin N D. Complex network from time series based on phase space reconstruction[J]. Chaos,2009,19(3):033137.
[122]Watts D J, Strogatz S H. Collective dynamics of small-world networks[J]. Nature,1998,393(1):440-442.
[123]Barabási, Albert László, Albert, et al. Emergence of scaling in random networks[J]. Science,1999,286(5439):509-512.