微电网的基本概念.doc

微电网的基本概念 1引言电力行业是关系国计民生的基础性行业。随着全球资源环境压力的不断增大、电力市场化进程的不断深入，可再生能源等分布式发电单元的数量不断增加，用户对电能质量要求的不断提升，传统网络已经难以满足社会发展需求，建设更加安全、可靠、环保、经济的现代电网成为全球电力行业的共同目标。现代电网的内涵包括实现以抵御事故扰动为主的安全稳定运行，降低大规模停运风险；使分布式电源得到有效的利用；提高用户用电的效率和电能质量；提高电网资产的利用率等等。其中，突出自愈功能的智能电网研究与发展被认为是当今世界电力系统发展变革的最新动向，是21世纪电力系统的重大科技创新和发展趋势1-6。近年来，分布式发电供能系统得到快速发展，由于分布式发电采用就地能源，可以实现分区分片灵活供电，通过合理的规划设计，在灾难性事件发生导致大电网瓦解的情况下，可以保证对配电网内重要负荷的供电，并有助于大电网快速恢复供电，降低大电网停电造成的社会经济损失。另外，分布式发电供能系统与大电网并网运行，有助于克服一些分布式电源的间歇性给用户负荷造成的影响，提高系统供电的电能质量7。因此，具有自愈能力、兼容多种发电资源、具有自愈能力的智能配电网研究将有助于分布式发电技术的推广应用，也有助于防止大面积停电，提高电力系统的安全性和可靠性，增强电网抵御自然灾害的能力。传统的电力系统由发电、输电、配电系统组成。据统计，8090以上的重要用户停电是由城市配电系统故障引起的。配电网结构复杂、数量不断增加，用户对电能质量要求的不断提升，传统网络已经难以满足社会发展需求，建设更加安全、可靠、环保、经济的现代电网成为全球电力行业的共同目标。现代电网的内涵包括实现以抵御事故扰动为主的安全稳定运行，降低大规模停运风险；使分布式电源得到有效的利用；提高用户用电的效率和电能质量；提高电网资产的利用率等等。其中，突出自愈功能的智能电网研究与发展被认为是当今世界电力系统发展变革的最新动向，是21世纪电力系统的重大科技创新和发展趋势1-6。传统的电力系统由发电、输电、配电系统组成。据统计，8090以上的重要用户停电是由城市配电系统故障引起的。配电网结构复杂、电压等级多样、且配电系统投资巨大。因此，在建设坚强国家电网骨干网架的同时，把握当前城市配电网大规模建设的良好时机，及时地解决好分布式发电供能系统接入配电网的关键技术问题，为实现健壮的智能配电网奠定坚实的理论基础，无疑具有重大的社会效益和经济价值。本文概述了智能电网的技术内涵，重点指出智能配电网所面临的技术难点与发展需求，对智能配电网保护控制系统提出了设计思路，并探讨了该领域的相关研究课题。2智能电网的技术内涵智能电网涵盖了电力系统所有的领域，是一项长期、庞大的科研课题与工程实践。它以智能一次设备和二次设备为基础，并集合了通信、计算机、电力电子技术的发展。从技术层面讲，未来的智能电网和目前电网的主要区别体现在电力市场化、新能源发电与储能技术、电能质量、以及电网自愈能力等各个方面。以电网结构及运行特征来分，智能电网的研究与实施主要包括以下内容，如图1所示。由于电力系统的运行、控制以及管理是依靠跨越各个分布系统或终端的信息交换来实现的。因此，智能电网的实施必然是建立在电力系统与通信系统高度集成和发展的基础上形成的智能电网体系。由图1可知，智能电网主要包含智能输电网和智能配电网两大组成部分。而智能变电站则是实现电网智能化运行与控制的关键组成部分。在此基础上，智能化电力调度的目标则包括建立一个基于同步信息的广域保护和紧急控制一体化理论与技术，协调电力系统元件保护和控制、区域稳定控制系统、紧急控制系统、解列控制系统和恢复控制系统等具有多道安全防线的综合防御体系。与此同时，智能电力交易及价格形成机制也是电力市场化的关键技术。鉴于电力系统具有广域动态的特征，智能电网的保护控制必然要进一步在数据信息交换的基础上，解决全局与局部的功能协调和速度协调，实现广域控制与分布保护控制的协调性。图1智能电网的技术内涵Fig.1 Technical content of smart grid3智能配电网的特点随着城市工业与社会生活的不断发展与需求，构建安全、稳定的城市电网是未来电网发展中至关重要的基础环节之一。而城市电网所涉及的内容正是智能配电网所面临的课题与发展方功能的自然扩展，从功能上必须能够支持现有结构所不能支持的两个基本要求2：（1）支持综合考虑终端用户和总体配电系统控制，以达到系统性能的优化、期望的安全稳定性与向。未来智能配电网结构应该是现有系统结构和期望的电能质量。（2）支持高比重的分布式能源接入电网，以提高系统的整体性、效率和灵活性。例如：通过协同的、分布式的控制，可以利用分布式能源来优化系统性能，并且在发生重大系统故障时利用分布式电源进行局部控制（微型电网、cell电网）。城市工业体系中数字化产业比重逐渐增大，对电能质量要求苛刻。为保证供电的连续性与可靠性，国外较为成功的配电网结构中实现了多分段、多连接的供电模式。甚至为了个别重要的用电负荷，局部配电网从辐射状网络向闭环环网的运行模式转变。而在这个发展变化的过程中，传统的保护配置方式及原理显然不能适应这种变化，进而形成了一些应用上的障碍。随着分布式发电资源以及微电网技术的发展，城市配电网受端系统出现发电单元并且其发电能力得到不断提升。因此智能配电网内的电力供应模式将发生改变，即从单一的由大型注入点单向供电的模式，向大量使用受端分布式发电设备的多源多向模块化模式转变。在微电网概念引入之前，世界各国一般均不允许分布式电源孤岛运行，采用系统故障时主动将分布式电源退出的保护控制方案。但随着微网技术的发展，在未来智能配电网中，微电网与配电网的协调运行以及其孤岛运行能力无疑是提高供电可靠性的有效措施之一8, 9。因此，智能配电网需具备分层分块结构，使每个微型电网模块具备孤岛运行能力，从而提高城市电网的健壮性、运行的灵活性。如图2所示的微电网典型结构。图2典型的微电网拓扑结构（来源：IEEE P1547.4）Fig.2 Typical micro-grid topologies根据微电网的结构、特征及其应用和负荷特征，微电网包括与配电网直接相连的微电网，以及面向工商业或居民供电的小型微电网等。不同的微电网特性不同、归属权不同，因此其运行方式灵活多变，相应的保护控制方式应根据这些一般特征有所侧重。4智能配电网的保护控制系统4.1 智能配电网的保护控制系统设计基于智能配电网结构的特点以及其所要实现的运行目标，智能配电网的保护控制应具有突出的自愈能力10, 11。所谓自愈，是指自我预防和自我恢复的能力，体现在以下两方面：（1）预防控制为主要的控制手段，及时发现、诊断和消除故障隐患；（2）具有故障情况下维持系统连续运行的能力，不造成系统的运行损失，并且通过自治修复功能从故障中尽可能恢复供电。自愈是智能配电网最突出的特点，一般智能配电网也往往被称为自愈电网。显然，智能配电网的自愈能力必须依靠可靠、协调的保护控制方案来支撑。根据前文中对智能配电网特点及其对保护控制系统的要求，本文提出的保护控制系统设计方案如图3所示。就电网结构看，图中包括与配电网直接相连的微电网系统（微电网-1），以及面向工商业或居民供电的小型微电网系统（微电网-2）。对于含微网的智能配电网来说，各分布式电源均有各自的控制器，尤其是逆变型电源的电力电子接口可以使分布式电源的运行更加智能化。它可以利用本地信息对其输出电压和频率进行控制，这对提高微电网自身的供电质量起到了重要的支撑作用（如微电网-2所示）。另一方面，对于微电网来说，同样需要保护控制系统以实现对各分布式电源有功和无功出力的监测，并要求实现对分布式电源及负荷的投切控制，从而达到最优的微电网与配电网的并网运行模式或孤岛运行模式（如微电网-1所示）。其中还包括孤岛运行方式下微电网与配电网的同步运行控制以及并网技术等等。图3所示的保护控制系统设计主要包括以下几部分：（1）面向电子式互感器、光互感器以及数字量输入的合并单元，可实现面向变电站或本地的多信息采集。（2）冗余的通信网络体系结构。这两部分是数据采集和分散控制的设备基础。同时，GPS信息的引入，不仅为基于本地信息的传统保护控制方案提供参考时标，更为面向区域或广域信息的控制策略提供了必要的技术基础12。（3）面向智能配电网的保护控制系统，该部分不仅包括配电网的保护控制方案，更应考虑微电网引入后的保护控制策略（图3的设计中侧重了微网引入后配网的保护控制）。其中，分别包括微电网1、2的保护控制方案及其相互关系，最终通过通信网络实现对分布式电源及负荷的分散控制。设计方案中通过通信网关与其他非本地的保护控制单元进行通信将实现更高层次的优化控制。正如前文所述，随着通信技术的成熟与设备的逐渐完善，在智能配电网内实现区域集中控制策略与本地保护控制的相互协调将得到深入的发展。I. INTRODUCTION 引论Economic, technology and environmental incentives are changing the face of electricity generation and随着技术的进步，经济的发展，社会的前进，环境保护的升级，传统的电力系统的发电，输电的方式在发生了变化。集中式的大规模发电设备逐步被小规模的分布式发电系统所渗透。transmission. Centralized generating facilities are giving way to smaller, more distributed generatio partially due to the loss of traditional economies of scale. Penetration of distributed generation across the US has not yet reached significant levels. However that situation is changing rapidly and requires attention to issues related to high penetration of distributed generation within the distribution system. Distributed generation encompasses a wide range of prime、mover technologies, such as internal combustion (IC) engines, gas turbines, micro-turbines, photovoltaic, fuel cells and wind power.。Most emerging technologies such as micro-turbines, photovoltaic, fuel cells and gas internal combustion engines with permanent interface with the electrical distribution system. These emerging technologies have lower emissions and the potential to have lower cost negating traditional economies of scale. The applications include power support at substations, deferral of T&D upgrades and onsite generation. Indiscriminant application of individual distributed generators can cause as many problems as it may solve. A better way to realize the emerging potential of distributed generation is to take a system approach which views generation and associated loads as a subsystem or a “microgrid”. This approach allowsfor local control of distributed generation thereby reducing or eliminating the need for central dispatch. During disturbances, the generation and corresponding loads can separate from the distribution system to isolate the microgrids load from the disturbance (and thereby maintaining high level of service)without harming the transmission grids integrity. Intentionalislanding of generation and loads have the potential to providea higher local reliability than that provided by the powersystem as a whole. The smaller size of emerging generationtechnologies permits generators to be placed optimally inrelation to heat loads allowing for use of waste heat. Suchapplications can more than double the overall efficiencies ofthe systems.Most current microgrid implementations combine loads withsources, allow for intentional islanding and try to use theavailable waste heat. These solutions rely on complexcommunication and control and are dependent on keycomponents and require extensive site engineering. Ourapproach is to provide generator-based controls that enable aplug-and-play model without communication or customengineering for each site.II. EMERGING GENERATION TECHNOLOGIESDistributed-power applications favor natural-gas technologiesdue to the potential of low air emissions. Diesel-fueledsystems still dominate in standby and short-run applications,but currently natural gas is better at combining availability,price and environmental compliance.Reciprocating engine technology has been driven by economicand environmental pressures for power-density improvements,increased fuel efficiency and reduced emissions. Emissions ofnatural-gas engines have improved significantly through betterdesign and control of the combustion process. Advanced leanburnnatural-gas engines produce nitrogen oxide (NOx) levelsas low as 50 ppmv, which is an enormous improvement butin most applications still requires the use of exhaust catalysts.Efficiencies are around 39% with a goal of 50%.Unfortunately high efficiency and low emissions are notcurrently achieved simultaneously.Microturbines are an important emerging technology. They aremechanically simple, single shaft devices with air bearingsiand no lubricants. They are designed to combine the reliabilityof commercial aircraft auxiliary power units with the low costof automotive turbochargers. The generator is usually apermanent magnet machine operating at variable speeds(50,000-100,000 rpm). This variable speed operation requirespower electronics to interface to the electrical system.Examples include; Capstones 30-kW and 60- kW systems andproducts from European manufacturers Bowman and Turbec.Sophisticated combustion systems, low turbine temperaturesand lean fuel-to-air ratios results in NOx emissions of lessthan 10 ppmv and inherently low-carbon monoxide emissions.Larger gas turbines, reciprocating engines, and reformers allinvolve higher temperatures that result in much higher NOxproduction. Microturbines can operate using a number ofdifferent fuels including natural-gas and such liquid fuels asgasoline with efficiencies in the 28-30% range.Fuel cells, which produce electricity from hydrogen andoxygen, emit only water vapor. NOx and CO2 emissions areassociated with the reforming of natural gas or other fuels toproduce the fuel cells hydrogen supply. Fuel cells offer higherefficiency than microturbines with low emissions but arecurrently expensive. Phosphoric acid cells are commerciallyavailable in the 200-kW range, and high temperature solidoxideand molten-carbonate cells have been demonstrated andare particularly promising for distributed applications. Amajor development effort by automotive companies hasfocused on the possibility of using on-board reforming ofgasoline or other common fuels to hydrogen, to be used inlow temperature proton exchange membrane (PEM) fuel cells.Automotive fuel cells will have a major impact on stationarypower if the automotive cost goal of $100/kW is achieved.III. TOTAL SYSTEM EFFICIENCYMost existing power plants, central or distributed, deliverelectricity to user sites at an overall fuel-to-electricityefficiency in the range of 28-32%. This represents a loss ofaround 70% of the primary energy provided to the generator.To reduce this energy loss it is necessary to either increase thefuel-to-electricity efficiency of the generation plant and/or usethe waste heat.Combined power cycles technology can attain efficienciesapproaching 60% with ratings in the hundreds of millionwatts 1. On the other hand if the waste heat from generatorswith much lower efficiency (28-32%) can be utilized throughheat exchangers, absorption chillers and/or desiccantdehumidification the overall fuel-to-useful energy efficiencycan be higher than 80%. Currently Capstone markets a 60kWmicroturbine using waste heat to heat water. This system has afuel to useful energy efficiency approaching 90% 2.This use of waste heat through co-generation or combined heatand power (CHP) implies an integrated energy system, whichdelivers both electricity and useful heat from an energy sourcesuch as natural gas 3. Unlike electricity, heat, usually in theform of steam or hot water, cannot be easily or economicallytransported long distances, so CHP systems typically provideheat for local use. Because electricity is more readilytransported than heat, generation of heat close to the locationof the heat load will usually make more sense than generationof heat close to the electrical load.The size of emerging generation technologies permitsgenerators to be placed optimally in relation to heat loads.The scale of heat production for individual units is small andtherefore offers greater flexibility in matching to heatrequirements. An ideal system could be constructed from themost economic combination of waste-heat-producinggenerators and non-waste-heat producing generators so that thecombined generation of electricity and heat is optimized. In anextreme example, fuel cells could be placed on every floor of ahospital to meet each floors hot water needs and provideelectricity to local loads.For stationary energy users needing both electricity and usefulheat there are two basic systems available: separate generationof electricity and heat, and combined heat and power (CHP)systems located near the heat load. To find the total energyefficiency for systems with separate generation of electricityand heat the generation efficiencies as well as the loading ratioof thermal energy-to-electrical energy needs to be known. ForCHP systems the electrical and thermalrecovery efficiencies along with the maximum loading levelare required.Figure 1 illustrates the total energy efficiency as a function ofthe loading ratio for different systems. Two systems assumeseparate generation of electricity and heat, the third is a CHPsystem. The assumed thermal generation efficiency for theFigure. 1. Total Energy Efficiencynon-CHP examples is 85%, while the electrical efficiencies are60% and 30% respectively. If a loading ratio of one isassumed the overall efficiencies of the separate systems are70% and 44%.For the case where the waste heat is near the heat load thewaste heat can be used instead of fuel to provide the requiredheat. Typical thermal recovery efficiencies can range from 20%to 80%. Unlike separate systems for heat and electricity themaximum loading ratio is limited. For example if theelectrical efficiency is 30% resulting in 70% of the fuel beingwaste heat. If this waste heat can be converted to useful heatassuming a thermal recovery efficiency of 40% the totalenergy efficiency is 58% with a loading (ratio of thermalenergy-to-electrical energy) of 1.00. This is the maximumloading and maximum total efficiency for this particularsystem. If the system is not loaded to this level the totalefficiency drops linearly. It is apparent from figure 1 thatCHP systems can greatly improve the total energy efficiencydepending on loading levels and the thermal recoveryefficiencies. Note that three thermal recovery efficiencies areshown on the plot for the CHP system.IV. MICROGRID CONCEPTTo realize the emerging potential of distributed generationone must take a system approach which views generation andassociated loads as a subsystem or a “microgrid” 4. Duringdisturbances, the generation and corresponding loads canseparate from the distribution system to isolate themicrogrids load from the disturbance (and therebymaintaining service) without harming the transmission gridsintegrity.The difficult task is to achieve this functionality withoutextensive custom engineering and still have high systemreliability and generation placement flexibility. To achievethis we promote a peer-to-peer and plug-and-play model foreach component of the microgrid. The peer-to-peer conceptinsures that there are no components, such as a mastercontroller or central storage unit that is critical for operation ofthe microgrid. This implies that the microgrid can continueoperating with loss of any component or generator. With oneadditional source (N+1) we can insure complete functionalitywith the loss of any source. Plug-and-play implies that a unitcan be placed at any point on the electrical system without reengineeringthe controls.Plug-and-play functionality is much akin to the flexibility onehas when using a home appliance. That is it can be attached tothe electrical system at the location where it is needed. Thetraditional model is to cluster generation at a single point thatmakes the electrical application simpler. The plug-and-playmodel facilitates placing generators near the heat loads therebyallowing more effective use of waste heat without complexheat distribution systems such as steam and chilled water This ability to island generation and loads together has thepotential to provide a higher local reliability than thatprovided by the power system as a whole. Smaller units,having power ratings in thousands of watts, can provide evenhigher reliability and fuel efficiency. These units can createmicrogrid services at customer sites such as office buildings,industrial parks and homes. Since the smaller units aremodular, site management could decide to have more units(N+) than required by the electrical/heat load, providing local,online backup if one or more of the operating units failed. It isalso much easier to place small generators near the heat loadsthereby allowing more effective use of waste heat.Basic Microgrid architecture is shown in figure 2. Thisconsists of a group of radial feeders, which could be part of adistribution system or a buildings electrical system. There isa single point of connection to the utility called point ofcommon coupling. Some feeders, (Feeders A-C) havesensitive loads, which require local generation. The noncriticalload feeders do not have any local generation. In ourexample this is Feeder D. Feeders A-C can island from thegrid using the static switch which can separate in less than acycle 5. In this example there are four microsources at nodes8, 11, 16 and 22, which control the operation using only localvoltages and currents measurments. When there is a problemwith the utility supply the static switch will open, isolatingthe sensitive loads from the power grid. Feeder D loads ridethrough the event. It is assumed that there is sufficientgeneration to meet the loads demand. When the Microgrid isgrid-connected power from the local generation can be directedto feeder D.Figure 2 MicrogridPoint ofCommonCouplingStaticSwitchFeeder AFeeder BFeeder CFeeder D22168 11AI. BBV. MICROGRID CONTROLInverters can provide the control and flexibility required forplug-and-play functionally. Microgrid controls need to insurethat; new microsources can be added to the system withoutmodification of existing equipment, the Microgrid can connectto or isolate itself from the grid in a rapid and seamlessfashion, reactive and active power can be independentlycontrolled, and can meet the dynamic needs of the loadsMicrosource controller techniques described below rely on theinverter interfaces found in fuel cells, microturbines, andstorage technologies. A key element of the control design isthat communication among microsources is un