Wind power forecasting & prediction methods
基于卡尔曼滤波原理对风电功率短期预测
1基于卡尔曼滤波原理的风电功率预报林可薇(西安交通大学电气工程学院，陕西省，西安市，710000)Prediction of wind power based on the principle of Calman filterLIN Ke-wei(School of Electrical Engineering, Xi'an Jiaotong University, Xi'an710000, Shanxi Province, China)ABSTRACT:focuses on the principle of Calman filter and Calman filter algorithm,Understanding both based applications in the wind power forecast.Wind power forecasting system based on digital weather forecast meteorological parameters related to output close to the ground can not accurately predict the output power,Departure from the principle of Kalman filtering,Kalman filter algorithm can take advantage of wind capacity to be corrected digital output weather forecast,Improve forecast accuracy.KEY WORD:Kalman filter;Power prediction;Accuracy 摘要：本文重点介绍卡尔曼滤波原理和卡尔曼滤波算法，了解基于两者在风电预测的应用。
风能英文简介
风能英⽂简介Wind powerWind power is the conversion of wind energy into a useful form of energy, such as using wind turbines to make electricity, wind mills for mechanical power, wind pumps for pumping water or drainage, or sails to propel ships.At the end of 2009, worldwide nameplate capacity of wind-powered generators was 159.2 gigawatts (GW).(By June 2010 the capacity had risen to 175 GW.) Energy production was 340 TWh, which is about 2% of worldwide electricity usage; and has doubled in the past three years. Several countries have achieved relatively high levels of wind power penetration, such as 20% of stationary electricity production in Denmark, 14% in Ireland and Portugal, 11% in Spain, and 8% in Germany in 2009. As of May 2009, 80 countries around the world are using wind power on a commercial basis.Wind power: worldwide installed capacity 1996-2008Large-scale wind farms are connected to the electric power transmission network; smaller facilities are used to provide electricity to isolated locations. Utility companies increasingly buy back surplus electricity produced by small domestic turbines. Wind energy, as an alternative to fossil fuels, is plentiful, renewable, widely distributed, clean, and produces no greenhouse gas emissions during operation. However, the construction of wind farms is not universally welcomed because of their visual impact but any effects on the environment are generally among the least problematic of any power source. The intermittency of wind seldom creates problems when using wind power to supply a low proportion of total demand, but as the proportion rises, increased costs, a need to upgrade the grid, and a lowered ability to supplant conventional production may occur. Power management techniques such as exporting and importing power to neighboring areas or reducing demand when wind production is low, can mitigate these problems.Burbo Bank Offshore Wind Farm, at the entrance to the River Mersey in North West England.HistoryHumans have been using wind power for at least 5,500 years to propel sailboats and sailing ships. Windmills have been used for irrigation pumping and for milling grain since the 7th century AD in what is now Afghanistan, India, Iran and Pakistan.In the United States, the development of the "water-pumping windmill" was the major factor in allowing the farming and ranching of vast areas otherwise devoid of readily accessible water. Windpumps contributed to the expansion of rail transport systems throughout the world, by pumping water from water wells for the steam locomotives. The multi-bladed wind turbine atop a lattice tower made of wood or steel was, for many years, a fixture of the landscape throughout rural America. When fitted with generators and battery banks, small wind machines provided electricity to isolated farms.Medieval depiction of a wind millIn July 1887, a Scottish academic, Professor James Blyth, undertook wind power experiments that culminated in a UK patent in 1891. In the United States, Charles F. Brush produced electricity using a wind powered machine, starting in the winter of 1887-1888, which powered his home and laboratory until about 1900. In the 1890s, the Danish scientist and inventor Poul la Cour constructed wind turbines to generate electricity, which was then used to produce hydrogen. These were the first of what was to become the modern form of wind turbine.Small wind turbines for lighting of isolated rural buildings were widespread in the first part of the 20th century. Larger units intended for connection to a distribution network were tried at several locations including Balaklava USSR in 1931 and in a 1.25 megawatt (MW) experimental unit in Vermont in 1941.The modern wind power industry began in 1979 with the serial production of wind turbines by Danish manufacturers Kuriant, Vestas, Nordtank, and Bonus. These early turbines were small by today's standards, with capacities of20–30 kW each. Since then, they have increased greatly in size, with the Enercon E-126 capable of delivering up to 7 MW, while wind turbine production has expanded to many countries.Windmills are typically installed in favourable windy locations. In the image, wind power generators in Spain near an Osborne bullWind energyThe Earth is unevenly heated by the sun, such that the poles receive less energy from the sun than the equator; along with this, dry land heats up (and cools down) more quickly than the seas do. The differential heating drives a global atmospheric convection system reaching from the Earth's surface to the stratosphere which acts as a virtual ceiling. Most of the energy stored in these wind movements can be found at high altitudes where continuous wind speeds of over 160 km/h (99 mph) occur. Eventually, the wind energy is converted through friction into diffuse heat throughout the Earth's surface and the atmosphere.The total amount of economically extractable power available from the wind is considerably more than present human power use from all sources. The most comprehensive study as of 2005 found the potential of wind power on land and near-shore to be 72 TW, equivalent to 54,000 MToE (million tons of oil equivalent) per year, or over five times the world's current energy use in all forms. The potential takes into account only locations with mean annual wind speeds ≥ 6.9 m/s at 80 m. The study assumes six 1.5 megawatt, 77 m diameter turbines per square kilometer on roughly 13% of the total global land area (though that land would also be available for other compatible uses such as farming). The authors acknowledge that many practical barriers would need to be overcome to reach this theoretical capacity.Map of available wind power for the United States. Color codes indicate wind power density classThe practical limit to exploitation of wind power will be set by economic and environmental factors, since the resource available is far larger than any practical means to develop it.Distribution of wind speedThe strength of wind varies, and an average value for a given location does not alone indicate the amount of energy a wind turbine could produce there. To assess the frequency of wind speeds at a particular location, a probability distribution function is often fit to the observed data. Different locations will have different wind speed distributions. The Weibull model closely mirrors the actual distribution of hourly wind speeds at many locations. The Weibull factor is often close to 2 and therefore a Rayleigh distribution can be used as a less accurate, but simpler model.Distribution of wind speed (red) and energy (blue) for all of 2002 at the Lee Ranch facility in Colorado. The histogram shows measured data, while the curve is the Rayleigh model distribution for the same average wind speedBecause so much power is generated by higher wind speed, much of the energy comes in short bursts. The 2002 Lee Ranch sample is telling; half of the energy available arrived in just 15% of the operating time. The consequence is that wind energy from a particular turbine or wind farm does not have as consistent an output as fuel-fired power plants.Electricity generationIn a wind farm, individual turbines are interconnected with a medium voltage (often 34.5 kV), power collection system and communications network. At a substation, this medium-voltage electric current is increased in voltage with a transformer for connection to the high voltage electric power transmission system.Typical components of a wind turbine (gearbox, rotor shaft and brake assembly) being lifted into positionThe surplus power produced by domestic microgenerators can, in some jurisdictions, be fed into the network and sold to the utility company, producing a retail credit for the microgenerators' owners to offset their energy costs.Grid managementInduction generators, often used for wind power, require reactive power for excitation so substations used in wind-power collection systems include substantial capacitor banks for power factor correction. Different types of wind turbine generators behave differently during transmission grid disturbances, so extensive modelling of the dynamic electromechanical characteristics of a new wind farm is required by transmission system operators to ensure predictable stable behaviour during system faults (see: Low voltage ride through). In particular, induction generators cannot support the system voltage during faults, unlike steam or hydro turbine-driven synchronous generators. Doubly-fed machines generally have more desirable properties for grid interconnection. Transmission systems operators will supply a wind farm developer with a gridcode to specify the requirements for interconnection to the transmission grid. This will include power factor, constancy of frequency and dynamic behavior of the wind farm turbines during a system fault.Capacity factorSince wind speed is not constant, a wind farm's annual energy production is never as much as the sum of the generator nameplate ratings multiplied by the total hours in a year. The ratio of actual productivity in a year to this theoretical maximum is called the capacity factor. Typical capacity factors are 20–40%, with values at the upper end of the range in particularly favourable sites. For example, a 1 MW turbine with a capacity factor of 35% will not produce8,760 MW·h in a year (1 × 24 × 365), but only 1 × 0.35 × 24 ×365 = 3,066 MW·h, averaging to 0.35 MW. Online data is available for some locations and the capacity factor can be calculated from the yearly output.Unlike fueled generating plants, the capacity factor is limited by the inherent properties of wind. Capacity factors of other types of power plant are based mostly on fuel cost, with a small amount of downtime for maintenance. Nuclear plants have low incremental fuel cost, and so are run at full output and achievea 90% capacity factor. Plants with higher fuel cost are throttled back to follow load. Gas turbine plants using natural gas as fuel may be very expensive to operate and may be run only to meet peak power demand. A gas turbine plant may have an annual capacity factor of 5–25% due to relatively high energy production cost.In a 2008 study released by the U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy, the capacity factor achieved by the wind turbine fleet is shown to be increasing as the technology improves. The capacity factor achieved by new wind turbines in 2004 and 2005 reached 36%.PenetrationWind energy "penetration" refers to the fraction of energy produced by wind compared with the total available generation capacity. There is no generally accepted "maximum" level of wind penetration. The limit for a particular grid will depend on the existing generating plants, pricing mechanisms, capacity for storage or demand management, and other factors. An interconnected electricity grid will already include reserve generating and transmission capacity to allow for equipment failures; this reserve capacity can also serve to regulate for the varying power generation by wind plants. Studies have indicated that 20% of the total electrical energy consumption may be incorporated with minimal difficulty. These studies have been for locations with geographically dispersed wind farms, some degree of dispatchable energy, or hydropower with storage capacity, demand management, and interconnection to a large grid area export of electricity when needed. Beyond this level, there are few technical limits, but the economic implications become more significant. Electrical utilities continue to study the effects of large (20% or more) scale penetration of wind generation on system stability and economics.At present, a few grid systems have penetration of wind energy above 5%: Denmark (values over 19%), Spain and Portugal (values over 11%), Germany and the Republic of Ireland (values over 6%). But even with a modest level of penetration, there can be times where wind power provides a substantial percentage of the power on a grid. For example, in the morning hours of 8 November 2009, wind energy produced covered more than half the electricity demand in Spain, setting a new record. This was an instance where demand was very low but wind power generation was very high.Variability and intermittencyElectricity generated from wind power can be highly variable at several different timescales: from hour to hour, daily, and seasonally. Annual variation also exists, but is not as significant. Related to variability is the short-term (hourly or daily) predictability of wind plant output. Like other electricity sources, wind energy must be "scheduled". Wind power forecasting methods are used, but predictability of wind plant output remains low for short-term operation.Because instantaneous electrical generation and consumption must remain in balance to maintain grid stability, this variability can present substantial challenges to incorporating large amounts of wind power into a grid system. Intermittency and the non-dispatchable nature of wind energy production can raise costs for regulation, incremental operating reserve, and (at highpenetration levels) could require an increase in the already existing energy demand management, load shedding, or storage solutions or system interconnection with HVDC cables. At low levels of wind penetration, fluctuations in load and allowance for failure of large generating units requires reserve capacity that can also regulate for variability of wind generation. Wind power can be replaced by other power stations during low wind periods. Transmission networks must already cope with outages of generation plant and daily changes in electrical demand. Systems with large wind capacity components may need more spinning reserve (plants operating at less than full load).Wildorado Wind Ranch in Oldham County in the Texas Panhandle, as photographed from U.S. Route 385Pumped-storage hydroelectricity or other forms of grid energy storage can store energy developed by high-wind periods and release it when needed. Stored energy increases the economic value of wind energy since it can be shifted to displace higher cost generation during peak demand periods. The potential revenue from this arbitrage can offset the cost and losses of storage; the cost of storage may add 25% to the cost of any wind energy stored, but it is not envisaged that this would apply to a large proportion of wind energy generated. The 2 GW Dinorwig pumped storage plant in Wales evens out electrical demand peaks, and allows base-load suppliers to run their plant more efficiently. Although pumped storage power systems are only about 75% efficient, and have high installation costs, their low running costs and ability to reduce the required electrical base-load can save both fuel and total electrical generation costs.In particular geographic regions, peak wind speeds may not coincide with peak demand for electrical power. In the US states of California and Texas, for example, hot days in summer may have low wind speed and high electrical demand due to air conditioning. Some utilities subsidize the purchase of geothermal heat pumps by their customers, to reduce electricity demand during the summer months by making air conditioning up to 70% more efficient; widespread adoption of this technology would better match electricity demand to wind availability in areas with hot summers and low summer winds. Another option is to interconnect widely dispersed geographic areas with an HVDC"Super grid". In the USA it is estimated that to upgrade the transmission system to take in planned or potential renewables would cost at least $60 billion.In the UK, demand for electricity is higher in winter than in summer, and so are wind speeds. Solar power tends to be complementary to wind. On daily to weekly timescales, high pressure areas tend to bring clear skies and low surface winds, whereas low pressure areas tend to be windier and cloudier. On seasonal timescales, solar energy typically peaks in summer, whereas in many areas wind energy is lower in summer and higher in winter. Thus the intermittencies of wind and solar power tend to cancel each other somewhat. The Institute for Solar Energy Supply Technology of the University of Kassel pilot-tested a combined power plant linking solar, wind, biogas and hydrostorage to provide load-following power around the clock, entirely from renewable sources.A report on Denmark's wind power noted that their wind power network provided less than 1% of average demand 54 days during the year 2002. Wind power advocates argue that these periods of low wind can be dealt with by simply restarting existing power stations that have been held in readiness or interlinking with HVDC. Electrical grids with slow-responding thermal power plants and without ties to networks with hydroelectric generation may have to limit the use of wind power.[42]Three reports on the wind variability in the UK issued in 2009, generally agree that variability of wind needs to be taken into account, but it does not make the grid unmanageable; and the additional costs, which are modest, can be quantified. A 2006 International Energy Agency forum presented costs for managing intermittency as a function of wind-energy's share of total capacity for several countries, as shown: Increase in system operation costs, Euros per MW·h, for 10% and 20% wind share10% 20%Germany 2.5 3.2Denmark 0.4 0.8Finland 0.3 1.5Norway 0.1 0.3Sweden 0.3 0.7。
英国大学电力专业介绍
英国大学电力专业排名、地址及教授介绍1 Cambridge 剑桥大学地址：英格兰剑桥镇网址：教授介绍：主要都是电力电子方向Prof Gehan Amaratunga Director of EPECResearch interests include: Nanoscale materials and device design for electronics and energy conversion. Novel materials and device structures for low cost, high efficiently solar cells. Power electronics for optimum grid connection of large photovoltaic electric generation systems. Integrated and discrete semiconductor devices for power switching and control.Dr Richard McMahon Senior LecturerCurrent research focuses on low maintenance generators for wind turbines; linear generators for wave power and energy efficient power conversion for power supplies and electric appliances such as compact fluorescent lights.Dr Timothy Coombs Heads Superconductivity GroupResearch Interests * Electrical Machines * Electromagnetic Modelling * Engineering Applications of Superconductivity * MEMS2 Southampton 南安普顿大学地址：英格兰南部海滨的Hampshire郡，主校区(Highfield Campus)距离南安普敦市中心3英里网址：教授介绍：以下为主要负责人及链接，无法打开Professor Jan SykulskiElectrical Power Engineering3 Imperial College 伦敦大学帝国理工学院地址：帝国理工主校区坐落于伦敦标准的富人区南肯辛顿网址：教授介绍：Dr Balarko ChaudhuriHis areas of interest are power systems dynamics, stability and robust control. He is actively involved with ABB Corporate Research in the area of wide-area monitoring and control of power systemsProf Goran StrbacProf Strbac's research interests include Power system optimisation and economics; Integration of distributed energy resources and Intermittency.Dr Bikash PalPower System Stability; Dynamic Equivalencing and Coherency; State estimation in power distribution system, Robust Control of Power System Oscillations; FACTS Controllers; Distributed and Renewable Energy Modelling; Grid Integration of Marine Wave Generations; and Risk modelling and assessment in distribution system operation.Dr Imad Jaimoukharobust controller design for structured and unstructured uncertainties; controller reduction; model reduction for large-scale systems; hierarchical optimization in robust controller design, robust control design for power system and fault detection and isolation.4 Surrey 萨里大学地址：位于英国伦敦市郊的吉尔福德网址：教授介绍：没有electrical engineering5 Loughborough 拉夫堡大学地址：位于英格兰中部的拉夫堡镇网址：/教授介绍：Professor Philip C EamesHis research focuses on various aspects of renewable energy systems, energy in buildings and thermal energy storage.Professor Ivor R SmithFor several years recently his research interests have been in the pused power area, where he has been concerned with the Generation, Processing and Application of High-Energy pulsed of electrical energy.Dr Murray ThomsonRenewables into existing electrical power systems. Analysis of low-voltage distribution networks and the development of flexible demand as a means of grid balancing in future low-carbon power systems incorporating high penetrations of intermittent wind, marine and solar powerDr Simon J Watson Head of Wind and Water Power ResearchCondition Monitoring of Wind Turbines; Wind Resource Assessment; Wind Power Forecasting; Wake Modelling; Wave Power Device Modelling; Climate Change Impacts.6 Edinburgh 爱丁堡大学地址：位于爱丁堡市中心，爱丁堡则位于苏格兰海滨，是苏格兰首府网址：/教授介绍：Prof Robin Wallacenetwork integration of distributed renewable energy generation and marine energyProf Janusz BialekPower system economics: Transmission pricing; Modelling electricity markets and security of supply; Congestion management.Sustainable power generation and supply: Future Network Technologies; Flexible Network; Asset Management and Performance in Energy Networks; Autonomous Regional Active Network Management System; Smart Grid Oscillation Management for a Changing Generation Mix. Power systems dynamic and stabilityProf Ian BrydenMarine Renewable EnergyDr Markus MuellerThe design of novel generator topologies for direct drive wave energy, wind energy and tidal current energy converters7 Sheffield 谢菲尔得大学地址：谢菲尔得大学位于约克郡南边的谢菲尔得市网址：教授介绍：Emeritus Professor Barry ChambersSmart electromagnetic structures. Target signature management. Passive and active radar absorbing materials. Conducting polymers and composites. Optimisation using evolutionary computing techniques. Automated microwave measurement systems. Radomes and electromagnetic windows.Emeritus Professor David HoweElectrical technologies for aerospace, automotive and marine applications. High integrity electrical drive systems. Novel electromagnetic devices. Multi-physics modelingProfessor Geraint JewellSelf-bearing electrical machines. Power dense electrical machines and actuators for aerospace and marine applications. Valve actuation. Electromagnetic modelling of novel devices8 Bristol 布里斯托大学地址：大学的几个校园分布在极具活力的现代海滨城市－布里斯托市中心，布里斯托是英格兰西南部最大的城市网址：/教授介绍：Dr Dritan KaleshiCommunications and distributed systems performance; connectivity and performance issues in access and local networks. Self-organised systems, service discovery. Specification of distributed systems; specification conformance testing. Small-device networking, and in particular home networking systems. Interoperability: standardisation and autonomic systems. ICT solutions for SmartGrids and distributed energy management.Dr Dave DruryHardware-in-the-Loop and real-time substructuring (hybrid dynamic) testing methods Aircraft generation and power management systems Hybrid automative vehicle traction and generation systems Efficient control of electric machines Distributed control methods, using industrial purpose built networks and standard ethernet9 York 约克大学地址：约克大学位于历史名城约克郡约克市网址：教授介绍：没有electrical engineering10 Essex 埃塞克斯大学地址：埃塞克斯大学位于英国有史以来最古老的市镇科尔切斯特(Colchester)的郊外两英里处，该镇也是英国的第一个首都网址：主校区：/Southend校区：/southendEast 15 校区：教授介绍：没有electrical engineering11 Bath 巴斯大学地址：巴斯市郊网址：/教授介绍：Dr Miles Alexander Redfernthe control and protection of distribution systemsthe connection of embedded generationintegration of renewable energy systems into utility networks.high speed transient based protection schemescommunications for power system control and protectionnon-invasive techniques for the location of buried utilities.Dr Furong Liall aspects of power system planning, operation, analysis and power system economics.Professor Raj AggarwalProfessor Aggarwal’s research interests are in Electrical Power and Energy Systems. His research group focuses on the technology to support the development of a secure and stable electricity supply network that is able to accommodate new and renewable forms of energy generation.12 Glasgow 格拉斯哥大学地址：格拉斯哥市以西3英里网址：教授介绍：Prof. Enrique Achapower systems analysis and power electronics applications in renewable energy systems.Prof. T.J.E. MillerActive in the Power Systems & Power Engineering research areaProf. John O'Reillyfundamental trade-offs between transient stability and oscillation stability in multi-machine power systems, distributed renewable (wind) generation systems, and distribution level energy control and management.13 Queen's, Belfast 贝尔法斯特女王大学地址：学校位于北爱尔兰首府贝尔法斯特的绿树成荫的南部郊外，步行到市中心只要15分钟网址：教授介绍：Professor Brendan FoxHis interests are in power system analysis, modelling and operation. Current interests include system integration aspects of embedded generation, including wind farms, and power system dynamic stability.Professor Haifeng Wangpower systems modelling, analysis and control with power electronics and renewable generation.14 Leeds 利兹大学地址：利兹市是英格兰北部的金融以及工业中心网址：教授介绍：没有electrical engineering15 University College London 伦敦大学学院地址：伦敦市中心的Bloomsbury广场网址：教授介绍：没有electrical engineering=15 Strathclyde 斯特拉斯克莱德大学地址：格拉斯哥是苏格兰最大的城市，地处苏格兰中部，位于克莱德河两岸网址：/教授介绍：Prof James R McDonaldPower system operation, management and control, protection system analysis, design and modelling, artificial intelligence applications in power systems, energy management systems, electricity pricing techniques, power system planning; optical sensing techniques.Prof Stephen McArthurPower engineering applications of Artificial Intelligence: condition monitoring; diagnostics and prognostics for equipment and plant; active network management and Smart Grids; and monitoring and diagnosis of nuclear reactorsIntelligent and automated power system fault analysisIntelligent system methods: knowledge based systems; model based reasoning; case based reasoningMulti-Agent Systems and Intelligent Agents: agent based condition monitoring; agent based power system fault analysis; multi-agent methods, models, techniques and architectures for power engineering applicationsModel and simulation integrationDecision support environmentsProf Kwok L LoPower systems analysis, planning, operation, monitoring and control including the application of expert systems and artificial neural networks;transmission and distribution management systems and privatisation issues.Prof William E LeitheadDynamic analysis, simulation, modelling, control, integrated drive-train design of wind turbines. Analysis and design of multivariable control system. Analysis and simulation of stochastic systems.Prof David InfieldMy research interests are with electricity generation from renewable energy sources, in particular from wind and photovoltaics (PV), and the integration of these sources into electricity systems large and small. Associated with this central challenge I take an interest in energy storage technology and application, and demand side management.Dr Andrew J CrudenHydrogen and Fuel cell systems, electric vehicles, power electronics for fuel cells and rotating machines (e.g. wind turbines), and electrical machine design.Dr Graham AultDr. Ault's research is in the general area of power system planning and operations with particular emphasis on renewables grid integration, distributed energy resources, distribution systems and long-term transitions and scenarios.17 Manchester 曼彻斯特大学地址：曼彻斯特大学位于地理位置优越的曼彻斯特市中心，曼彻斯特市是伦敦以外英国最重要的商业、教育和文化中心，也是英格兰重要的交通枢纽网址：教授介绍：Prof Daniel Kirschen Head of the Electrical Energy and Power Systems Group in the School of Electrical and Electronic Engineering.The introduction of competitive electricity markets has created a whole new set of interesting and challenging problems in the operation and development of power systems.Prof Jovica MilanovicHis research and consultancy work is equally split between the areas of Power System Dynamics and Power Quality with a common, underlying stream of probabilistic / stochastic modelling of uncertainties of events and system parameters.Prof Vladimir TerzijaMy main research interests are application of intelligent methods to power system monitoring, control, and protection, as well as power system plant, particularly switchgears.=17 Heriot-Watt 赫瑞特瓦特大学地址：主校园位于爱丁堡的郊外。
关于风电不确定性对电力系统影响的评述_薛禹胜
randomness; wind power forecasting; wind farm control 摘要： 风电的波动和间歇行为都具有强烈的不确定性， 其对 电力可靠性、 电能质量、 经济性及社会福利的影响随着渗透 率的增加而越发突出。为此，讨论风电波动性、间歇性与随 机性的关系；归纳风电不确定性因素的构成、描述，及其对 电力系统功角稳定性、频率与电压可接受性、充裕性、电能 质量、 电力市场及减排等方面的影响； 并将其纳入广义阻塞 的框架。回顾对其机理的研究现状；讨论发电侧、电网、需 求侧及其综合的不确定性分析及协调控制； 提出计及风电不 确定性的电网三道防线； 强调量化和风险观点在上述研究中 的重要性。 关键词：风电；波动性；间歇性；随机性；风电功率预测； 风电场控制
A Review on Impacts of Wind Power Uncertainties on Power Systems
XUE Yusheng1, LEI Xing2, XUE Feng1, YU Chen1, DONG Zhaoyang3, WEN Fushuan4, JU Ping5
风电功率预测文献综述
风电功率预测方法的研究摘要由于风能具有间歇性和波动性性等特点，随着风力发电的不断开展风电并网对电力系统的调度和平安稳定运行带来了巨大的挑战。
进展风电功率预测并且不断提高预测准确度变得越来越重要。
通过对国内外研究现状的了解，根据已有的风电功率预测方法，按照预测时间、预测模型、预测方法等对现有的风电功率预测技术进展分类，着重分析几种短期风电功率预测方法的优缺点及其使用场合。
根据实际某一风电场的数据，选取适宜的风电预测模型进展预测，对结果予以分析和总结。
关键词：风电功率预测；电力系统；风力发电；预测方法；引言随着社会不断开展人们对能源需求越来越大而传统化石能源日益枯竭不可再生，以及化石能源带来了环境污染等问题影响人类生活，人们迫切需要新的清洁能源代替传统化石能源。
风能是清洁的可再生能源之一，大力开展风力发电成为各国的选择。
根据相关统计，截止至2021 年，全球风电产业新增装机63013MW，，同比增长22%[1]。
其中，中国风电新增装机容量达30500MW，占市场份额48.4%。
全球累计装机容量为432419MW，其中中国累计装机容量为145104,占全球市场份额的33.6%。
目前风力发电主要利用的是近地风能，近地风能具有波动性、间歇性、低能量密度等特点，因而风电功率也是波动的。
当接入到电网的风电功率到达一定占比时，风电功率的大幅度波动将破坏电力系统平衡和影响电能质量，给电力系统的调度和平安平稳运行带来严峻挑战。
根据风速波动对风力发电的影响按照时间长度可分为三类：一种是在几分钟之内的超短时波动，该时段内的波动影响风电机组的控制；另一种是几小时到几天内的短时波动，该时段内的波动影响风电并网和电网调度；最后一种是数周至数月的中长期波动，该时段内的波动影响风电场与电网的检修和维护方案。
本文主要研究不同的风电功率短期预测方法的优缺点。
通过对短期风电功率预测，能够根据风电场预测的出力曲线优化常规机组出力，降低运行本钱；增强电力系统的可靠性、稳定性；提升风电电力参与电力市场竞价能力。
基于均值出力的风电调峰成本量化方法
基于均值出力的风电调峰成本量化方法Wind power integration into the grid has been a hot topic in the energy sector due to its environmental benefits and potential to reduce reliance on fossil fuels. However, one of the challenges of wind power is its intermittency, which can lead to fluctuations in power generation. This issue becomes more pronounced during peak demand periods, when grid operators need to ensure a stable supply of electricity. To address this, strategies such as wind power curtailment or energy storage have been considered, but these solutions come with their own set of challenges, including cost implications.风力发电的融入电网一直是能源领域的热门话题，因为它具有环境优势，并且有潜力减少对化石燃料的依赖。
然而，风力发电面临的挑战之一是其间歇性，这可能导致发电量的波动。
在高峰期间，电网运营商需要确保稳定供电，这一问题变得尤为突出。
为了解决这个问题，人们考虑到了风力发电限电或储能等策略，但这些解决方案都伴随着各自的挑战，包括成本影响。
One method that has been proposed to quantify the cost of wind power peak shaving is based on the concept of mean dispatch cost.This approach involves calculating the difference between the average cost of dispatching electricity from wind power and the cost of dispatching electricity from other sources during peak periods. By comparing these costs, it is possible to estimate the financial implications of relying on wind power for peak shaving. However, this method may oversimplify the complexity of the electricity market and fail to capture all relevant factors that contribute to the cost of peak shaving.有一种提出的方法来量化风力调峰的成本，就是基于平均出力成本的概念。
风力发电原理英语作文
风力发电原理英语作文Title: The Principles of Wind Power Generation。
Wind power generation, a renewable energy source, harnesses the kinetic energy of wind to produce electricity. Understanding the principles behind wind power is crucialfor comprehending its significance in the renewable energy landscape.At its core, wind power generation involves the conversion of wind energy into electrical energy throughthe operation of wind turbines. These turbines consist of several key components, including blades, a rotor, a generator, and a tower.Firstly, the blades are designed to capture the kinetic energy present in the wind. Their shape and angle are optimized to maximize the efficiency of this process. Asthe wind blows, it exerts force on the blades, causing them to rotate.Next, the rotating blades turn a rotor connected to a generator. This rotor spins within a magnetic field, inducing an electric current in the generator's coils through electromagnetic induction. The generatedelectricity is then transmitted through cables connected to the turbine and subsequently integrated into the electrical grid for distribution.The height at which a wind turbine is positioned playsa crucial role in its efficiency. Wind speeds tend to increase with altitude due to reduced surface friction, making higher turbine placements more advantageous. Therefore, wind turbines are often installed on tall towers to capitalize on stronger and more consistent winds.Several factors influence the amount of electricity generated by a wind turbine. Wind speed is a primary determinant, as the power output of a turbine increaseswith higher wind speeds. Additionally, the size and designof the turbine, as well as the site's geographical location, terrain, and local weather patterns, all impact itsperformance.One of the key advantages of wind power generation isits sustainability. Unlike fossil fuels, wind energy is abundant and inexhaustible, making it a renewable resource with minimal environmental impact. By harnessing wind power, we can reduce our reliance on non-renewable energy sources and mitigate climate change.Furthermore, wind power is a form of clean energy, asit produces no greenhouse gas emissions or air pollutants during operation. This aspect is particularly crucial in combating global warming and air pollution, contributing to a healthier and more sustainable environment for future generations.However, it's essential to acknowledge the challenges associated with wind power generation. One such challengeis intermittency, as wind is not always predictable or consistent. To address this issue, advancements in energy storage technologies, grid integration, and forecasting methods are continuously being pursued to enhance thereliability and stability of wind power systems.In conclusion, wind power generation harnesses the natural force of wind to produce electricity, offering a sustainable and environmentally friendly alternative to conventional energy sources. By understanding the principles underlying wind power, we can appreciate its role in transitioning towards a cleaner and more sustainable energy future.。
基于改进型双向长短期记忆网络的风电超短期功率预测
基于改进型双向长短期记忆网络的风电超短期功率预测0引言风电功率预测及其在调度运行中的应用是促进新能源消纳的基础，研究风电功率预测技术有助于削弱风电功率并网时对电力系统带来的不利影响，降低电网的运行成本，提高系统运行的可靠性，有效保证电网安全[1]。
但是风力发电固有的高度依赖天气条件，随机性和波动性大，预测困难的特点，又限制了风力发电的大规模应用[2]。
风力发电的输出功率在很大程度上取决于风机所能接收到的风速大小，容易受到天气因素的影响。
安装在风电场的测风塔所测的风速，无法考虑整个风电场风速的波动性和随机性，其预测精度较低，在天气状况变化剧烈或者预测时间尺度较长时预测效果更差。
尤其是现有技术中基于实测风速测量历史值预测未来几个小时风速时，未能反应未来几个小时天气变化因素，从而导致风电场超短期功率预测不准确。
超短期风电功率预测是指预测风电场未来0~4小时的有功功率，时间分辨率不小于15分钟[3]。
文献[4]采用了长短期记忆（LSTM）单元对风电场的超短期风速进行了预测，并在此基础上进行了功率预测。
文献[5]提出风电场的风速之间存在较强的空间相关性，并给出了计及空间相关性的风速超短期预测方法。
文献[6]采用深度卷积神经网络和双向门控循环单元，分别提取风速、风向以及数值天气预报的时空特征，利用融合后的特征进行风速预测。
文献利用NWP信息，考虑多个风电场的空间相关性，提出了一种基于多位置NWP和门控循环单元的风电功率超短期预测。
本文提出一种基于改进型双向长短期记忆网络的风电超短功率预测方法。
首先以风电场实测瞬时风速、瞬时有功功率和风机状态为基础，进行数据异常点筛选，计算出风电场平均风速；根据风电场平均风速和对应时刻升压站并网点输出功率，进行二次数据筛选，从而减少因数据原因导致的误差。
然后，利用改进型双向长短期记忆网络对多变量时间序列与风电功率序列之间非线性关系进行冬天时间建模，构建预测模型。
最后，通过将历史样本划分成建模样本和测试样本，对网络模型进行有效的训练，完成风电场输出功率的超短期预测。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
IEEE, 9th International Conference on Environment and Electrical Engineering, Prague, Czech Republic, May 16 – 19, 2010
1
Wind Power Forecasting & Prediction Methods
II. FORECASTING & PREDICTIONS METHODS
There has also been much debate and discussion in regard to
1 A M Foley, Lecturer, Civil & Environmental Engineering Department, School of Engineering, University College Cork, College Rd., Cork, Republic of reland, Phone + 353 (0)21 427 2285 Email: aoife.foley@ucc.ie
2 P Leahy, Lecturer in Wind Energy Engineering, School of Engineering, University College Cork, College Rd., Cork, Republic of Ireland, Phone + 353 (0)21 427 2285 Email: p.leahy@ucc.ie
the costs associated with wind power integration. In the last five years a number of studies and reports have been commissioned and carried out by consultants and academics to investigate the costs [7 - 10]. Interestingly the costs associated with the integration of other renewable or non-renewable sources of power have gained less attention. This is probably due to the fact that wind power is relatively unpredictable. Traditionally grid systems are designed around fully ‘dispatchable’ power plant using measurable risk in terms of loss of load probability (LOLP), loss of load expectation (LOLE), energy not served (ENS), expected energy not served (EENS) and value of lost load (VOLL). The cost of backingup wind power with peaking plant with reserve power continues to be a difficulty for system operators.
3 E McKeogh, Senior Lecturer, Civil & Environmental Engineering Department, School of Engineering, University College Cork, College Rd., Cork, Republic of Ireland, Phone + 353 (0)21 427 2285 Email: e.mckeogh@ucc.ie
Keywords- Meteorology, Wind, Wind power forecasting, Wind power prediction
I. INTRODUCTION
Rapid growth in wind power has led to the need for advanced wind power forecasting and prediction methods. Accurate wind power forecasting and prediction reduces the risk in uncertainty and allows for better grid and power system integration. By and large wind forecasting and prediction is predominantly focused on the immediate short-term of seconds to minutes to the short-term of hours to typically two days and to the medium term of 2 to 7 days. Short-term wind power patterns are discussed in Reference [1]. Little research is available in relation to long term wind power forecasting and prediction. Long-term wind patterns and electricity demand is studied in Reference [2]. There are a number of studies which study long-term wind patterns, but these focus primarily on the effects of climate change [3 - 6]. The timescale of the model is related directly to the application of the forecasting and prediction. Very short-term forecasts (milliseconds to minutes) are usually associated with the active control of turbines at the wind farm and grid frequency and voltage stability. Whereas short-term forecasts and predictions (hours to 2 days) are power system planning (unit commitment and dispatch scheduling). Medium-term forecasts and predictions are identified for maintenance and outages planning of the wind farms, thermal generators and the grid system.
A. M. Foley1, Member, IEEE, P.G. Leahy2, Member, IEEE and E.J. McKeogh3, Member IEEE
Abstract — Globally on-shore wind power has seen considerable growth in all grid systems. In the coming decade offshore wind power is also expected to expand rapidly. Wind power is variable and intermittent over various time scales because it is weather dependent. Therefore wind power integration into traditional grids needs additional power system and electricity market planning and management for system balancing. This extra system balancing means that there is additional system costs associated with wind power assimilation. Wind power forecasting and prediction methods are used by system operators to plan unit commitment, scheduling and dispatch and by electricity traders and wind farm owners to maximize profit. Accurate wind power forecasting and prediction has numerous challenges. This paper presents a study of the existing and possible future methods used in wind power forecasting and prediction for both on-shore and off-shore wind farms.