本文選題：數(shù)據(jù)融合　切入點(diǎn)：風(fēng)電預(yù)測(cè)　出處：《華北電力大學(xué)》2017年碩士論文

【摘要】：科學(xué)技術(shù)飛速發(fā)展的同時(shí),能源消耗也在迅速增長(zhǎng),從而導(dǎo)致化石能源日益枯竭。因此,風(fēng)能作為一種清潔能源,得到廣泛的開發(fā)利用。在我們大規(guī)模利用風(fēng)能的同時(shí),也出現(xiàn)了很多問題,比如在風(fēng)電并網(wǎng)時(shí),由于風(fēng)能的不穩(wěn)定性對(duì)整個(gè)電網(wǎng)系統(tǒng)的穩(wěn)定運(yùn)行造成沖擊。目前為止,風(fēng)能的不穩(wěn)定性在很大程度上限制了風(fēng)電的發(fā)展。風(fēng)電企業(yè)信息系統(tǒng)中最為重要的一部分就是風(fēng)電功率預(yù)測(cè),因?yàn)轱L(fēng)能的不穩(wěn)定性,就導(dǎo)致風(fēng)電的波動(dòng)性和間歇性以及反調(diào)峰特性,實(shí)現(xiàn)風(fēng)電功率的準(zhǔn)確預(yù)測(cè),電網(wǎng)就可以根據(jù)預(yù)測(cè)功率,制定合理的功率調(diào)度計(jì)劃,保證電網(wǎng)的穩(wěn)定性,同時(shí)也可以最大效率的利用風(fēng)能,避免了因風(fēng)電功率不穩(wěn)定和調(diào)度問題從而導(dǎo)致風(fēng)電不能物盡其用。合理的風(fēng)電企業(yè)信息系統(tǒng)可以協(xié)助電網(wǎng)完成功率調(diào)度問題,同時(shí)促進(jìn)了風(fēng)能的利用率,減少了化石能源的消耗。本文研究了數(shù)據(jù)融合技術(shù),從數(shù)據(jù)融合技術(shù)的定義、分類和數(shù)據(jù)融合方法三個(gè)方面進(jìn)行研究。主要研究了像素級(jí)融合、特征層融合和決策層融合。還詳細(xì)介紹了數(shù)據(jù)融合中的貝葉斯方法、D_S證據(jù)推理、人工神經(jīng)網(wǎng)絡(luò)和模糊理論算法等經(jīng)典方法。此外本文還從風(fēng)電功率預(yù)測(cè)方法和風(fēng)電功率預(yù)測(cè)類型等方面認(rèn)真研究了風(fēng)電功率預(yù)測(cè)技術(shù)。本文通過對(duì)比現(xiàn)有的風(fēng)電功率預(yù)測(cè)技術(shù),結(jié)合數(shù)據(jù)融合技術(shù),本文設(shè)計(jì)了基于數(shù)據(jù)融合的風(fēng)電功率預(yù)測(cè)算法。該算法分為兩個(gè)階段,第一階段通過貼近度進(jìn)行篩選、清洗,最后將數(shù)據(jù)進(jìn)行融合。第二個(gè)階段通過四步:建立神經(jīng)網(wǎng)絡(luò),制定模糊控制規(guī)則,設(shè)計(jì)模糊推理,最后將輸出變量還原,得到風(fēng)電功率的預(yù)測(cè)值。該算法實(shí)時(shí)性較好,且排除了不精確數(shù)據(jù)的誤差。最后本文進(jìn)行了風(fēng)電企業(yè)信息系統(tǒng)的設(shè)計(jì)研究,該系統(tǒng)主要應(yīng)用與風(fēng)電功率的預(yù)測(cè),從而滿足本地的管理需求和并網(wǎng)需要。主要從系統(tǒng)總體的設(shè)計(jì)思想,需求分析和系統(tǒng)結(jié)構(gòu)與開發(fā)框架進(jìn)行分析研究。通過將數(shù)據(jù)融合算法應(yīng)用到系統(tǒng)中完成了風(fēng)電企業(yè)信息系統(tǒng)的設(shè)計(jì),并進(jìn)行了界面展示。本文將數(shù)據(jù)融合技術(shù)應(yīng)用于風(fēng)電數(shù)據(jù)處理,解決風(fēng)電企業(yè)信息化中的風(fēng)電預(yù)測(cè)問題,這應(yīng)用了數(shù)據(jù)融合技術(shù)可以全面的去除冗余數(shù)據(jù),綜合多樣信息,得到正確結(jié)果的特點(diǎn),方便風(fēng)電企業(yè)構(gòu)建合適的風(fēng)電企業(yè)信息系統(tǒng),解決風(fēng)電并網(wǎng)造成的電網(wǎng)不穩(wěn)定問題。
[Abstract]:With the rapid development of science and technology, energy consumption is also increasing rapidly, which leads to the depletion of fossil energy. Therefore, wind energy, as a clean energy, has been widely developed and utilized. There are also a lot of problems. For example, when wind power is connected to the grid, the instability of wind power has an impact on the stability of the entire power system. So far, The instability of wind energy limits the development of wind power to a large extent. The most important part of wind power enterprise information system is wind power prediction, because of the instability of wind energy. As a result of wind power fluctuation, intermittence and backpeak-shaving characteristics, and accurate prediction of wind power, the power grid can make a reasonable power dispatching plan according to the predicted power and ensure the stability of the power grid. At the same time, wind energy can be used efficiently to avoid the problem of wind power instability and dispatch, which leads to wind power not being able to make the best use of it. A reasonable wind power enterprise information system can help the power grid to complete the power dispatching problem. At the same time, it promotes the utilization of wind energy and reduces the consumption of fossil energy. In this paper, the data fusion technology is studied, including the definition, classification and data fusion methods of data fusion technology. Feature level fusion and decision level fusion. The Bayesian method in data fusion is also introduced in detail. The classical methods such as artificial neural network and fuzzy theory algorithm are also studied in this paper. In addition, wind power prediction methods and types of wind power prediction are carefully studied in this paper. Combined with data fusion technology, this paper designs a wind power prediction algorithm based on data fusion. The algorithm is divided into two stages. Finally, the data is fused. In the second stage, the neural network is established, the fuzzy control rules are formulated, the fuzzy reasoning is designed, and the output variables are restored to get the predictive value of the wind power. Finally, this paper studies the design of wind power enterprise information system, which is mainly used to predict wind power. In order to meet the local management needs and network connection needs. Mainly from the overall design ideas of the system, Through the application of data fusion algorithm to the system, the design of wind power enterprise information system is completed. The data fusion technology is applied to wind power data processing to solve the wind power forecasting problem in wind power enterprise informatization. This data fusion technology can remove redundant data comprehensively and synthesize various information. It is convenient for wind power enterprises to build appropriate wind power enterprise information system and solve the problem of power grid instability caused by wind power grid connection.
【學(xué)位授予單位】：華北電力大學(xué)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TP202;TP311.52;TM614

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 王新友;王晨華;張yN;魯江;;基于改進(jìn)型ESN的短期風(fēng)電功率預(yù)測(cè)[J];自動(dòng)化與儀器儀表;2017年02期

2 黎靜華;桑川川;甘一夫;潘毅;;風(fēng)電功率預(yù)測(cè)技術(shù)研究綜述[J];現(xiàn)代電力;2017年03期

3 江岳春;楊旭瓊;賀飛;陳禮鋒;何鐘南;;基于EEMD-IGSA-LSSVM的超短期風(fēng)電功率預(yù)測(cè)[J];湖南大學(xué)學(xué)報(bào)(自然科學(xué)版);2016年10期

4 王麗婕;厲虹;方市彬;;基于人工魚群優(yōu)化算法的支持向量機(jī)短期風(fēng)電功率預(yù)測(cè)模型[J];電氣工程學(xué)報(bào);2016年10期

5 楊茂;董駿城;;基于混合分布模型的風(fēng)電功率波動(dòng)特性研究[J];中國(guó)電機(jī)工程學(xué)報(bào);2016年S1期

6 蘭飛;桑川川;梁浚杰;黎靜華;;基于條件Copula函數(shù)的風(fēng)電功率區(qū)間預(yù)測(cè)[J];中國(guó)電機(jī)工程學(xué)報(bào);2016年S1期

7 季峰;蔡興國(guó);王超柱;;基于弱魯棒優(yōu)化的含風(fēng)電電力系統(tǒng)調(diào)度方法[J];中國(guó)電機(jī)工程學(xué)報(bào);2016年17期

8 張延龍;王俊勇;;多傳感器數(shù)據(jù)融合技術(shù)概述[J];艦船電子工程;2013年02期

9 戴晨鋮;;多傳感器信息融合綜述[J];科技視界;2012年26期

10 劉少剛;呂建偉;王士成;;基于數(shù)據(jù)融合與神經(jīng)網(wǎng)絡(luò)的火災(zāi)探測(cè)研究[J];應(yīng)用科技;2011年05期

相關(guān)博士學(xué)位論文 前1條

1 王碩禾;基于短期負(fù)荷預(yù)測(cè)技術(shù)的電能控制系統(tǒng)研究[D];天津大學(xué);2009年

相關(guān)碩士學(xué)位論文 前7條

1 劉桐歡;無線傳感器網(wǎng)絡(luò)多源數(shù)據(jù)融合技術(shù)研究[D];北京交通大學(xué);2014年

2 劉濤;信息融合算法及其應(yīng)用研究[D];南京郵電大學(xué);2013年

3 竇晨超;基于多源信息融合方法的火災(zāi)檢測(cè)系統(tǒng)的算法研究[D];華東理工大學(xué);2013年

4 劉小玲;三維GIS輸電線路運(yùn)行管理系統(tǒng)[D];中南大學(xué);2012年

5 周曉琳;基于神經(jīng)網(wǎng)絡(luò)的多傳感器數(shù)據(jù)融合火災(zāi)預(yù)警系統(tǒng)研究[D];長(zhǎng)春理工大學(xué);2012年

6 韓菁;多傳感器數(shù)據(jù)融合技術(shù)在火災(zāi)探測(cè)中的應(yīng)用研究[D];重慶理工大學(xué);2011年

7 袁剛;無線傳感器網(wǎng)絡(luò)數(shù)據(jù)融合系統(tǒng)設(shè)計(jì)[D];北京郵電大學(xué);2009年

，

本文編號(hào)：1687257