【摘要】：在能源匱乏和環(huán)境問題的大背景下,世界各國開始尋求低碳發(fā)展道路,競(jìng)相發(fā)展可再生能源,風(fēng)電便是其中之一。風(fēng)電資源豐富、裝機(jī)靈活,風(fēng)電技術(shù)相較其他可再生能源發(fā)電而言也更成熟,效率更高,能很好地替代化石能源,保證能源供應(yīng)的前提下更好地保護(hù)好環(huán)境。經(jīng)過近些年的快速發(fā)展,我國已經(jīng)成為世界上風(fēng)電裝機(jī)最大的國家。但是,風(fēng)電具有的隨機(jī)性、間歇性和反調(diào)峰特性,嚴(yán)重影響了我國風(fēng)電的大規(guī)模并網(wǎng)消納,導(dǎo)致了嚴(yán)重的棄風(fēng)問題。因此,對(duì)風(fēng)電短期功率預(yù)測(cè)的研究可以彌補(bǔ)風(fēng)電不穩(wěn)定的缺點(diǎn),有利于電網(wǎng)更加合理地安排調(diào)度計(jì)劃,使得更多的風(fēng)電得到消納,有效地緩解棄風(fēng)問題,對(duì)我國風(fēng)電產(chǎn)業(yè)健康持續(xù)的發(fā)展具有重要意義。另一方面,隨著風(fēng)場(chǎng)大數(shù)據(jù)的逐步崛起,利用大數(shù)據(jù)進(jìn)行風(fēng)電功率預(yù)測(cè)是未來發(fā)展的一個(gè)趨勢(shì)。而深度學(xué)習(xí)在大數(shù)據(jù)的挖掘中正在發(fā)揮著越來越突出的貢獻(xiàn)。其中,卷積神經(jīng)網(wǎng)絡(luò)(CNNs)發(fā)展最為成熟,在圖像識(shí)別、模式識(shí)別等領(lǐng)域取得了成功。本文首先基于相鄰風(fēng)場(chǎng)大數(shù)據(jù)的結(jié)構(gòu)特點(diǎn),通過真實(shí)的數(shù)據(jù)構(gòu)建了三維的實(shí)驗(yàn)數(shù)據(jù)集,并運(yùn)用統(tǒng)計(jì)分布、動(dòng)態(tài)關(guān)聯(lián)性分析等方法,研究了實(shí)驗(yàn)數(shù)據(jù)集的數(shù)據(jù)特點(diǎn),為后續(xù)預(yù)測(cè)建模打下基礎(chǔ)。接著,建立了風(fēng)電短期功率CNNs預(yù)測(cè)模型,利用多個(gè)CNNs網(wǎng)絡(luò)獨(dú)立運(yùn)行,實(shí)現(xiàn)模型多輸出的效果;通過重點(diǎn)闡釋風(fēng)電短期功率CNNs預(yù)測(cè)模型建立的全過程,詳細(xì)分析模型的預(yù)測(cè)效果,驗(yàn)證了風(fēng)電短期功率CNNs預(yù)測(cè)模型的實(shí)用性和可靠性。結(jié)果顯示,CNNs預(yù)測(cè)模型在誤差控制上有較好的效果,在整體預(yù)測(cè)精度提升的同時(shí),對(duì)不同時(shí)間節(jié)點(diǎn)、不同功率樣本的預(yù)測(cè)效果較傳統(tǒng)方法而言,也更為平均。最后,通過建立CNNs預(yù)測(cè)模型和物理預(yù)測(cè)模型的組合預(yù)測(cè)模型,采用分類式的結(jié)構(gòu)權(quán)重,充分發(fā)揮兩種方法在不同樣本中的優(yōu)勢(shì),進(jìn)一步降低風(fēng)電短期功率預(yù)測(cè)誤差。實(shí)際工作中,將組合模型權(quán)重確定問題轉(zhuǎn)化成參數(shù)優(yōu)化問題,利用遺傳算法(SC)快速求解,效率高。從實(shí)驗(yàn)結(jié)果看,組合預(yù)測(cè)模型誤差較CNNs預(yù)測(cè)模型降低約5%,分類式的結(jié)構(gòu)權(quán)重也較單一權(quán)重下的誤差要略小。通過本文的研究,一定程度上證明了CNNs網(wǎng)絡(luò)方法在處理風(fēng)電短期功率預(yù)測(cè)問題上的大數(shù)據(jù)時(shí),有較好的應(yīng)用前景。
[Abstract]:In the background of energy shortage and environmental problems, countries in the world began to seek low-carbon development path, and competing for renewable energy, wind power is one of them. Compared with other renewable sources, wind power technology is more mature and more efficient. It can replace fossil energy and protect the environment better on the premise of ensuring energy supply. After the rapid development in recent years, China has become the largest wind power installed country in the world. However, the randomness, intermittency and anti-peak-shaving characteristics of wind power seriously affect the large-scale grid-connected dissipation of wind power in China, resulting in a serious wind abandonment problem. Therefore, the research on short-term power forecasting of wind power can make up for the shortcomings of instability of wind power, help the power grid to arrange the dispatching plan more reasonably, make more wind power be absorbed, and effectively alleviate the problem of wind abandonment. It is of great significance to the healthy and sustainable development of wind power industry in China. On the other hand, with the gradual rise of wind field big data, using big data to forecast wind power is a trend in the future. And in-depth learning in big data's excavation is playing a more and more prominent contribution. Among them, convolutional neural network (CNNs) is the most mature and has been successful in image recognition and pattern recognition. Firstly, based on the structural characteristics of the adjacent wind field big data, a three-dimensional experimental data set is constructed through real data, and the data characteristics of the experimental data set are studied by means of statistical distribution, dynamic correlation analysis, and so on. It lays a foundation for the following prediction modeling. Then, a short-term wind power CNNs prediction model is established, which uses multiple CNNs networks to run independently to realize the effect of multi-output of the model. The whole process of the establishment of short-term wind power CNNs prediction model is explained, and the forecasting effect of the model is analyzed in detail. The practicability and reliability of the wind power short-term CNNs prediction model are verified. The results show that the CNNs prediction model has a good effect on error control. While the overall prediction accuracy is improved, the prediction effect of different time nodes and different power samples is more average than that of the traditional method. Finally, the combination forecasting model of CNNs prediction model and physical prediction model is established, and the classification structure weight is adopted to give full play to the advantages of the two methods in different samples, so as to further reduce the short-term power prediction error of wind power. In practical work, the weight determination problem of combinatorial model is transformed into parameter optimization problem, and genetic algorithm (SC) is used to solve the problem quickly, which has high efficiency. The experimental results show that the error of the combined prediction model is about 5% lower than that of the CNNs prediction model, and the error of the structure weight of the classification formula is slightly smaller than that of the single weight. Through the research in this paper, it is proved that the CNNs network method has a good application prospect in dealing with big data in the short-term power prediction of wind power.
【學(xué)位授予單位】：華北電力大學(xué)(北京)
【學(xué)位級(jí)別】：碩士
【學(xué)位授予年份】：2017
【分類號(hào)】：TM614

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 周勇;吳瑕;周為;狄宏林;;深度學(xué)習(xí)發(fā)展來源研究[J];數(shù)碼世界;2016年10期

2 周強(qiáng);汪寧渤;冉亮;沈薈云;呂清泉;王明松;;中國新能源棄風(fēng)棄光原因分析及前景探究[J];中國電力;2016年09期

3 張穎超;肖寅;鄧華;王璐;;基于OS-ELM的風(fēng)速修正及短期風(fēng)電功率預(yù)測(cè)[J];電子技術(shù)應(yīng)用;2016年02期

4 劉松;王俊;王端陽;李文華;邵丹;;基于灰色關(guān)聯(lián)分析和黑洞粒子群優(yōu)化算法的短期風(fēng)電功率預(yù)測(cè)[J];電器與能效管理技術(shù);2015年13期

5 王印松;蘇子卿;;一種基于相鄰風(fēng)機(jī)測(cè)量數(shù)據(jù)相關(guān)性分析的風(fēng)速預(yù)測(cè)方法[J];華北電力大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年02期

6 鐘海旺;夏清;張健;張國強(qiáng);;激勵(lì)風(fēng)電場(chǎng)提升功率預(yù)測(cè)精度的機(jī)制設(shè)計(jì)[J];電力系統(tǒng)自動(dòng)化;2015年05期

7 劉愛國;薛云濤;胡江鷺;劉路平;;基于GA優(yōu)化SVM的風(fēng)電功率的超短期預(yù)測(cè)[J];電力系統(tǒng)保護(hù)與控制;2015年02期

8 朱軍;胡文波;;貝葉斯機(jī)器學(xué)習(xí)前沿進(jìn)展綜述[J];計(jì)算機(jī)研究與發(fā)展;2015年01期

9 錢曉東;劉維奇;;基于時(shí)間序列分析的風(fēng)電功率預(yù)測(cè)模型[J];電力學(xué)報(bào);2014年04期

10 肖遷;李文華;李志剛;劉金龍;劉會(huì)巧;;基于改進(jìn)的小波-BP神經(jīng)網(wǎng)絡(luò)的風(fēng)速和風(fēng)電功率預(yù)測(cè)[J];電力系統(tǒng)保護(hù)與控制;2014年15期

相關(guān)碩士學(xué)位論文 前5條

1 鄭文書;基于時(shí)空相關(guān)性的區(qū)域風(fēng)電場(chǎng)群風(fēng)速預(yù)測(cè)研究[D];華北電力大學(xué);2014年

2 王建成;短期風(fēng)電功率預(yù)測(cè)方法研究[D];華南理工大學(xué);2013年

3 于安興;風(fēng)電場(chǎng)短期風(fēng)電功率預(yù)測(cè)研究[D];華東理工大學(xué);2013年

4 夏冬;基于時(shí)間序列分析的大型風(fēng)電場(chǎng)功率預(yù)測(cè)方法研究[D];北京交通大學(xué);2012年

5 賀電;大型風(fēng)電場(chǎng)短期功率預(yù)測(cè)研究[D];北京交通大學(xué);2011年

，

本文編號(hào)：2467224