發(fā)布時(shí)間：2018-05-25 22:43

  本文選題：風(fēng)力發(fā)電 + 功率預(yù)測(cè)��； 參考：《華北電力大學(xué)》2017年碩士論文

【摘要】：風(fēng)電出力具有波動(dòng)性和間歇性,大規(guī)模風(fēng)電接入系統(tǒng)會(huì)對(duì)電網(wǎng)的電能質(zhì)量帶來(lái)不良影響,對(duì)電網(wǎng)的安全、穩(wěn)定運(yùn)行帶來(lái)嚴(yán)峻挑戰(zhàn),精確的風(fēng)電場(chǎng)輸出功率預(yù)測(cè)是應(yīng)對(duì)大規(guī)模風(fēng)電并網(wǎng)問(wèn)題,提高風(fēng)電比例的有效手段之一。當(dāng)前,國(guó)內(nèi)對(duì)風(fēng)電功率預(yù)測(cè)還處于理論研究階段,開(kāi)發(fā)并應(yīng)用于實(shí)際的成熟可靠的預(yù)測(cè)系統(tǒng)較少,實(shí)踐經(jīng)驗(yàn)缺乏�？梢�(jiàn),對(duì)風(fēng)電輸出功率預(yù)測(cè)方法進(jìn)行研究具有重要的理論意義和現(xiàn)實(shí)意義,本文選擇風(fēng)電場(chǎng)輸出功率短期預(yù)測(cè)方法進(jìn)行研究。本文針對(duì)風(fēng)電功率具有復(fù)雜的非線性、非平穩(wěn)的特性,提出了一種基于經(jīng)驗(yàn)?zāi)B(tài)分解和極端學(xué)習(xí)機(jī)相結(jié)合的風(fēng)電功率預(yù)測(cè)方法。該方法首先運(yùn)用經(jīng)驗(yàn)?zāi)B(tài)分解法對(duì)原始風(fēng)電功率時(shí)間序列進(jìn)行分解處理,然后根據(jù)各個(gè)分量的特點(diǎn)分別建立合適的極端學(xué)習(xí)機(jī)預(yù)測(cè)模型,最后將各分量的預(yù)測(cè)值疊加得到最終預(yù)測(cè)值。結(jié)果表明,EMD處理降低了建模和預(yù)測(cè)的難度,提高了風(fēng)電功率預(yù)測(cè)精度,并且極端學(xué)習(xí)機(jī)在學(xué)習(xí)速度和泛化性能上比傳統(tǒng)BP神經(jīng)網(wǎng)絡(luò)具有更大的優(yōu)勢(shì)。為了改善極端學(xué)習(xí)機(jī)算法隨機(jī)選取隱含層參數(shù)造成的模型不穩(wěn)定問(wèn)題,本文提出了一種基于經(jīng)驗(yàn)?zāi)B(tài)分解和核極端學(xué)習(xí)機(jī)的預(yù)測(cè)方法。仿真結(jié)果表明,核極端學(xué)習(xí)機(jī)算法引入核函數(shù)映射代替極端學(xué)習(xí)機(jī)算法的隱含層映射,預(yù)測(cè)模型在穩(wěn)定性和預(yù)測(cè)精度上都有了較大改善。為了進(jìn)一步提高模型的預(yù)測(cè)精度,結(jié)合多核學(xué)習(xí)算法,本文提出了一種基于經(jīng)驗(yàn)?zāi)B(tài)分解和多核極端學(xué)習(xí)機(jī)的功率預(yù)測(cè)方法。多核函數(shù)集合了多種基礎(chǔ)核函數(shù)的特點(diǎn),能更好的提取數(shù)據(jù)樣本間的特征信息,具有更強(qiáng)的學(xué)習(xí)能力。仿真結(jié)果表明,該方法的風(fēng)電功率預(yù)測(cè)效果得到有效提高,模型的預(yù)測(cè)精度更高,泛化性能更強(qiáng),驗(yàn)證了該方法在風(fēng)電功率預(yù)測(cè)中的有效性。
[Abstract]:Wind power output is volatile and intermittent. Large-scale wind power access system will bring adverse impact on power quality of power grid, and bring severe challenges to the security and stability of power grid. Accurate prediction of wind farm output power is one of the effective methods to solve the problem of large-scale wind power grid connection and improve wind power ratio. At present, wind power prediction in China is still in the stage of theoretical research, the development and application of practical mature and reliable forecasting system is less, and practical experience is lacking. Therefore, it is of great theoretical and practical significance to study the prediction method of wind power output power. In this paper, the short-term prediction method of wind farm output power is chosen to study. Aiming at the complex nonlinear and non-stationary characteristics of wind power, a wind power prediction method based on empirical mode decomposition and extreme learning machine is proposed in this paper. In this method, the original wind power time series is decomposed by empirical mode decomposition method, and then an appropriate extreme learning machine prediction model is established according to the characteristics of each component. Finally, the final prediction value is obtained by superposing the predicted values of each component. The results show that the EMD processing reduces the difficulty of modeling and prediction, improves the precision of wind power prediction, and the extreme learning machine has more advantages than the traditional BP neural network in learning speed and generalization performance. In order to improve the model instability caused by random selection of hidden layer parameters in extreme learning machine, a prediction method based on empirical mode decomposition and kernel extreme learning machine is proposed in this paper. The simulation results show that the kernel function mapping is introduced into the kernel extreme learning machine algorithm instead of the hidden layer mapping of the extreme learning machine algorithm, and the stability and prediction accuracy of the prediction model are greatly improved. In order to further improve the prediction accuracy of the model, a power prediction method based on empirical mode decomposition and multi-core extreme learning machine is proposed. The multi-kernel functions have the characteristics of many basic kernel functions, which can extract the feature information between the data samples better, and have a stronger learning ability. The simulation results show that the prediction effect of the proposed method is improved effectively, the prediction accuracy of the model is higher and the generalization performance is better. The effectiveness of the proposed method in wind power prediction is verified.
【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)】：TM614;TP181

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 李冬輝;閆振林;姚樂(lè)樂(lè);鄭宏宇;;基于改進(jìn)流形正則化極限學(xué)習(xí)機(jī)的短期電力負(fù)荷預(yù)測(cè)[J];高電壓技術(shù);2016年07期

2 包廣清;宋澤;吳國(guó)棟;徐海龍;;基于經(jīng)驗(yàn)?zāi)B(tài)分解和形態(tài)學(xué)的風(fēng)電并網(wǎng)電壓故障檢測(cè)[J];農(nóng)業(yè)工程學(xué)報(bào);2016年11期

3 李軍;李大超;;基于優(yōu)化核極限學(xué)習(xí)機(jī)的風(fēng)電功率時(shí)間序列預(yù)測(cè)[J];物理學(xué)報(bào);2016年13期

4 路濤;趙靚;;2015年中國(guó)風(fēng)電開(kāi)發(fā)主要數(shù)據(jù)匯總[J];風(fēng)能;2016年04期

5 ;2015年中國(guó)風(fēng)電裝機(jī)容量統(tǒng)計(jì)[J];風(fēng)能;2016年02期

6 雷亞國(guó);陳吳;李乃鵬;林京;;自適應(yīng)多核組合相關(guān)向量機(jī)預(yù)測(cè)方法及其在機(jī)械設(shè)備剩余壽命預(yù)測(cè)中的應(yīng)用[J];機(jī)械工程學(xué)報(bào);2016年01期

7 杜占龍;李小民;鄭宗貴;張國(guó)榮;毛瓊;;基于正則化與遺忘因子的極限學(xué)習(xí)機(jī)及其在故障預(yù)測(cè)中的應(yīng)用[J];儀器儀表學(xué)報(bào);2015年07期

8 王長(zhǎng)路;王偉功;張立勇;喬雪濤;;中國(guó)風(fēng)電產(chǎn)業(yè)發(fā)展分析[J];重慶大學(xué)學(xué)報(bào);2015年01期

9 王新迎;韓敏;;多元混沌時(shí)間序列的多核極端學(xué)習(xí)機(jī)建模預(yù)測(cè)[J];物理學(xué)報(bào);2015年07期

10 賀彥林;王曉;朱群雄;;基于主成分分析-改進(jìn)的極限學(xué)習(xí)機(jī)方法的精對(duì)苯二甲酸醋酸含量軟測(cè)量[J];控制理論與應(yīng)用;2015年01期

相關(guān)博士學(xué)位論文 前5條

1 陳昊;基于波動(dòng)性模型的風(fēng)電功率預(yù)測(cè)研究[D];東南大學(xué);2015年

2 李小冬;核極限學(xué)習(xí)機(jī)的理論與算法及其在圖像處理中的應(yīng)用[D];浙江大學(xué);2014年

3 熊濤;基于EMD的時(shí)間序列預(yù)測(cè)混合建模技術(shù)及其應(yīng)用研究[D];華中科技大學(xué);2014年

4 楊易e，

本文編號(hào)：1934926