【摘要】：隨著化石能源的全面緊張和環(huán)境污染的日益加劇,世界各國(guó)普遍把開發(fā)、利用可再生能源作為其重要的能源發(fā)展戰(zhàn)略。而在可再生能源中,風(fēng)能是發(fā)展最快的清潔能源,風(fēng)力發(fā)電方式也最具規(guī)模開發(fā)需求和商業(yè)發(fā)展前景。就目前情況來看,風(fēng)力發(fā)電設(shè)備與技術(shù)已相對(duì)成熟,但是由于風(fēng)力發(fā)電過程中的隨機(jī)性、波動(dòng)性和間歇性,造成風(fēng)電輸出功率穩(wěn)定性較差,因此限電棄風(fēng)問題嚴(yán)重制約風(fēng)電并網(wǎng)。目前存在的主要問題是如何提高風(fēng)電功率預(yù)測(cè)的精度,特別是未來24小時(shí)的預(yù)測(cè)精度�；谝陨媳尘�,本文主要就以下幾個(gè)方面展開研究。(1)風(fēng)力發(fā)電系統(tǒng)歷史數(shù)據(jù)的真實(shí)可靠是進(jìn)行風(fēng)電功率預(yù)測(cè)的基礎(chǔ),而在風(fēng)力發(fā)電系統(tǒng)運(yùn)行或數(shù)據(jù)采集、測(cè)量、傳輸、轉(zhuǎn)換等環(huán)節(jié),尤其是人為的限電棄風(fēng),歷史數(shù)據(jù)中不可避免的存在異常數(shù)據(jù)。本文在分析風(fēng)電場(chǎng)異常數(shù)據(jù)特征的基礎(chǔ)上,采用四分位數(shù)法對(duì)風(fēng)電系統(tǒng)棄風(fēng)數(shù)據(jù)進(jìn)行前期預(yù)處理,提高歷史數(shù)據(jù)的準(zhǔn)確率。(2)相比于其他智能預(yù)測(cè)算法,人工神經(jīng)網(wǎng)絡(luò)在自學(xué)習(xí)性、自適應(yīng)性、魯棒性、容錯(cuò)性和推廣能力方面性能表現(xiàn)突出。當(dāng)前風(fēng)電功率預(yù)測(cè)中使用較多的人工神經(jīng)網(wǎng)絡(luò)是靜態(tài)神經(jīng)網(wǎng)絡(luò),而利用靜態(tài)型神經(jīng)網(wǎng)絡(luò)進(jìn)行預(yù)測(cè),造成風(fēng)電功率序列喪失時(shí)變特性能力,因此預(yù)測(cè)精度不高。所以,本文選用能夠更好反映風(fēng)電功率動(dòng)態(tài)特征的Elman神經(jīng)網(wǎng)絡(luò),給出Elman神經(jīng)網(wǎng)絡(luò)風(fēng)電功率預(yù)測(cè)算法。(3)Elman神經(jīng)網(wǎng)絡(luò)所采用的網(wǎng)絡(luò)參數(shù)會(huì)影響網(wǎng)絡(luò)性能,而當(dāng)前在神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)階段,普遍采用固定梯度變化方向的梯度下降方法,采用這種方法會(huì)存在收斂速度慢、易陷入局部最優(yōu)解等缺陷,這些缺陷限制了網(wǎng)絡(luò)的尋優(yōu)能力。因此,本文采用具有全局尋優(yōu)性能的改進(jìn)布谷鳥搜索算法優(yōu)化Elman神經(jīng)網(wǎng)絡(luò)的權(quán)值和閾值,目的在于提高Elman神經(jīng)網(wǎng)絡(luò)的穩(wěn)定性和泛化能力。(4)針對(duì)傳統(tǒng)單機(jī)計(jì)算資源和存儲(chǔ)資源不能很好的滿足短期風(fēng)電功率預(yù)測(cè)的實(shí)際需求,本文將改進(jìn)型布谷鳥搜索算法和Elman神經(jīng)網(wǎng)絡(luò)進(jìn)行并行化設(shè)計(jì),在Spark云平臺(tái)上對(duì)算法進(jìn)行性能測(cè)試。實(shí)驗(yàn)分析表明,預(yù)測(cè)精度和實(shí)時(shí)性均優(yōu)于傳統(tǒng)單機(jī)功率預(yù)測(cè)算法。
[Abstract]:With the development of fossil energy and the increasing environmental pollution, renewable energy is regarded as an important strategy of energy development in the world. In renewable energy, wind energy is the fastest developing clean energy, and wind power generation has the most large-scale development demand and commercial development prospects. As far as the current situation is concerned, wind power generation equipment and technology have been relatively mature, but due to the randomness, volatility and intermittency of wind power generation process, the stability of wind power output is poor. As a result, the problem of limiting the wind power abandonment seriously restricts the wind power grid. At present, the main problem is how to improve the accuracy of wind power prediction, especially in the next 24 hours. Based on the above background, this paper mainly studies the following aspects: (1) the real and reliable historical data of wind power generation system is the basis of wind power prediction, and in the operation of wind power system or data acquisition, measurement, transmission, Conversion and other links, especially artificial power restriction wind, historical data inevitably exist abnormal data. On the basis of analyzing the characteristics of abnormal wind farm data, this paper uses the quartile method to pre-process the abandoned wind data of wind power system in order to improve the accuracy of historical data. (2) compared with other intelligent prediction algorithms, The performance of artificial neural network is outstanding in the aspects of self-learning, adaptability, robustness, fault tolerance and generalization ability. At present, more artificial neural networks are used in wind power prediction, but static neural network is used to forecast wind power series, which results in the loss of time-varying capability of wind power series, so the prediction accuracy is not high. Therefore, the Elman neural network which can better reflect the dynamic characteristics of wind power is chosen in this paper, and the wind power prediction algorithm based on Elman neural network is given. (3) the network parameters used by Elman neural network will affect the performance of the network. At present, in the learning phase of neural networks, the gradient descent method with fixed gradient change direction is generally used. This method will have some defects such as slow convergence rate, easy to fall into local optimal solution and so on, which limit the optimization ability of the network. Therefore, the improved cuckoo search algorithm with global optimization performance is used to optimize the weights and thresholds of the Elman neural network. The purpose of this paper is to improve the stability and generalization ability of Elman neural network. (4) the traditional single-machine computing resources and storage resources can not meet the actual demand of short-term wind power prediction. In this paper, the improved cuckoo search algorithm and Elman neural network are designed in parallel, and the performance of the algorithm is tested on the Spark cloud platform. Experimental analysis shows that the prediction accuracy and real-time performance are better than the traditional single-machine power prediction algorithm.
【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)】：TP18;TM614

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 錢政;裴巖;曹利宵;王婧怡;荊博;;風(fēng)電功率預(yù)測(cè)方法綜述[J];高電壓技術(shù);2016年04期

2 王詔遠(yuǎn);王宏杰;邢煥來;李天瑞;;基于Spark的蟻群優(yōu)化算法[J];計(jì)算機(jī)應(yīng)用;2015年10期

3 宋寶燕;王俊陸;王妍;;基于范德蒙碼的HDFS優(yōu)化存儲(chǔ)策略研究[J];計(jì)算機(jī)學(xué)報(bào);2015年09期

4 趙敏;尤冬梅;;基于FOA_Elman神經(jīng)網(wǎng)絡(luò)的微網(wǎng)短期負(fù)荷預(yù)測(cè)[J];智能電網(wǎng);2015年09期

5 丁明;劉志;畢銳;朱衛(wèi)平;;基于灰色系統(tǒng)校正-小波神經(jīng)網(wǎng)絡(luò)的光伏功率預(yù)測(cè)[J];電網(wǎng)技術(shù);2015年09期

6 段學(xué)偉;王瑞琪;王昭鑫;郎澄宇;孫樹敏;趙鵬;鄭偉;;風(fēng)速及風(fēng)電功率預(yù)測(cè)研究綜述[J];山東電力技術(shù);2015年07期

7 徐曉飛;劉志中;王忠杰;閔尋優(yōu);劉睿霖;王海芳;;S-ABC——面向服務(wù)領(lǐng)域的人工蜂群算法范型[J];計(jì)算機(jī)學(xué)報(bào);2015年11期

8 胡俊;胡賢德;程家興;;基于Spark的大數(shù)據(jù)混合計(jì)算模型[J];計(jì)算機(jī)系統(tǒng)應(yīng)用;2015年04期

9 薛禹勝;郁琛;趙俊華;Kang LI;Xueqin LIU;Qiuwei WU;Guangya YANG;;關(guān)于短期及超短期風(fēng)電功率預(yù)測(cè)的評(píng)述[J];電力系統(tǒng)自動(dòng)化;2015年06期

10 馬友忠;孟小峰;;云數(shù)據(jù)管理索引技術(shù)研究[J];軟件學(xué)報(bào);2015年01期

相關(guān)博士學(xué)位論文 前1條

1 馮春時(shí);群智能優(yōu)化算法及其應(yīng)用[D];中國(guó)科學(xué)技術(shù)大學(xué);2009年

相關(guān)碩士學(xué)位論文 前1條

1 陳辰;基于卡爾曼濾波算法的短期風(fēng)電功率預(yù)測(cè)[D];新疆大學(xué);2015年

，

本文編號(hào)：2444825