基于BP神經(jīng)網(wǎng)絡(luò)-馬爾科夫鏈的光伏發(fā)電預(yù)測 

吳雪蓮，都洪基 
(南京理工大學(xué)，江蘇 南京 210094) 
 

摘 要：為了減少發(fā)電量隨機(jī)性對(duì)電力系統(tǒng)的影響，需要對(duì)發(fā)電功率預(yù)測進(jìn)行研究。通過分析影響光伏發(fā)電功率的因素，基于BP神經(jīng)網(wǎng)絡(luò)理論，在Matlab軟件中建立預(yù)測模型，實(shí)現(xiàn)了對(duì)輸出功率的短期預(yù)測，并給出了基于馬爾科夫鏈的改進(jìn)預(yù)測精度的方法。
關(guān)鍵詞：光伏發(fā)電；預(yù)測模型；神經(jīng)網(wǎng)絡(luò)；馬爾科夫鏈
中圖分類號(hào)：TP183 文獻(xiàn)標(biāo)識(shí)碼：A 文章編號(hào)：1007-3175(2014)03-0022-06

Photovoltaic Power Generation Forecast Based on BPNeural Network and Markov Chain 

WU Xue-lian, DU Hong-ji 
（Nanjing University of Science and Technology, Nanjing 20094, China） 
 

Abstract: In order to reduce impact of randomness generation capacity on power system, there is a need to study generation power forecast. Via analysis of factors of impact on photovoltaic generation power, this paper established the forecast model in Matlab based on BP neural network theory, realizing the short term forecast of output power and giving the method of improved forecast accuracy based on Markov chain.
        Key words: photovoltaic power generation; prediction model; neural network; Markov chain

參考文獻(xiàn)
[1] 張艷霞，趙杰.基于反饋型神經(jīng)網(wǎng)絡(luò)的光伏系統(tǒng)發(fā)電功率預(yù)測[J].電力系統(tǒng)保護(hù)與控制，2011，39(15)：96-101.
[2] 陳允平，王旭蕊，韓寶亮.人工神經(jīng)網(wǎng)絡(luò)原理及其應(yīng)用[M].北京：中國電力出版社，2002.
[3] Chen T Y，Athienitis A K.Ambient Temperature and Solar Radiation Prediction for Predictive Control of HVAC Systems and a Methodology for Optimal Building Heating Dynamic Operation[J].ASHRAE Transaction, 1996, 102(1)：26-36.
[4] 姜僑娜，陳中.BP_馬爾科夫組合預(yù)測方法在光伏發(fā)電量預(yù)測中的應(yīng)用[J].電力需求側(cè)管理，2011，13(6)：21-24.
[5] 董繼征，王桓，何怡剛，等.加權(quán)馬爾可夫鏈在負(fù)荷預(yù)測中的應(yīng)用[J].繼電器，2006，34(3)：32-36.
[6] 王義民，于興杰，暢建霞，等.基于BP神經(jīng)網(wǎng)絡(luò)馬爾科夫模型的徑流量預(yù)測[J].武漢大學(xué)學(xué)報(bào)，2008，41(5)：14-17.