基于FOA-Elman神經(jīng)網(wǎng)絡(luò)的光伏發(fā)電功率預(yù)測模型
李蕓，李萍，麻利新
（寧夏大學 物理與電子電氣工程學院，寧夏 銀川 750021）
摘 要：光伏發(fā)電功率對光伏發(fā)電的可靠性起著決定性作用。針對Elman神經(jīng)網(wǎng)絡(luò)收斂速度慢、訓練時間較長的問題，利用果蠅算法（FOA）來優(yōu)化Elman神經(jīng)網(wǎng)絡(luò)的權(quán)值和閾值，從而提高運行效率。建立了基于FOA-Elman神經(jīng)網(wǎng)絡(luò)的光伏發(fā)電功率預(yù)測模型，并給出了算法設(shè)計及編碼方案。仿真實驗結(jié)果表明，F(xiàn)OA-Elman模型預(yù)測精度比傳統(tǒng)Elman神經(jīng)網(wǎng)絡(luò)模型預(yù)測精度高，更適合于光伏發(fā)電功率預(yù)測。
關(guān)鍵詞：光伏發(fā)電；功率預(yù)測；果蠅算法；Elman 神經(jīng)網(wǎng)絡(luò)；預(yù)測精度
中圖分類號：TM615 文獻標識碼：A 文章編號：1007-3175(2019)12-0001-04
Prediction Model of Photovoltaic Power Generation Based on FOA-Elman Neural Network
LI Yun, LI Ping, MA Li-xin
（School of Physics and Electronic-Electrical Engineering, Ningxia University, Yinchuan 750021, China）
Abstract: The photovoltaic power generation plays an important role in the reliability of photovoltaic power system.Aiming at the slow convergence speed and long training time of Elman neural network, this paper used the fruit fly optimization algorithm (FOA) to optimize the weights and thresholds of Elman neural network to improve the operation efficiency. A photovoltaic power prediction model based on FOAElman neural network was established, and the algorithm, design and coding scheme were given. The simulation results show that the prediction accuracy of FOA-Elman model is higher than that of traditional Elman neural network model, more suitable for the photovoltaic power prediction.
Key words: photovoltaic power generation; power prediction; fruit fly optimization algorithm; Elman neural network; prediction accuracy
參考文獻
[1] 段然. 光伏發(fā)電儲能系統(tǒng)研究[D]. 徐州：中國礦業(yè)大學，2016.
[2] 程啟明，李明，陳根，等. 小型離網(wǎng)光伏發(fā)電系統(tǒng)的設(shè)計[J]. 上海電力學院學報，2014，30(3)：199-202.
[3] 張星. 并網(wǎng)光伏發(fā)電系統(tǒng)建模及其參數(shù)辨識技術(shù)研究[D]. 合肥：合肥工業(yè)大學，2018.
[4] 王可舜. 關(guān)于風力發(fā)電與光伏發(fā)電并網(wǎng)問題的探討[J]. 自動化應(yīng)用，2016(5)：75-76.
[5] 楊育剛. 基于光伏發(fā)電與負荷預(yù)測的微電網(wǎng)儲能系統(tǒng)控制策略研究[D]. 保定：華北電力大學，2015.
[6] MORENO-MUNOZ A, ROSA J J, LOPEZ M A,et al. Grid interconnection of renewable energy sources: Spanish legislation[J].Energy for Sustainable Development，2010，14(2)：104-109.
[7] BENTH F E, IBRAHIM N A.Stochastic modeling of photovoltaic power generation and electricity prices[J].Journal of Energy Markets，2017，10(3)：1-33.
[8] 周同旭，周松林. 光伏發(fā)電功率區(qū)間概率預(yù)測[J].銅陵學院學報，2017(2)：108-110.
[9] 楊書佺. 果蠅算法在電力通信光纖網(wǎng)絡(luò)恢復優(yōu)化中的應(yīng)用[J]. 信息與電腦，2017(22)：65-67.
[10] 徐杏芳. 基于果蠅算法優(yōu)化的BP 神經(jīng)網(wǎng)絡(luò)[J]. 福建電腦，2017，33(6)：27-28.
