Suzhou Electric Appliance Research Institute
期刊号: CN32-1800/TM| ISSN1007-3175

Article retrieval

文章检索

首页 >> 文章检索 >> 往年索引

基于交叉熵理论的光伏发电功率组合预测方法

来源:电工电气发布时间:2022-04-20 14:20 浏览次数:345

基于交叉熵理论的光伏发电功率组合预测方法

陈丽霞
(国网福建省电力有限公司福州供电公司,福建 福州 350007)
 
    摘 要:光电功率预测对电网的安全稳定运行以及调度等方面具有重要意义。针对单一预测方法精度较低的问题,提出了一种基于交叉熵理论的光伏发电功率组合预测方法,以单一预测方法和最小化组合预测方法的“差值”为依据,动态地改变不同预测方法的权重,提高组合预测的精度。以某光伏电场为算例进行分析,结果表明,该模型针对不同的天气,具有较强的预测适应性,可以提高预测精度,减少预测误差的出现。
    关键词:光伏发电;功率预测;交叉熵;权重;组合预测
    中图分类号:TM615     文献标识码:A     文章编号:1007-3175(2022)04-0017-04
 
A Combination Forecasting Method of Photovoltaic Power Generation
Based on Cross-Entropy Theory
 
CHEN Li-xia
(Fuzhou Power Supply Company, State Grid Fujian Electric Power Co., Ltd, Fuzhou 350007, China)
 
    Abstract: Photoelectric power prediction has significant meaning to the safe and stable operation and dispatching of power grids. This paper proposed a combined prediction method of photovoltaic power generation based on cross-entropy theory to solve the problem of the low accuracy of the single prediction method. This combined prediction method is based on the“gap”between the single forecast method and the minimized combined forecast method. It changed the weight of different forecasting methods dynamically and improved the accuracy of combined forecasting. This paper took a photovoltaic field as an example to analyze. The result shows that the model has strong prediction adaptability, and it could improve prediction accuracy and reduce prediction errors.
    Key words: photovoltaic power generation; power prediction; cross-entropy; weight; combined prediction
 
参考文献
[1] 李旭. 基于典型日出力特性分析的光伏电站功率预测研究[D]. 北京:华北电力大学,2016.
[2] BOYLE Godfrey.Renewable electricity and thegrid[M].London :Earthscan Publications Ltd,2007.
[3] 龚莺飞,鲁宗相,乔颖,王强. 光伏功率预测技术[J]. 电力系统自动化,2016,40(4) :140-151.
[4] 朱永强,田军. 最小二乘支持向量机在光伏功率预测中的应用[J]. 电网技术,2011,35(7) :54-59.
[5] FERNANDEZ-JIMENEZ L A, TERREROS-OLARTE S,ZORZANO-SANTAMARIA P J, MENDOZA-VILLENA M,GARCIA-GARRIDO E.Probabilistic photovoltaic power forecasting model based on deterministic forecasts[C]//E3S Web of Conferences,2020:152.
[6] 丁明,徐宁舟. 基于马尔可夫链的光伏发电系统输出功率短期预测方法[J] . 电网技术,2011,35(1) :152-157.
[7] 杨锡运,刘欢,张彬,陈嵩. 基于熵权法的光伏输出功率组合预测模型[J] . 太阳能学报,2014,35(5) :744-749.
[8] 杨锡运,任杰,肖运启. 基于粗糙集的光伏输出功率组合预测模型[J] . 中国电力,2016,49(12) :133-138.
[9] 陈嵩. 组合预测技术及其在功率预测中的应用[D].北京:华北电力大学,2015.
[10] WANG Yue, GUO Chuangxin, WU Q H.A cross-entropy-based three-stage sequential importance sampling for composite power system short-term reliability evaluation[J].IEEE Transactions on Power Systems,2013,28(4) :4254-4263.
[11] 王越,郭创新,文云峰,李树青,袁翔. 一种三段式序贯交叉熵重采样方法及其在电力系统短期可靠性评估中的应用[J] . 中国电机工程学报,2013,33(28) :94-100.
[12] 周军妮,杨润玲,王燕妮,江莉. 一种结合交叉熵和投影特征的图像匹配算法[J] . 小型微型计算机系统,2013,34(2) :405-408.
[13] 陈宁,沙倩,汤奕,朱凌志. 基于交叉熵理论的风电功率组合预测方法[J] . 中国电机工程学报,2012,32(4) :29-34.
[14] 高阳,张碧玲,毛京丽,刘勇. 基于机器学习的自适应光伏超短期出力预测模型[J] . 电网技术,2015,39(2) :307-311.