[1]张颖超,成金杰,邓华,等.基于相似日和特征提取的短期风电功率预测[J].郑州大学学报(工学版),2020,41(05):44-49.[doi:10.13705/j.issn.1671-6833.2020.02.023]
　ZHANG Yingchao,CHENG Jinjie,DENG Hua,et al.Short-Term Wind Power Prediction Based on Similar Day and Feature Extraction[J].Journal of Zhengzhou University (Engineering Science),2020,41(05):44-49.[doi:10.13705/j.issn.1671-6833.2020.02.023]
点击复制

基于相似日和特征提取的短期风电功率预测()
分享到：

《郑州大学学报(工学版)》[ISSN:1671-6833/CN:41-1339/T]

卷:
41
期数:
2020年05期
页码:
44-49
栏目:
出版日期:
2020-10-01

文章信息/Info

Title:
Short-Term Wind Power Prediction Based on Similar Day and Feature Extraction
作者:
张颖超; 成金杰; 邓华; 宗阳; 章璇
南京信息工程大学自动化学院,江苏南京210044, 南京信息工程大学气象灾害预报预警与评估协同创新中心,江苏南京210044, 南京信息工程大学自动化学院,江苏南京210044

Author(s):
ZHANG Yingchao1; 2;  CHENG Jinjie1;  DENG Hua1; 2;  ZONG Yang1;  ZHANG Xuan1
1.School of Automation, Nanjing University of Information Science & Technology, Nanjing 210044, China; 2.Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing 210044, China
关键词:
Keywords:
discrete Fréchet distance;  similar day;  kernel entropy component analysis;  support vector machine;  power prediction
DOI:
10.13705/j.issn.1671-6833.2020.02.023
文献标志码:
A
摘要:
为提高短期风电功率预测精度,增强预测模型对特定天气状况的代表性和适应性,提出一种基于离散Fréchet距离与核熵成分分析(KECA)相结合的数据处理方法。通过引入离散Fréchet距离,建立匹配相似日的数学模型,提取与预测日相似的样本,使用KECA从多种气象要素中提取合适的非线性主.元作为支持向量机(SVM)模型的输入。实验结果表明:所提出的方法明显提高了预测精度并具有一定的适用性 。
Abstract:
In order to improve the short-term wind power prediction accuracy and enhance the representativeness and adaptability of the prediction model to specific weather conditions, a data processing method was proposed based on discrete Fréchet distance and kernel entropy component analysis (KECA).Using Fréchet distance, this paper established a mathematical model matching similar days, extracted samples similar to the prediction date, and then used KECA to extract suitable nonlinear principal elements from various meteorological elements as input of support vector machine (SVM) model. The simulation verification showed that the proposed method could significantly improve the prediction’s accuracy and has certain applicability.

参考文献/References:

[1] 王健, 严干贵, 宋薇,等. 风电功率预测技术综述[J]. 东北电力大学学报, 2011, 31(3):20-24.

[2] 李军, 卓泽赢. 基于经验小波变换和多核学习的风电功率短期预测[J]. 信息与控制, 2018, 47(4):437-447.
[3] 孙晓燕, 时良振, 徐瑞东,等. 基于区间样本和回声状态网络的风电功率不确定性预测[J]. 郑州大学学报(工学版), 2017, 38(1):56-62.
[4] 阎洁, 许成志, 刘永前,等. 基于风速云模型相似日的短期风电功率预测方法[J]. 电力系统自动化, 2018, 42(6):53-59.
[5] 孟洋洋, 卢继平, 孙华利,等. 基于相似日和人工神经网络的风电功率短期预测[J]. 电网技术, 2010, 34(12):163-167.
[6] 倪鹏, 孙富荣. 基于相似日和风速连续性的风电场短期功率预测[J]. 山东电力技术, 2016, 43(11):39-43.
[7] 张宜阳, 严欢. 基于分段分层相似日搜索和自适应脊波神经网络的风电功率多步预测[J]. 电网与清洁能源, 2015, 31(4):124-131.
[8] YESILBUDAK M. Clustering analysis of multidimensional wind speed data using K-means approach[C]//2016 IEEE International Conference on Renewable Energy Research and Applications (ICRERA).Birmingham, United Kingdom： IEEE, 2016: 961-965.
[9] 丁志勇, 杨苹, 杨曦,等. 基于连续时间段聚类的支持向量机风电功率预测方法[J]. 电力系统自动化, 2012, 36(14):131-135.
[10] 郑婷婷, 王海霞, 吕泉,等. 基于相似日选取样本的短期风电功率混沌预测[J]. 电网与清洁能源, 2013, 29(3):74-79.
[11] 郁琛, 薛禹胜, 文福拴,等. 按时序特征优化模型后在线选配的超短期风电预测[J]. 电力系统自动化, 2015, 39(8):5-11.
[12] GANGER D, ZHANG J S, VITTAL V. Statistical characterization of wind power ramps via extreme value analysis[J]. IEEE transactions on power systems, 2014, 29(6):3118-3119.
[13] 叶小岭, 陈浩, 郭晓杰,等. 基于风速升降特征的短期风电功率预测[J]. 电力系统保护与控制, 2016, 44(19):56-62.
[14] ALT H, GODAU M. Computing the Fréchet distance between two polygonal curves[J]. International journal of computational geometry & applications, 1995, 5(1/2):75-91.
[15] EITER T, MANNILA H. Computing discrete fréchet distance[R]. Viena: Information System Department, Technical University of Vienna， 1994.
[16] 李思同, 庄强, 金琳,等. 基于时频特征核熵成分分析的局部放电模式识别方法[J]. 高压电器, 2018，54(6):125-131.
[17] 彭怀午, 杨晓峰, 刘方锐. 基于SVM方法的风电场短期风速预测[J]. 电网与清洁能源, 2009, 25(7):48-52.
[18] CHANG C C, LIN C J. LIBSVM: a library for support vector machines[J]. ACM transactions on intelligent systems and technology, 2011, 2(3): 27.

更新日期/Last Update: 2020-10-23