基于特征选择谱聚类和优化支持向量机的短期风速预测

被引:10
作者
张国维
王丙乾
机构
[1] 华北电力大学经济与管理学院
关键词
随机森林; 特征选择; 谱聚类; 支持向量机; 短期风速预测;
D O I
暂无
中图分类号
TM614 [风能发电];
学科分类号
080811 [新能源发电与电能存储];
摘要
提出一种基于随机森林算法、谱聚类算法和支持向量机的短期风速预测方法。首先,利用小波变换对原始风速进行去噪,剔除原始风速中不规则波动信息。然后利用随机森林算法进行特征选择,选择出最优的特征输入。再利用谱聚类算法对特征输入进行聚类分析,得出各个训练样本的聚类标签,提高模型训练样本的有效性;利用支持向量机对各个聚类标签分别进行建模,并使用遗传算法优化支持向量机的参数,提高模型的泛化能力。最后确定预测点的聚类标签,并使用相应标签的预测模型得到最终的预测结果。以某风电场的实际数据研究表明,所提出模型在短期风速预测中具有较高的精度。
引用
收藏
页码:9 / 14
页数:6
相关论文
共 25 条
[1]
Wind speed prediction using the hybrid model of wavelet decomposition and artificial bee colony algorithm-based relevance vector machine.[J].Sheng-wei Fei;Yong He.International Journal of Electrical Power and Energy Systems.2015,
[2]
Wind direction forecasting with artificial neural networks and support vector machines.[J].F. Tagliaferri;I.M. Viola;R.G.J. Flay.Ocean Engineering.2015,
[3]
Wind speed prediction using reduced support vector machines with feature selection.[J].Xiaobing Kong;Xiangjie Liu;Ruifeng Shi;Kwang Y. Lee.Neurocomputing.2015,
[4]
Short-term wind speed forecasting based on spectral clustering and optimised echo state networks.[J].Da Liu;Jilong Wang;Hui Wang.Renewable Energy.2015,
[5]
Feature selection in wind speed prediction systems based on a hybrid coral reefs optimization – Extreme learning machine approach.[J].S. Salcedo-Sanz;A. Pastor-Sánchez;L. Prieto;A. Blanco-Aguilera;R. García-Herrera.Energy Conversion and Management.2014,
[6]
Noise model based ν -support vector regression with its application to short-term wind speed forecasting.[J].Qinghua Hu;Shiguang Zhang;Zongxia Xie;Jusheng Mi;Jie Wan.Neural Networks.2014,
[7]
Short-term wind speed forecasting using wavelet transform and support vector machines optimized by genetic algorithm.[J].Da Liu;Dongxiao Niu;Hui Wang;Leilei Fan.Renewable Energy.2014,
[8]
Short-term prediction of wind power with a clustering approach [J].
Kusiak, Andrew ;
Li, Wenyan .
RENEWABLE ENERGY, 2010, 35 (10) :2362-2369
[9]
一种新的风电场风速时间序列建模及超短期预测方法 [J].
李卫 ;
席林 .
电网与清洁能源, 2015, 31 (09) :78-82
[10]
基于小波分解和极限学习机的短期风速组合预测研究 [J].
汪小明 ;
肖猛 ;
杨楠 ;
尹笋 .
可再生能源, 2015, (08) :1159-1163