基于粗糙集理论-主成分分析的 Elman 神经网络短期风速预测.pdf

文档介绍：第 42 卷第 11 期电力系统保护与控制
2014 年 6 月 1 日 Power System Protection and Control Jun. 1, 2014
基于粗糙集理论-主成分分析的 Elman 神经网络短期风速预测
尹东阳,盛义发,蒋明洁,李永胜,谢曲天
(南华大学电气工程学院,湖南衡阳 421001)
摘要:为了解决传统静态前馈神经网络(FNN)在短期风速预测中易陷入局部最优值及动态性能的不足,引入Elman动态神经网
络建立风速预测模型,采用主成分分析法(PCA)对原始风速数据进行特征提取以优化神经网络的输入,改进激励函数和网
络结构以寻求函数收敛速度和预测精度的最优解。针对Elman神经网络预测模型在风速波动的峰值处预测误差较大及预测精
度存在波动性,提出采用粗糙值理论对模型预测值进行修正与补偿,进一步提高预测精度。实验证明:所提出的方法能有效
提高预测精度,增强神经网络模型的泛化能力,具有较好的实用性。
关键词:风速预测;Elman 神经网络;主成分分析;粗糙集理论;预测值修正
Short-term wind speed forecasting using Elman work based on rough set theory
and ponents analysis
YIN Dong-yang, SHENG Yi-fa, JIANG Ming-jie, LI Yong-sheng, XIE Qu-tian
(College of Electrical Engineering, University of South China, Hengyang 421001, China)
Abstract: Because the traditional static feed forward works (FNN) are easy to fall into local optimum and lack of dynamic
performance, the wind speed prediction model using Elman work (ElmanNN) is established, the ponent
analysis (PCA) is used to extract the feature of wind speed data, which optimizes the inputs of ElmanNN. Furthermore, excitation
function and the structures work are improved to search for the optimum solution of function convergence rate and prediction
accuracy. To solve large error and prediction accuracy fluctuations of the ElmanNN model at the peak value of wind speed, the rough
set theory is proposed pe