摘要
由于风速具备很好的时序性和随机性,因此选用时间序列的ARMA(p,q)模型对风速进行预测。为了进一步提高预测精度,先利用小波多分辨分解对全风速进行高低频分解,提取出低频平缓信号,然后对平缓后的数据进行时间序列预测。并以某风电场实测风速为例,实际算例表明:该组合模型可以有效提高风速预测精度,具有一定实用性。
Wind speed is of good succession and randomness, So this paper uses ARMA (p,q) model of times series to forecast the wind speed. For further improving the prediction precision, this paper first uses multi-resolution wavelet analysis to pick up the low frequency parts through the decomposition of the whole wind speed, then uses times series method to forecast the wind speed on the gentled data. This paper takes the wind speed directly measured from a certain wind farm as a case, which shows that this combination model can effectively improve the wind speed prediction accuracy.
出处
《陕西电力》
2011年第12期36-38,49,共4页
Shanxi Electric Power
基金
河北省科技支撑计划项目(10213901D)
关键词
小波分析
时间序列
ARMA模型
风速
预测
预测精度
wavelet analysis
times series method
ARMA model
wind speed
forecasting
prediction accuracy