摘要
针对风速序列具有很强的随机性和波动性,提出一种基于完备总体经验模态分解(CEEMD)、长短期记忆网络(LSTM)和自回归差分移动平均(ARIMA)的组合预测模型来对短期风速进行准确预测。首先利用CEEMD算法将原始风速序列分解为多个模态分量,降低风速序列的复杂度;然后通过排列熵(PE)把风速子模态分为高频序列和低频序列,对高频序列和低频序列分别建立LSTM和ARIMA预测模型;最后把子序列预测结果叠加起来,得到最终的风速预测值。实验结果表明,该预测模型的平均绝对百分比误差(MAPE)和均方根误差(RMSE)分别为8.68%、0.389 m/s,与其他预测模型的预测结果相比,预测精度有了一定提高。
In view of the strong randomness and volatility of wind speed series,a combined forecasting model based on complete ensemble empirical mode decomposition(CEEMD),long short-term memory network(LSTM)and autoregressive integrated moving average(ARIMA)is proposed to accurately predict short-term wind speed.First,the CEEMD algorithm is used to decompose the original wind speed sequence into multiple modal components to reduce the complexity of the wind speed sequence.Then,through permutation entropy(PE),the wind velocity sub-modes are divided into high-frequency and low-frequency sequences,and LSTM and ARIMA prediction models are established for the high-frequency and low-frequency sequences respectively.Finally,the sub-sequence prediction results are superimposed to obtain the final wind speed prediction value.The experimental results show that the mean absolute percentage error(MAPE)and root mean square error(RMSE)of the prediction model are 8.68%and 0.389 m/s,respectively.Compared with the prediction results of other prediction models,the prediction accuracy has been improved to a certain extent.
作者
李秉晨
于惠钧
丁华轩
刘靖宇
LI Bingchen;YU Huijun;DING Huaxuan;LIU Jingyu(College of Railway Transportation,Hunan University of Technology,Zhuzhou 412007,China)
出处
《中国测试》
CAS
北大核心
2022年第2期163-168,共6页
China Measurement & Test
关键词
风速预测
完备总体经验模态分解
长短期记忆网络
自回归差分移动平均
wind speed prediction
complete ensemble empirical mode decomposition
long short-term memory
autoregressive integrated moving average