CEEMD-WT和CNN在短期风速预测中的应用研究

由于风速存在随机性和不稳定性，为了提高短期风速预测的精度，提出了一种基于完备总体经验模态分解（CEEMD）、小波变换（WT）和卷积神经网络（CNN）的短期风速预测混合模型。首先，CEEMD算法把原始风速序列分解成一些相对平稳的固有模态函数和一个残差序列；然后，WT算法对每个固有模态函数进行二次去噪，进一步消除噪声对固有模态函数的影响；最后，卷积神经网络对每个固有模态函数、残差序列和影响风速的5个属性训练预测得到各自的预测结果，对所有的预测结果重构得到最终的预测结果。通过实验与其他4个风速预测模型进行比较，所提出的模型预测的绝对平均百分比误差（MAPE）最小，为2.484%，表明在短期风速预测方面CEEMD-WT-CNN模型有较好的性能。

Application research on complete ensemble empirical mode decomposition,wavelet transform and convolutional neural networks in short-term wind speed forecasting

Because there are randomness and uncertainty in wind speed, this paper proposes a hybrid model of Complete Ensemble Empirical Mode Decomposition（CEEMD）, Wavelet Transform（WT） and Convolutional Neural Networks（CNN） to improve forecasting accuracy. Firstly, CEEMD decomposes original wind speed into some relatively stable intrinsic mode functions and a residual sequence. Then, WT makes secondary noise elimination to eliminate effects of noise on each intrinsic mode function. Finally, the final result is obtained by refactoring forecasting results that CNN trains each intrinsic mode function, residual sequence and five attribute to obtain respectively. Compared with other four wind speed forecasting model, the Mean Absolute Percentage Error（MAPE） is 2.484% in the proposed model. This indicates that model of CEEMD-WT-CNN exists better performance in terms of short-term wind speed forecasting.