说明：  利用新陈代谢灰色预测、样本自适应BP 神经网络和时间序列分析分别进行风电功率实时预测和日前预测，并采用熵值取权法确定组合权重，引入自控机制，构建反馈，提出组合预测法和基于时间序列的卡尔曼滤波法。研究结果表明，组合预测模型能减少各预测点较大误差的出现，而卡尔曼滤波能大幅消减原始序列的波动影响。
(Use of metabolic gray forecast, sample adaptive BP neural network and time sequence analysis respectively conducted wind electric power real-time forecast and before prediction, and using the entropy take the right method to determine a combination of the right weight, the introduction of self-control mechanism, to build feedback, proposed combination forecasting method and based on the time sequence Kalman filtering method. The research results show that the combination forecasting model can reduce the prediction point error appears, the Kalman filter can significantly abatement fluctuations of the original sequence.)