The Investigation And Application Of Time Series Modelsbased On Ensemble Empirical Mode Decomposition

Wind energy, as a non-polluting, renewable resources, increasingly attracts people’s attention. However, because of its intermittency, wind power has a significant impact on grid reliability and the stability of power system. Therefore, accurate prediction of wind energy is very important. In the prediction of wind energy, wind speed is one of the most important parameters. In order to improve the accuracy of wind speed forecasting, this paper presents a new hybrid model which combines ensemble empirical mode decomposition and artificial intelligence model named SVM. Three wind field of China’s Gansu region are used to verify the model. The results show that, compared with other traditional models and the neural network not using the ensemble empirical mode decomposition, the proposed method can obtain better forecasting results in the prediction of wind speed.For electric power system, along with the electric power industry restructuration, the electricity price becomes an important signal to all market participants and the focus of commercial activities in electricity market. Because of the volatility and non-stationarity of price, the existing single model is difficult to obtain satisfactory forecasting precision; this paper presents empirical mode decomposition method based on aggregation, seasonal adjustment and ARMA combination model in order to obtain a better prediction performance. The results show that, compared with other traditional model and other forecasting methods without ensemble empirical mode decomposition, the proposed method can achieve better prediction results in electricity price forecasting...