A Hybrid Forecasting Model for Wind Energy based on the Complementary Ensemble Empirical Mode Decomposition and Whale Optimized Back Propagation Neural Network

Renewable energy is drawing increasing attention to economic development. Accurate wind energy prediction has a significant impact on the rational dispatch and safe operation of the power system. However, the fluctuation and randomness characteristics of wind speed and direction lead to the problem of poor forecasting accuracy. This paper proposes a novel forecasting method of wind speed and wind direction based on complementary ensemble empirical mode decomposition (CEEMD) and whale optimized algorithm back propagation neural network (WOA-BPNN) is proposed. First, the original wind speed and direction data series are decomposed into some relatively stable components by the CEEMD algorithm. Then we use the WOA-BPNN method to predict each component respectively. Finally, all the single prediction results are combined to generate the final prediction result. As a result, the experimental analysis shows that the hybrid forecasting model has higher prediction accuracy in ultra-short-term (15 minutes) wind speed and wind direction.