A Short-Term Wind Speed Forecasting Model Based on EMD/CEEMD and ARIMA-SVM Algorithms

In order to ensure the driving safety of vehicles in windy environments, a wind monitoring and warning system is widely used, which speed prediction algorithm with better stability sufficient accuracy one key factors smooth operation system. this paper, novel short-term forecasting model, combining complementary ensemble empirical mode decomposition (CEEMD), auto-regressive integrated moving average (ARIMA), support vector machine (SVM) technology, proposed. Firstly, EMD CEEMD are used decompose measured sequence into finite number intrinsic functions (IMFs) decomposed residual. Each IMF subseries has linear characteristics. The ARIMA adopted predict each subseries. Then, new reconstructed using sum predicted errors all high nonlinear features error modeled SVM, suitable process data. Finally, superposition results performed obtain final speed. To verify two typhoon datasets, from south coast China, test proposed methods. show that hybrid model predictive ability than single models other combined models. root mean squared (RMSEs) for three datasets 0.839, 0.529, 0.377, respectively. combination contributes most performance model. It feasible apply prediction.