Short-term local prediction of wind speed and wind power based on singular spectrum analysis and locality-sensitive hashing

摘要: With the growing penetration of wind power in systems, more accurate prediction speed and is required for real-time scheduling operation. In this paper, a novel forecast model short-term proposed, which based on singular spectrum analysis (SSA) locality-sensitive hashing (LSH). To deal with impact high volatility original time series, SSA applied to decompose it into two components: mean trend, represents tendency fluctuation component, reveals stochastic characteristics. Both components are reconstructed phase space obtain trend segments component segments. After that, LSH utilized select similar segments, then employed local forecasting, so that accuracy efficiency can be enhanced. Finally, support vector regression adopted prediction, where training input synthesis corresponding Simulation studies conducted series from four databases, final results demonstrate proposed stable comparison other models.