A Novel Decomposition-Based Approach for Non-Stationary Hub-Height Wind Speed Modelling

Abstract

An accurate description of hub-height wind speed characteristics is indispensable to offshore wind resource exploitation. However, given the assumption of stationarity in wind speeds that is violated, the commonly used stationary statistical models are inappropriate, and a non-stationary frequency analysis is required. This paper proposed a novel decomposition-based approach for non-stationary hub-height wind speed modelling. A combination of signal decomposition method and recurrence quantification analysis was applied to separate the sampled time series into deterministic and stochastic components. Five signal decomposition methods for time series separation were evaluated under various represent non-stationary scenarios via a set of numerical experiments. The decomposed deterministic and stochastic components were respectively investigated by discrete functions and traditional frequency analysis, and the non-stationary model was established by aggregating the modelled two components. To examine the proposed approach, 120 m hub-height wind speed data at six positions distributed along China’s coastline were used. The results suggest that the non-stationarity in wind speeds is significant and varies with geographical locations, emphasizing the importance of the consideration of non-stationarity. More importantly, it is clearly confirmed that the non-stationary model has advantages over the commonly used stationary models, and the proposed decomposition-based method outperforms other non-stationary modelling approaches.

Keywords: Non-stationary modelling, hub-height wind speed, Signal decomposition method, Recurrence quantification analysis, Deterministic and stochastic components