Improved Long-Term Wind Energy Assessment Accounting for the Non-Stationarity of Wind Speed in Sea Area Along China's Coastline

Abstract

An accurate wind energy assessment is essential for the effective utilization of offshore wind resources. Conventionally, energy assessment factors are estimated based on parametric models under the assumption of wind speed stationarity. However, given the subjection of seasonal variability and long-term trends, the hypothesis is violated, especially in the context of climate change. In this paper, with the aim of improving the accuracy of a wind energy assessment, wind speeds were modelled using the decomposition-based approach to consider the non-stationarity, and then corresponding procedures for energy factor estimation were designed. Using 30-year ERA5 data, long-term wind resource characteristics across the sea area along China’s coastline were investigated. Results demonstrated that the established non-stationary models are superior to commonly used stationary models, allowing the estimation of resource characteristics at arbitrary time scales after a one-time parametrization of parametric distributions. Significant trends exist in wind energy resources especially over the last two decades, when the magnitudes reach 4 W/m2/year in the Taiwan Strait. Moreover, the neglect of non-stationarity can lead to an underestimation of about 30% of wind power density in southeast sea areas. Thus, an adequate consideration of the non-stationarity of wind speed data in future wind energy research is required.

Keywords: Offshore wind energy, wind speed, non-stationary model, China Sea, long-term variation