Short-term Wind Speed forecasting Based on Weighted Spatial Correlation and Improved GWO-GBRT algorithm

Abstract

Energy transformation and development is a major issue that the international community urgently needs to implement, many countries develop wind power industry energetically. However, the uncertainty of wind has been a huge challenge for wind power generation. In this paper, a spatial correlation wind speed forecasting model based on weighted grey correlation analysis model and improved gradient boosting regression tree algorithm is proposed to evaluate and forecast wind resources. The coefficient of variation method was employed to obtain the weight of spatial correlation wind speed information, and the wind speed at reference points selected based on the weighted grey relational degree were input into the gradient boosting regression tree (GBRT) to build a forecasting model. In addition, the GBRT parameters were optimized by the improved grey wolf optimization (IGWO) algorithm before the model was used to perform wind speed and direction forecasting. The experimental results show that the improved forecasting algorithm has high accuracy and real-time, thus having research and practical value.