Short-term wind power prediction based on GPR-BSO model

Wind power forecasting is a crucial part for the safe and stable operation of wind integration, which under influence different factors such as speed, direction, atmospheric pressure. These bring randomness volatility to makes it less predictable. While, there are very limited studies on describing uncertainty power. Therefore, providing additional information volatility, kernel-based Gaussian Process Regression (GPR) incorporating hyper-parameters intelligent optimization method proposed in this paper. Firstly, solution GPR formulated nonlinear with constraints. Then, an algorithm named Brain-storming (BSO) adopted obtain optimal GPR. Furthermore, performance examined short-term data. Most importantly, BSO can avoid at local optimum.