Short-term wind power prediction based on nutrosophic clustering and GA-ELM

Wind farm NWP data has regularity and difference, and making full use of the information contained in NWP data is the key to wind power prediction. A short-term forecasting method for wind power based on nutrosophic clustering and GA-ELM is proposed. Firstly, the wind farm NWP data is divided into several weather types by the Chinese wisdom clustering method, and then the GA-ELM model is established for different weather types. The Gaussian index method is used to classify the forecast data, and then the different types of forecast data are substituted into the corresponding model predictions. Taking a 14MW wind farm in Northeast China as an example, the experiment shows that the nutrosophic clustering method reduces the influence of boundary points and abnormal points on the clustering center. Compared with the traditional method, the method has higher precision and universality.