NARMAX identification for space weather prediction using polynomial radial basis functions

Solar storms can damage transformers, electrical networks, and satellites. In this paper, we use system identification methods to construct nonlinear time-series models that are used to predict solar wind conditions with a 27-day prediction horizon. To identify nonlinear time-series models, we use a set of basis functions to represent the nonlinear mapping. For these basis functions, we propose an alternative class of radial basis functions, which have fewer parameters that needs to be tuned by the user. Finally, we compare the predictions obtained using identified models with predictions obtained with existing models.