Hurricane Forecasting Using by Parallel Calculations & Machine Learning

This research is devoted to determinate the causal relationship between the flow of particles that are coming from the Sun and emergence of the hurricanes Irma, Jose, and Katia. Five parameters i.e. characteristics of solar activity (Radio Flux 10.7, the flows of protons and electrons with maximum energy, speed of solar wind particles, and density of solar wind particles) were chosen as model input, while wind speed and air pressure of Irma, Jose, and Katia hurricanes were used as model output. Input data were sampled to six hours interval in order to adapt time interval to the observed data about hurricanes, in the period between 28 September and 21 December 2017. As a result of the preliminary analysis of 12 274 264 linear and 594 Neural Networks models using by parallel calculations, the six of them were choose as bests. The identified lags were the basis for refinement of models with the artificial neural networks. Multilayer perceptrons with back propagation have been chosen as common used artificial neural networks which are based on learning from real data. Comparison of the accuracy of both linear and artificial neural networks results confirmed the adequacy of these models. Sensitivity analysis has shown that Radio Flux 10.7 has the greatest impact on the wind speed of the hurricanes. Despite low sensitivity of pressure to change the parameters of solar wind, their strong fluctuations can cause a sharp decrease in pressure, and therefore the appearance of hurricanes.