Solar Cycle 25 Prediction Using an Optimized Long Short-Term Memory Mode with F10.7

Abstract

In this paper, an optimized long short-term memory (LSTM) model is proposed to deal with the smoothed monthly $F_{10.7}$ F 10.7 data, aiming to predict the peak amplitude of $F_{10.7}$ F 10.7 and the occurring time for Solar Cycle 25 (SC-25) to obtain the maximum amplitude of sunspot number (SSN) and the reaching time. The “re-prediction” process in the model uses the latest prediction results obtained from the previous prediction as the input for the next prediction calculation. The prediction errors between the predicted and observed peak amplitude of $F_{10.7}$ F 10.7 for SC-23 and SC-24 are 2.87% and 1.09%, respectively. The predicted peak amplitude of $F_{10.7}$ F 10.7 for SC-25 is 156.3, and the maximum value of SSN is calculated as 147.9, which implies that SC-25 will be stronger than SC-24. SC-25 will reach its peak in July 2025.

Performance evaluation of IRI-Plas 2017 model with ionosonde data measurements of ionospheric parameters

“…As reported by Okoh and Okoro [ 49 ], the solar activity indexes of SSN and F10.7 are well corrected with a correlation coefficient of 0.95 between the periods December 1963 and January 2019, representing solar cycles 20 to 24. The conclusion of this study shows that there is a remarkable difference between them during different solar cycles and solar activity phases, but the result shows that there is a better fit between SSN and F10.7 in high solar activity than low solar activity years [ 7 , 50 ]. Due to this reason, in this work, we chose the SSN option.…”

Interplanetary shock data base

“…Observations show that SC23 was significantly stronger than SC24 based on the overall sunspot numbers, which is reflected on the total number of shocks observed in the corresponding periods (432 and 187, respectively). Zhu et al (2022) predicted that SC25 will be stronger than SC24 and will reach its maximum value around July 2025. Therefore, according to these predictions, it is reasonable to expect that SC25 will have a similar number of shocks with respect to SC24.…”

Forecasting solar energetic proton integral fluxes with bi-directional long short-term memory neural networks

“…In time series forecasting, it is a common practice to take a continuous set of data points from the main dataset to be the validation set and another smaller chunk of data to be the test set, for instance in Pala & Atici (2019), Benson et al (2020), Zhang et al (2022), andZhu et al (2022). From our experiments, we got decent results when we applied the same data split method, but the results were a bit biased toward the end of the solar cycle 24 and the testing set was biased towards a quiet period.…”