anthonylefranc/helios

Helios: Sunspot and Solar Radiation Prediction

Introduction

Helios is a final Bootcamp's project aimed at predicting solar activity, specifically focusing on the number of sunspots (SSN) and the solar radio flux (F10.7, measured in SFU) over a 10-year period. Utilizing advanced time series forecasting techniques, including Meta's Prophet model and the Seasonal Autoregressive Integrated Moving Average (SARIMA) model, Helios provides accurate and valuable insights into solar behavior. This project leverages the comprehensive dataset provided by the National Oceanic and Atmospheric Administration (NOAA). Features

Sunspot Prediction (SSN): Uses Prophet and SARIMA models to forecast the number of sunspots, providing insights into solar magnetic activity. Solar Radiation Forecast (F10.7): Another instance of the Prophet model predicts the solar radio flux, crucial for understanding solar emissions. 10-Year Forecasting: Long-term predictions offering a decade of foresight into solar activities. NOAA Dataset Utilization: Employs authoritative and detailed solar activity data from NOAA.

Requirements

Python 3.x
Pandas
NumPy
Plotly
FB Prophet
Statsmodels (for SARIMA)

Installation

To set up Helios, follow these steps:

Clone the repository:

bash

git clone https://github.com/anthonylefranc/helios.git

Install the required Python packages:

pip install -r requirements.txt

Usage

To run the prediction models, execute the main notebooks :

• 10_years_ARIMA_model.ipynb
• 10_years_PROPHET_model.ipynb
• 10_years_f10_prophet.ipynb

Source dataset

The dataset used in Helios is sourced from NOAA's Solar Data Services, encompassing historical records of sunspot numbers and solar flux. Contributing

Acknowledgments

-National Oceanic and Atmospheric Administration (NOAA) for providing the solar activity dataset.

• Meta Platforms, Inc. for the Prophet forecasting model.
• The Statsmodels community for the SARIMA model implementation.