HELIO SURFERS

WE DISCLOSE RADIATIONS BEFORE IT AFFLICTS OUR EARTH

P R NIRANJANA BALAMUGESH M MISTRY MARGI RESHMA K R

SAVE THE EARTH FROM ANOTHER CARRINGTON EVENT

​

OUR OBJECTIVE AND APPROACH

• The objective of our project is to create a machine learning algorithm to detect any upcoming solar storms due to intense coronal mass ejection and thus produce signs of warning to save the Earth from another Carrington event.
• Our approach to this challenge involves the fact that a strong solar storm like the Carrington event.
• If occurred today, would cost anywhere around $0.6 to $2.6 trillion worth loss and around months and years for recovery from the internet apocalypse caused due to the electromagnetic phenomena.
• We aim to detect any possible solar flares at least 45 mins prior to the event with maximum precision.

​

CARRINGTON EVENT:

• The Carrington Event was the most intense geomagnetic storm in recorded history, with peak values from 1 to 2 September 1859 during solar cycle 10, just a few months before the solar maximum of 1860.
• The outer solar atmosphere, the corona, is structured by strong magnetic fields. When a CME arrives at Earth it buffets the magnetosphere. If the arriving solar magnetic field is directed southward, it interacts strongly with the oppositely oriented magnetic field of the Earth.
• At the Earth's surface , it causes a rapid drop in the magnetic field strength, and this is called the solar storm. It usually lasts about 6 to 12 hours, after which the magnetic field gradually recovers over a period of several days.

​

DSCOVR- FARADAY CUP

• Sunspot activity rises and falls on an 11-year cycle, and we're currently approaching the next solar maximum in 2025. So, there is a high probability of an arbitrary range solar storm occurring anywhere between 2024- 2025.

• Scientists at the National Oceanic and Atmospheric Administration (NOAA)'s Space Weather Prediction Centre analyze sunspot regions daily to assess the threats. Solar System Exploration.

​

​

• DSCOVR (Deep Space Climate Observatory), a joint mission between NASA, NOAA, and the USAF, monitors changes in the solar wind, providing space weather alerts and forecasts for geomagnetic storms that could disrupt power grids, satellites, telecommunications, aviation and GPS. DSCOVR was launched on Feb. 11, 2015, and orbits about a million miles from Earth at Sun-Earth Lagrange point

​

• The Faraday Cup of DSCOVR carries Plasma-Magnetometer (PlasMag), Earth Polychromatic Imaging Camera (EPIC) and National Institute of Standards and Technology Advanced Radiometer (NISTAR) to provide real time solar wind observations for solar forecasts.

​

01

DATA COLLECTION:

• DSCOVR- FARADAY CUP DATA FROM

NOAA, SWPC,NGDC.

. IMF, SOLAR WIND DATA,SYM-H INDEX

TRACKING SOLAR STORM:

. PREDICTING CARRINGTON EVENT

. SOLAR STORM MODEL

ML ALGORITHM:

. ANN

.

​

​

EXPLORATORY DATA ANALYSIS

• USING NUMPY, PANDAS, SKLEARN
• DATA AVERAGED TO 5 INTERVALS

​

MODEL REGISTRY

​

03

04

05

02

APPROACH

VARIATION IN MAGNETIC PARAMETERS AND SOLAR WIND DATA

MODEL ANALYSIS

LINEAR REGRESSION, RANDOM FOREST

MACHINE LEARNING MODEL

• ARIMA model (Autoregressive Integrated Moving Average) is used which combines three methods including autoregressive, moving average and integration.

​

• We also used logistic regression and TensorFlow(for data normalization using max scalar) for supervised prediction.

​

• Using sklearn library we had split the dataset to training, testing and validation and created a model which is giving an accuracy > 80%. for prediction with limited dataset.

​

• For live prediction we planned to collect the real-time data using web scraping tools like selenium, beautiful soup and pandas and feed those values to the pretrained model that we have generated using the above-mentioned datasets. So, if any solar storm occurs, we can predict that 45 min prior the coronal mass radiations traverse the Bx GSM of earth and can generate a warning or an alert to different authorities from the model.

​

• By utilizing more dataset if we can forecast the changes for the above parameters like before 10 days the warning system can be improved which can reduce the impact of the event.

FUTURE WORK

DSCOVR DATA ANALYSIS

• For user given inputs for the above-mentioned parameters a UI with React and Flask API will collect the user input and using a GET request it will be given to the model and prediction is made and it will display the graph for the parameters Bt, Bx GSM, By GSM, Bz GSM, velocity, temperature and density against time along with the predicted value whether event will occur or not. If will occur an alert to the user.

• Due to the deficient time duration, we were not able to completely format the desired results with our output. We are aiming to create an exclusive algorithm with the capacity to model the network to provide warning signs for solar storms at least 48 hours prior to occurrence.