High-Level Project Summary
We developed a supervised machine learning algorithm that uses NASA's given data sets as an input , we believe that if we can predict the next solar event so we could take our precaution and get ready to save the earth .Here goes the idea: our solution tracks the speed of solar wind and predicts if there any solar event will be coming .the journey will not be ended here , public awareness should be spread .from our side we made a user interface of website that simulates our idea,
Link to Final Project
Link to Project "Demo"
Detailed Project Description
Our project has 2 points of view, The first one is the Machine Learning Algorithm platform, This one is only between the DSCOVR spacecraft and the agency o Earth, The Model takes the historical data sets and live records from the spacecraft sensors the uses these data to predict the next solar event then send a warning to the agency to take precautions and to avoid financial losses.
The second section in our project is a User Interface of a website to increase awareness and spread the knowledge of space weather in terms of articles, history of solar events and their effects on Earth, provide info about NASA'S spacecrafts like DSCOVR spacecraft.
We have a future plans to provide these info to children in an easy way and colorful UI.
Space Agency Data
We used mainly the Wind ion's parameters data set provided by NASA in the challenge resources, We thought about the best way to use the data and we decided to take slices of intervals of time of the data that have a known event eg: 2012 solar storm, 2022 February geomagnetic storm. we took days before and after the event to allow the model to train at different situations.
NASA can put more data into the model in the future.
Hackathon Journey
At the first day we started our day by registration and we took our lovely gifts from NASA space apps Cairo, then we went to the mosque to do the jumaa prayer, After that we started our work by taking time to make a plan, we decided to preprocess the data first as it is very important part in the challenge, so we worked in parallel trying to read the data as it was a CDF format not CSV, We converted successfully the formats from CDF to CSV .
Then started merging the data together and cleaning it by using pandas and Numpy and with assistance of Microsoft Excel.
During the day The organizers gave us our dinner meal, it was so delicious.
then our Machine Learning journey begins.
we went to take some rest and after that we made our demo video.
References
Microsoft Excel
Wind Ion's Parameters Dataset
Jupyter notebook
Google colabs
CDF tool from NASA website
Tags
#Sun #solar_wind #solar_storm #solar_flare #machine_learning #Artificial Intelligence #ML_models #Awareness #people #software