E.N.I.A.C

https://github.com/gabriel-ac6/FinalCodes

https://youtube.com/shorts/vss8nLoUScU?feature=share

Background of the Carrington Event:

• It happened in 1859

• It was a geomagnetic solar storm that happened during the peak of the solar cycle

• The intensity was so high that cases have been reported where the storm generated an electric current that was carried by the telegraph transmission wires and managed to run the machines - some even caught fire

• A storm of this magnitude, in the current world context, would cause serious problems, as it would compromise the operation of electromagnetic devices, on which society is extremely dependent

Features:

• Predict with considerable accuracy up to a maximum of 48h

• Catalog the possible threats, note the data from CME's and flares obtained by NASA

• Provide warnings of threats that have been catalogued (that will occur)

• Make the data available for anyone to use (for those who don't pay)

• Early warnings for VIP customers (who pay for use)

• Estimated time for interference or damage to reach targets (for VIP customers)

• Website for viewing real-time data and predictions regarding CME's

The benefits:

• Allows an action plan for the protection of electronic components to be carried out in advance, avoiding major disruption and damage - both material and human.

• Reduction of the complexity of the theoretical mathematical models used for the correction of the anomalies

• Reduction of the operating cost of data collection, because it eliminates the need for great efforts for the calibration and maintenance of the probes responsible for the collection

• A system of low cost of execution that does not require labor to maintain it and the server, because besides being an autonomous neural network, all other services such as database and hosting are outsourced

• High precision for prevention of possible instabilities in communication and broadband connections

• High accuracy to prevent possible damage to physical structures: either satellites or ground-based devices

Project Value Proposition:

Replace the current prediction model flow that relies on theoretical models that attempt to compensate, unsatisfactorily, for the presence of noise present in the data used to create it.

The new flow would count on a neural network model that would use the existing and concrete historical data, which cannot suffer changes in values and dates, to treat in a more adequate way how each of the CME's (that generate the interference problems). That is, how they occur and when they will occur, taking into consideration another method of how they will be consumed by the prediction model, thus a better chance of increasing accuracy and precision.

And subsequently create an interface that allows anyone to have access to this data and can use it or in their lives whoever they are.

Technical Specification:

For the back-end integration the database chosen was MySQL, the MySQL database is essential to perform the database management in a table, which makes the interconnection between all data and builds a much more practical and agile data administrative system, ensuring that all information is well organized and secure.

The software was built in python language 3.8.0, using the DONKI API to obtain the information for machine learning and a document made available by Stanford containing a robust database to complement the first, the prediction model was a dense neural network (with more than one neuron) and using the sklearn library in communion with others such as sun-py (for studies focused on solar activity)

As the software is mostly robust lines of code with no possibility to visualize the data in a comfortable way, keeping an interested person out of the programming area would be relatively complicated, so we developed a website that allows the integration of the captured data through the python document with a nice interface developed in js, Html5, php, react and CSS3.

External technologies:

We use some external technologies, not in the development of the project, but for later use, some of these technologies are:

-AWS

-Google Cloud

-Azure

Database:

• API DONKI (CME Analises e Flare)

• Stanford Document

• sun-py database

Challenges we faced:

The biggest challenge I faced in this hackathon was precisely that I had never worked with react, I had only used the Laravel framework, but I had never used js as a back-end and I had never in my life heard of js. And while the mentors' assistance was good, I wanted to develop this project without getting tips just by myself without using much of their advice to improve development performance, which ended up making it a bit difficult as well.

What we learned:

I learned how to use new technologies like js more broadly, react, and python, all languages I thought I could never learn in my life, but learned that it is possible and consequently learned how to use them. I also learned to use my university's resources in a more versatile way and mostly to listen more to the advice of my mentors.

Why we chose this project:

We chose this challenge precisely because of the level of difficulty that this idea provided us and at first could provide us, outside that many people have no idea of the importance of this theme, the CME's are responsible for causing numerous damages in various devices and structures, which every day provides several challenges for various areas of society and consequently for several companies, And we cannot forget how much these loads of energy / radiation can provide in a chaotic scenario, there is speculation that a giant CME's would stop all devices on Earth for 8s, thinking about the chaos that this could be we decided to develop a software to give a chance to human beings to face the worst of existing scenarios.

Material de referência:

• [Página do desafio](https://2022.spaceappschallenge.org/challenges/2022-challenges/carrington-event/details)

• Vídeo do desafio

• [https://youtu.be/m_pDSJive-E](https://youtu.be/m_pDSJive-E)

• Um pouco sobre a [metodologia utilizada na calibração e análise de dados](https://www.spaceappschallenge.org/space-apps-challenge-2022-example-resource-save-the-earth-from-another-carrington-event/) da telemetria da sonda

• Barnes, W. T.; Bradshaw, S. J.; Viall, N. M.

#Machine Learning, #SUN,#NASA,#CME,#Carrington,#Ciencias de dados,#Future