Inspiration

Coronal mass ejections can disrupt satellite communications and cause power outages. We thought it would be helpful and fun to solve this problem by predicting when they would occur. Solar flares are good predictors of these events, so we implemented an algorithm to classify the intensity of solar flares. Predicting these events saves money and preparation time.

What it does

Our algorithm reads in solar flare data and classifies them based on intensity. There are three classes total, and the highest class (M-class) has the highest chance of causing or predicting coronal mass ejections.

How we built it

We read multiple research papers about quantum classifiers. After reading a few papers, we implemented an algorithm on a real quantum computer in the cloud. We then used our implementation to predict solar flares.

Challenges we ran into

Due to the small size of the data set, the capabilities of quantum were unable to be fully explored as quantum computing thrives on very large data sets

Accomplishments that we're proud of

Our contribution to solar flare predictions is something that is novel and worthy of being proud of. We appreciate the amount of knowledge we gained from this experience and are excited to apply this to future challenges.

What we learned

We spent a lot of time researching quantum computing. We learned about qubits, quantum annealing, quantum tunneling, and various methods to implement these.

What's next for Using Quantum Computing to Classify and Detect Solar Flares

Applying the same principles, we can create and run an algorithm on a much larger data set, allowing us to truly exploit the benefits of Quantum computing.

Built With

Share this project:

Updates