sciencetech-blog

comments

Artificial intelligence is helping astronomers predict deadly solar flares that have the potential to cause havoc on Earth.

US researchers say their super computer can provide advance warning of solar eruptions, which can release energy equivalent to 100 billion atomic bombs.

The flares arise from twisted magnetic fields that occur all over the sun's surface, and they increase in frequency every 11 years - a cycle that is now at its maximum.

Scroll down for video 

Artificial intelligence is helping astronomers predict deadly solar flares that have the potential to cause havoc on Earth. This image released by Nasa shows the sun emitting a mid-level solar flare, to the right of the sun

On Earth, they can cause widespread power outages and severely damaging critical infrastructure. 

Using artificial intelligence, Stanford University solar physicists Monica Bobra and Sebastien Couvidat have automated the study solar flares.

They looked at the largest ever set of solar observations at the Solar Dynamics Observatory (SDO) to find patterns in something known as 'vector magnetic lines'.

These describe the strength and direction of magnetic fields on the solar surface.

'Machine learning is a sophisticated way to analyse a ton of data and classify it into different groups,' Professor Bobra said.

Machine learning software attaches information to a set of established categories.

The software looks for patterns and tries to see which information is relevant for predicting a particular category.

'We had never worked with the machine learning algorithm before, but after we took the course we thought it would be a good idea to apply it to solar flare forecasting,' Professor Couvidat said.

The researchers used the AI machine to study the features of the two strongest classes of solar flares, M and X.

Solar Flares: See the sun as you've never seen it before

Though others have used machine learning algorithms to predict solar flares, nobody has done it with such a large set of data and or with vector magnetic field observations.

M-class flares can cause minor radiation storms that might endanger astronauts and cause brief radio blackouts at Earth's poles. X-class flares are the most powerful.

The researchers catalogued flaring and non-flaring regions from a database of more than 2,000 active regions and then separated those regions by 25 features such as energy, current and field gradient.

They then fed the learning machine 70 per cent of the data, to train it to identify relevant features.

Machine learning confirmed that the energy stored in the magnetic field is very relevant to predicting solar flares.

Using just a few of the 25 features, the super computer discriminated between active regions that would flare and those that would remain calm.

The next step in solar flare prediction would be to incorporate data from the sun's atmosphere, Professor Bobra said.

'It's exciting because we not only have a ton of data, but the images are just so beautiful,' she said. 'And it's truly universal. Creatures from a different galaxy could be learning these same principles.'

Nasa's Solar Dynamics Observatory (SDO), illustration shown, was launched on 11 February 2010. Researchers looked at the largest ever set of solar observations at the Solar Dynamics Observatory to find patterns in something known as 'vector magnetic lines'