The most sun Complex mysteries could soon be resolved thanks to artificial intelligence. On August 20, IBM and NASA announced the launch of Surya, a foundation model for the sun. Having been trained on large sets of solar activity data, this AI tool aims to deepen the humanity of the solar weather and predict solar eruptions – the explosions of electromagnetic radiation emitted by our star which threaten the two astronauts in the infrastructures in orbit and communication on earth.
Surya was formed with nine years of data collected by the NASA solar dynamic observatory (SDO), an instrument that has orbited the sun since 2010, taking high resolution images every 12 seconds. The SDO captures sun observations at different electromagnetic wavelengths to estimate the temperature of the stars. Precise measurements of the Magnetic Field of the Sun are also needed – essential data to understand how energy moves through the star and to predict solar storms.
Historically, the interpretation of this large amount of various and complex data was a challenge for heliophysicists. To take up this challenge, IBM says that Surya developers have used SDO data to create a digital sun twin – a dynamic virtual replica of the star which is updated when new data is captured, and which can be manipulated and more easily studied.
The process began by unifying the different data formats fed in the model, which allows it to treat them in a coherent way. Then, a long -range vision transformer was used – AI architecture which allows a detailed analysis of images with very high resolution and the identification of relationships between their components, whatever their distance.
The model performance has been optimized using a mechanism called spectral trigger, which reduces the use of memory up to 5% by filtering noise in data, thus increasing the quality of the information processed.
More precise predictions in less time
Its developers say that this conception gives Surya a significant advantage: unlike other algorithms that require in -depth labeling of the data which is powered to them, Surya can directly learn raw data. This allows it to adapt quickly to different tasks and provide reliable results in less time.
During the tests, Surya demonstrated its versatility in the integration of data from other instruments, such as the Solar PARKER probe and the solar and helospheric observatory (SOHO), two other spacecrafts that observe the sun. Surya has also proven to be effective in various predictive functions, in particular by predicting the activity of thrusts and the speed of the solar wind.
According to IBM, traditional prediction models can only predict one hour escape in advance based on signals detected in specific sun regions. On the other hand, “Surya has provided a two -hour advance using visual information. The model is considered the first to provide a warning of this type. During the early tests of the model, the team said it has achieved an improvement of 16% of the precision of the classification of solar pushes, a marked improvement in relation to existing methods,” the company said in a press release.
NASA stresses that, although the model has been designed to study heliophysics, its architecture is adaptable to different fields, from planetary science to the observation of the earth. “By developing a foundation model formed on NASA heliophysics, we facilitate the analysis of the complexities of the sun behavior with unprecedented speed and precision,” said Kevin Murphy, Director of Data Science at NASA, in a press release. “This model allows a more in -depth understanding of the impact of solar activity on critical systems and technologies on which we all count here on Earth.”
The risk posed by abnormal solar activity is not a minor. A major solar storm could directly affect global telecommunications, the collapse of electrical networks and disrupt GPS navigation, satellite operations, Internet connections and radio transmissions.
Andrés Muñoz-Jaramillo, solar physicist at the Southwest Research Institute in San Antonio, Texas, and the main scientist on the project, stressed that the object of Surya is to maximize the accounting period of these possible scenarios. “We want to give the land as long as possible. Our hope is that the model learned all the critical processes behind the evolution of our star in time so that we can extract usable information. ”
This story originally appeared on Cable in Spanish and was translated from Spanish.
