A U.S. company is speeding up the path to practical fusion energy by using Google’s vast computing power.
By using software that can improve on its own, TAE Technologies has reduced a task that once took two months to just a few hours.
Google has lent the firm its expertise in “machine learning” to help speed up the timeline for mergers.
Nuclear fusion promises an abundant supply of low-carbon energy, using the same process that drives the Sun.
Existing nuclear power is based on fission, in which heavy chemical elements are broken down to give a lighter one. Nuclear investment works by combining two lightweight elements to make it heavier.
For a coalition to be economically viable, it must generate more energy than the amount put in. But no one has yet reached this level, despite eight decades of efforts to “build stars on Earth”. The challenges are enormous, but some in the coalition community hope new thinking and disruptive technologies can help shatter this paradigm.
Controlling plasma at tens of millions of degrees requires a finely tuned system. Google’s expertise in machine learning where computer algorithms improve with experience has been used to optimize TAE combination devices.
Optimization, or tuning for best performance, is done when something on the device changes, such as new hardware added. The process used to take about two months, but with machine learning, “now we can optimize in fractions of an afternoon,” explains Dr. Binderbauer.
Credit to: https://bbc.in/3gKfENH