Nuclear Fusion 34 - Google And Tri Alpha Energy Have Developed A New Algorithm To Help Stabalize Plasmas
I have been blogging about nuclear fusion lately as a source of energy. In nuclear fusion, light nuclei like hydrogen and helium are fused into heavier nuclei. An enormous amount of energy is released when nuclei fuse. Nuclear fusion research has absorbed billions of dollars over the last sixty years but has yet to yield a commercial power generation system. If nuclear fusion can be made into practical system for the generation of electricity, a lot of current problems involving energy generation will be reduced or eliminated. These include pollution, dangerous waste, carbon dioxide release and the danger of accidents at power plants.
In fusion reactors, ionized gas called a plasma is subjected to great pressure and temperature to trigger the fusion reaction. One of the big problems is controlling and stabilizing the confinement of the heated plasma. There are various approached to solving this problem but they all must deal with non-linear behavior including turbulence which can dissipate the plasma and interfere with the fusion.
Google which is famous for advanced artificial intelligence research and a fusion research company named Tri Alpha Energy have joined forces to work on the problem of dealing with non-linear behavior in plasmas. Together they have developed something called the Optometrist algorithm. The basic idea is to combine advanced AI and power computation with human judgement in order to develop better solutions to these complex problems. The computer presents the human with options and the human makes an intuitive choice. This works in much the same way as a human being at an Optometrist’s making a choice on which lense shows a clear version of a letter.
The new system was used by Tri Alpha Energy on their C2-U experimental machine and resulted in accomplishing in a few hours what would have previously taken a month. The algorithm allowed the operators to discover new and unexpected ways of handling the plasma in the machine. They were able to cut energy losses in half and increase total plasma energy. The president of Tri Alpha said “Results like this might take years to solve without the power of advanced computation.”
The C2-U machine operated on a cycle of once each eight minutes. Inside the machine, the plasma was subjected to a beam of hydrogen atoms which kept it spinning in a magnetic field for up to ten milliseconds. The purpose of the series of experiments was to see if the actual behavior of the plasma matched the theoretical models developed by the researchers. The use of the new algorithm led to the discovery of a configuration in which the cooling losses in the machine were balanced by the hydrogen beam. This resulted in an increase in the total energy of the plasma for around two milliseconds.
The C2-U machine has been replaced with a more complex and powerful machine called “Norman.” It began operations this month. If the experiments with the Optometrist algorithm are successful in the Norman machine, the next step will be to work on a prototype power generator. Tri Alpha Energy hopes to be able to actually produce electricity from a commercial fusion reactor within ten years.
The CEO of the Tokamak Energy has said with respect to the Tri Alpha fusion project, “While publicly funded laboratories excel at fundamental research, the private sector can innovate and adopt new technologies much more rapidly.”
