Masayuki Yokoyama, Professor, National Institute for Fusion Science, Inter-University Research Institute Corporation, National Institutes of Natural Sciences
We are attempting to model the heat transport phenomena of nuclear fusion plasma using a "big data" approach based on the results of numerous experiments and analyses. This research, which regards tens of thousands of plasma experiments as a source of big data and is undertaken using JMP, is fundamentally different in concept from conventional heat transport modeling based on physical mechanisms (first principles).
If this research is successful, it will be possible to present heat transport modeling that is effective over a wide range of plasma parameters, without any consideration of heat transport caused by collisions of ions and electrons in the plasma, heat transport caused by turbulence, the types of fluctuations that cause turbulent transport, etc. It will also be a promising method for easily and quickly predicting the temperature to be reached in the operation control of a fusion reactor.
