Mathematical modeling plasma transport in tokamaks.

摘要

In this work, we have applied a systematic calibration, validation and application procedure based on the methodology of mathematical modeling to international thermonuclear experimental reactor (ITER) ignition studies. The multi-mode plasma transport model used here includes a linear combination of drift wave branch and ballooning branch instabilities with two a priori uncertain constants to account for anomalous plasma transport in tokamaks. A Bayesian parameter estimation method is used including experimental calibration error/model offsets and error bar rescaling factors to determine the two uncertain constants in the transport model with quantitative confidence level estimates for the calibrated parameters, which gives two saturation levels of instabilities. This method is first tested using a gyroBohm multi-mode transport model with a pair of DIII-D discharge experimental data, and then applied to calibrating a nominal multi-mode transport model against a broad database using twelve discharges from seven different tokamaks. The calibrated transport model is then validated on five discharges from JT-60 with no adjustable constants. The results are in a good agreement with experimental data. Finally, the resulting class of multi-mode tokamak plasma transport models is applied to the transport analysis of the ignition probability in the next generation machine, ITER. The ignition probability of ITER for engineering design activity (EDA) parameters can be formally as high as 99.9% in the present context. The same probability for conceptual design activity (CDA) parameters of ITER, which has smaller size and lower current, is only 62.6%. This suggests that EDA parameters for ITER tokamak are very likely to achieve the self-sustained thermonuclear reaction, but CDA parameters are risky for the realization of ignition.