Construction of Neoclassical Transport Database for Large Helical Device Plasma Applying Neural Network Method

A neoclassical transport database for the large helical device (LHD) plasma, DCOM/NNW, is constructed using the neural network method. Monoenergetic neoclassical transport coefficients evaluated by the Monte Carlo code, DCOM, are used as training data of the neural network. The databases for two typical magnetic field configurations in LHD, namely, standard and inward-shifted configurations, are constructed and transport coefficients for thermal plasma are evaluated. The plasma parameter dependencies and the ambipolar radial electric field are investigated.