Assessing time-dependent temperature profile predictions for high-performing NSTX plasmas with TRANSP

Spherical tokamaks such as NSTX-U pose unique challenges for transport modeling due to high beta, low aspect ratio, and non-steady conditions. Hence, it is important to assess the reliability of reduced transport models in order to facilitate scenario development on NSTX-U. To this end, predictive TRANSP simulations have been performed on a large set of high performing NSTX shots in order to evaluate the accuracy of time-dependent Te and Ti profile predictions relative to experimental measurements. It is found that the MMM anomalous transport model is reasonably accurate, generally overpredicting the Te and Ti profiles by 20 – 30%. Whereas the electrostatic version of TGLF available in TRANSP at the time of this work has a much more significant Te overprediction, TGLF predicts the Ti profile only moderately worse than MMM. For both transport models, TRANSP predicts better energy confinement than was inferred from the experimental measurements, corresponding to overly steep temperature profiles. In general, the transport models tend to perform best for NSTX plasmas with higher beta, better energy confinement, and broader Te profiles. Future work will involve comparing the TRANSP predictions against a hierarchy of models featuring a range of physics fidelity and computational costs.