The project objective is to bridge the predictive science and engineering application gap in real-time fusion system control. This work develops ML/AI methods to translate computationally expensive predictive science to DNN surrogates that are computationally efficient at deployment as engineering control models.
