Automatic Identification of MHD Modes in Magnetic Fluctuations Spectrograms using Deep Learning Techniques
This tool provides a fast and reliable method for fusion scientists to extract mode information from large experimental databases, aiding in the control and mitigation of MHD oscillations.
This paper addresses the problem of identifying MHD oscillation modes in fusion devices, which are crucial for control and mitigation. The authors developed a deep learning-based software tool using Convolutional Neural Networks, achieving an AUC score of 0.99 on a test set of Mirnov coil spectrograms from the TJ-II stellarator.
The control and mitigation of MHD oscillations modes is an open problem in fusion science because they can contribute to the outward particle/energy flux and can drive the device away from ignition conditions. It is then of general interest to extract the mode information from large experimental databases in a fast and reliable way. We present a software tool based on Deep Learning that can identify these oscillations modes taking Mirnov coil spectrograms as input data. It uses Convolutional Neural Networks that we trained with manually annotated spectrograms from the TJ-II stellarator database. We have tested several detector architectures, resultingin a detector AUC score of 0.99 on the test set. Finally, it is applied to find MHD modes in our spectrograms to show how this new software tool can be used to mine other databases.