Speaker
Description
In future fusion power plants, full diagnostic coverage may not always be available due to radiation damage, access limitations, or cost constraints. To explore profile control under such conditions, we tested a real-time control scheme on DIII-D that is robust against the loss of primary kinetic diagnostics. The system uses RTCAKENN[1], a neural network trained to infer seven kinetic profiles—including density, temperature, and rotation—based only on real-time-compatible inputs. In experiments, we evaluated its performance by selectively removing inputs from diagnostics such as Thomson scattering or charge exchange. Even with missing data, RTCAKENN continued to provide profile estimates with sub-5 ms latency and promising agreement with available measurements. These inferred profiles were used by a model predictive controller to adjust actuators like neutral beam injection and gas fueling. This approach may offer a practical solution for profile control in reactor environments where some diagnostics are unavailable or degraded.
This work was supported by the National Research Foundation of Korea (NRF), funded by the Korean government (Ministry of Science and ICT) under RS-2023-00255492. It was also supported by the U.S. Department of Energy (DOE), Office of Science, Office of Fusion Energy Sciences, through the DIII-D National Fusion Facility under Award DE-FC02-04ER54698, and under Awards DE-SC0015480 and DE-AC02-09CH11466.
Disclaimer: This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.
References:
[1] R. Shousha et al. NF 64 026006, 2024
