PhD típus:
Fizikai Tudományok Doktori Iskola
Év:
2026/2027/2
Munkahely neve:
Institute of Nuclear Techniques
Munkahely címe:
1111 Budapest, Műegyetem rakpart 9.
Leírás:
Power-plant-scale tokamak fusion reactors face numerous challenges in plasma physics, engineering, and plasma control. Power exhaust is a central issue because the heat and particle loads reaching plasma-facing components must be maintained within acceptable limits to enable long-pulse and, ultimately, steady-state operation. Detached divertor operation is a well-established power-exhaust mitigation strategy on present-day devices and is also foreseen for the International Thermonuclear Experimental Reactor (ITER). Detachment is promoted by increasing the divertor neutral density through gas fuelling and/or impurity seeding, thereby enhancing volumetric radiation, atomic processes, and plasma–neutral interactions that reduce the heat and particle fluxes reaching the divertor targets. Monitoring and assessing the detachment state and divertor conditions are therefore essential elements of plasma control. On present-day devices, detachment is commonly characterized using diagnostic systems such as Langmuir probes, spectroscopy, bolometry, and infrared thermography. ITER is expected to require a multi-diagnostic approach in which the sensitivity and limitations of individual diagnostics need to be understood with regard to detachment monitoring, together with the effects of signal degradation and diagnostic aging in the reactor environment. Synthetic diagnostics combined with neural network-based inference methods offer a potential route towards robust, computationally efficient detachment monitoring, a topic of active research.
The doctoral candidate will adapt and, where necessary, develop a multi-diagnostic synthetic-diagnostics framework for ITER-relevant detachment and divertor monitoring. Selected synthetic diagnostics of the framework may be validated using data from present-day fusion devices, such as ASDEX Upgrade, TCV, or WEST. The work may include the development of neural-network-based surrogate or inference models to be compatible with real-time operations. The candidate will consider synthetic diagnostics corresponding to detachment-relevant systems expected to be available during ITER’s Start of Research Operations phase. An ITER SOLPS simulation database will be used to assess the sensitivity, operating range, and limitations of individual diagnostics for detachment monitoring, while accounting for possible reductions in signal quality caused by continued exposure to the reactor environment. The candidate will also investigate neural-network-enhanced surrogate modeling approaches capable of reproducing relevant SOLPS outputs with sufficient accuracy and computational speed for large-scale database generation and real-time-oriented applications. The candidate will investigate synergies between complementary diagnostics and evaluate how multi-diagnostic combinations can improve the robustness and accuracy of detachment-state estimation. Depending on the results, selected components of the synthetic-diagnostics framework may be accelerated or supplemented using neural-network methods to enable real-time execution. Where appropriate, relevant synthetic diagnostics may also be adapted to ITER geometry and integrated into the IMAS framework.
Elvárások:
Advanced skills in computational physics
Advanced English language skills
Finished basic courses in plasma physics
Bsc/Msc in physics or equivalently fluid dynamics, engineering
Állapot:
Végleges
Stipendicum Hungaricum:
No

