The ability to provide fusion burn control without requiring physical access through the first wall and fuel breeding blankets, would be vital for any future, magnetically confined fusion power reactor. A multi-sensor, fusion fuel cycle exhaust, neutral gas analysis system on JET, capable of delivering real time data, and accessing only the sub-divertor region, provides an excellent example of such capability. Optimized for and operated during the deuterium–tritium experimental campaigns 2 and 3 (DTE2, DTE3), it is proving valuable for planning to explore fusion reactor burn control in ITER with a comparable diagnostic system called the Diagnostic Residual Gas Analyzer (DRGA). This paper aims to show feasibility of developing model-based controllers for ITER and next generation, reactor-relevant devices, by building both on the empirical experience in JET-DTE2, and on the already emerging experience on developing such models specifically for ITER. The paper begins with a specific use-case from JET-DTE2, pertaining to the observed sensitivity of the fusion neutron yield on the concentration of isotopic helium-3 (3He), with data from one of the high-performance DT shots exhibited with emphasis on the 3He measurement via the sub-divertor. Then, a first model is developed and then explored with simulations that aim to discover how well the controllers in the model react to either insufficient levels of 3He or excessive levels of 3He. The simulations then explore potential impact from a delay in the measurement (or the response) that would be comparable to the ∼1 s, conductance limited response for the ITER DRGA system, currently in its final design. The simulations show that control is feasible, and that its effectiveness is not significantly impacted by such delay.
This paper explores the use of a multi-sensor system to analyze the neutral gas in the sub-divertor region of a fusion reactor, which could enable real-time control of the fusion burn without physical access to the reactor core. The system was tested on the JET tokamak and could be adapted for use on ITER, providing valuable insights for developing model-based controllers for future fusion reactors.