Journal of Fusion Energy最新文献

The effects of divertor pumping on divertor detachment and core performance are investigated using SOLPS-ITER simulations for the H-mode discharges with argon (Ar) seeding on EAST. The simulation results show that a relatively low pumping speed (S) is advantageous for achieving divertor detachment due to an increased Ar density and enhanced radiative dissipation. On the other hand, a relatively low S results in a high Ar density in the core region, which is detrimental to the core performance. Increasing S improves the particle removal capacity, which is not conducive to obtaining detachment but is conducive to reducing the Ar accumulation in the core region. In order to evaluate the compatibility of detachment and high core plasma performance, the impact of S on the divertor Ar retention (measured by impurity compression C_Ar and enrichment E_Ar) and the corresponding physical mechanisms were analyzed. In cases with relatively low and medium upstream density (({n}_{mathrm{e},mathrm{sep}}^{mathrm{OMP}})), a higher S is beneficial to increase C_Ar and E_Ar, i.e. better core-divertor compatibility, mainly due to the diminished neutral diffusion towards the upstream and the enhanced net force on the Ar towards the target. At relatively high ({n}_{mathrm{e},mathrm{sep}}^{mathrm{OMP}}), both C_Ar and E_A show no clear change with increasing S. This is because the negative contribution of the reduced relative distance between the Ar ionization front and the velocity stagnation point of the Ar ions to the Ar retention becomes significant with increasing S, which can even offset the positive contribution of the neutral diffusion and the net force.

The aim of this study is to investigate the behaviour of liquid metal flow in Capillary Porous System (CPS) under strong external magnetic field. Overlapping simple cubic (SC) periodic array of electrically non-conducting spheres with diameter 6 mm and distance between spheres centres 5.6 mm is 3D printed from PLA electrically non-conducting filament. At room temperature, flow of up to 50 mL/s of In–Ga–Sn in pore space in magnetic field of superconducting magnet up to 5T is investigated. Three orientations of magnetic field in relation to the main flow in SC cell are considered—colinear with main flow and perpendicular to it. The values of Reynolds, Hartmann and Stuart numbers in experiment are up to 1160, 90 and 350, respectively. The results indicate that parallel to the main flow orientation of magnetic field has little influence on the flowrate, while perpendicular orientation strongly reduces flowrate with dependence close to 1/Ha, which agrees well with ANSYS Fluent simulations in a unit SC cell, resembling results for channel flow in magnetic field.

The liquid metal magnetohydrodynamic (MHD) flow through coupled ducts with conducting walls under inclined transversal gradient magnetic field is an important physical flow phenomenon, which has the unknown physical mechanism about the interaction between the electromagnetic coupling effect and the three-dimensional (3D) MHD effect. To reveal this physical mechanism, 3D numerical simulations based on a customized solver in the OpenFoam environment are conducted to systematically study the effect of inclined gradient magnetic field on the MHD flow states through coupled conducting ducts. Then the mechanism behind the generation of the 3D MHD effect in the gradient magnetic field zones has been discussed in detail. It is found that the electromagnetic coupling effect can enhance this 3D MHD effect in the co-flow case, but suppress it in the counter-flow case. Moreover, the strong electromagnetic coupling effect in the counter-flow case will induce a “self-circulation” flow region in the duct when the external magnetic field is inclined, and the inclined angle also has a great influence on the area of this flow region, which reduces with the increase of the inclined angle. These results are important for the in-depth fundamental understanding of the 3D MHD effect of liquid metal flowing through coupled conducting ducts under inclined gradient magnetic field, and also helpful for the future design of the liquid metal blanket of fusion reactor.

Liquid metals corrode structure materials in fusion, fission, and spallation applications. A duct strongly cooled on the outside surface is proposed to mitigate or eliminate the corrosion problem. A solidified metal layer between the cool duct (T_duct<T_melt) and the liquid metal could serve as an interface to protect the duct from corrosion.

When the local heat flux exceeds specified flux limit, tungsten PFC surfaces can be damaged, which is not acceptable for a reliable reactor operations. The divertor PFCs are typically designed for a specific heat flux limit usually assuming an average steady-state heat flux which is typically 5–10 MW/m². However, in addition to steady-state heat flux, fusion reactor divertor PFCs could experience transient heat fluxes such as ELMs and/or other magnetic reconnection events which can deposit large transient heat fluxes onto the divertor PFCs. The transient divertor heat flux could be significantly larger than the steady-state heat flux which could damage the solid PFC surfaces. The divertor heat flux can be subjected to additional complications such as the uncertainties in the the divertor strike point heat flux projection. Moreover, there are additional experimental observations of non-axisymmetric power flux which can occur under non-axisymmetric magnetic perturbations. The liquid lithium (LL) PFCs is more resilient against such transient heat fluxes as they could evaporate LL as needed and the lost LL can be then replenished afterward. In this paper, we analyze a case for a transient divertor heat pulse of 1 MJ in 10 ms for a ITER-size reactor. This is a small perturbation (~ 0.1%) to the expected plasma stored energy compared to the previously analyzed case of 20 MJ heat pulse. Even with this relatively modest heat pulse, the LL surface undergoes ~ 100 °C temperature rise. However, the resulting LL surface heating without rapid cooldown mechanism could lead to excessive LL evaporation continuing well after the transient heat flux resulting in a significant Li injection of ~ 0.6 mol in about a 200 ms period. This amount of Li injection could cause plasma dilution and performance degradation. On the other hand, an active Li injection capability if optimized could prevent the LL surface temperature rise and thus reducing subsequent Li evaporation into the plasma by a factor of 7 compared to the passive LL PFC case. A crucial tool of active Li injection is a rapid response pellet injector which could inject light impurity pellets before the excessive heat flux could reach the divertor plate causing serious damage. A simple pellet ablation model suggests a favorable pellet deposition profile for smaller ~ 0.1 mm radius pellet with ~ 10–20 m/s velocity. Moreover, if it is possible to inject from the private flux region, the pellet injection efficiency into the high heat flux strike point region can be as high as 80% compared to ~ 50% for the injection from outer radius region. The pellet deposition efficiency can be further improved by designing a shell-pellet which can burst when a certain ablation fraction is reached. A possible implementation technique using an inductive pellet injector with a rapid time response of a few msec is proposed here which can be tested in NSTX-U.

Harsh heat load conditions on plasma-facing components (PFCs) in steady-state and transient phenomena (e.g., disruptions and ELMs) in DEMO fusion reactors question the feasibility of current approaches based on solid targets made of tungsten. This issue calls for the development of innovative plasma-facing components. Liquid metal PFCs with strong convection enhance heat removal capability and resilience after the transient phenomena. However, transporting liquid metal across magnetic fields gives rise to MHD drag. MHD drag for the case of uniform B, estimated analytically, is acceptable. Grad-B MHD drags with straight ducts could seriously drag the LM flow across non-uniform B. Expanding the duct along B and shrinking the duct in a perpendicular direction could make electromotive force |vBh| approximately constant along the duct and significantly reduces the grad-B MHD drag. Here v denotes the flow velocity along the duct, B is the magnetic field strength, and h is the vertical duct size. Three-dimensional simulations for internal and free surface thermo-MHD phenomena have demonstrated that the proposed duct design reduces the total pressure drop along the duct.

A new control system of shattered pellet injection (SPI) has been successfully developed and implemented in the experimental advanced superconducting tokamak. The control system comprises four functional modules responsible for vacuum acquisition, temperature regulation, gas supply, and system protection, which facilitate the safe and stable operation of the SPI. The software framework employed for the SPI control system incorporates experimental physics and industrial control system and Phoebus. Utilizing these integrated control systems, the gun barrel temperature and material gas pressure could be accurately controlled during pellet forming phase. Also, it could cooperatively control the various types of valves to achieve material gas supply, propellant gas supply and timely pumping. Finally, the pellet was successfully generated, separated from the gun barrel, and accelerated into the plasma vacuum vessel controlled by this system. In addition, the issue of extended delay time was observed in SPI experiments, and a potential solution is also proposed in the paper.

When the breakdown of acceleration grid occurs, the core snubber can consume fault energy and limit the arc current. In this paper, the expression of equivalent circuit model is optimized. The value of magnetization inductance at da₁/dt = 0 is analyzed and the equivalent resistance expression eddy current coefficient is revised from 2.5 to 1. The simulation model with gyrator and the high voltage short circuit experiment platform are described. The rationality of the model hypothesis is verified. Then the effectiveness of equivalent circuit model established in this paper is verified. Also, the dynamic response of the discharge circuit is obtained and the influence of system leakage inductance on the surge suppression process is analyzed. At the same time, in this paper, the distributed capacitance of the transmission line is extracted, and the structure parameters of core snubber suitable for −1MV voltage level are designed. Based on the structure design parameters of snubber, distributed capacitance values of transmission line and power supply parameters, the overall configuration scheme of passive protection components in −1MV five-stage voltage system is analyzed.