EAST托卡马克L-H模转换动力学及阈值功率的实验研究

发布时间：2018-06-04 05:42

本文选题：L-H和H-L模转换 + 动力学　；参考：《中国科学技术大学》2017年博士论文

【摘要】：L-H模转换机理是磁约束聚变界35年没有完全解决的难题,理解其机理将有助于实现和稳态维持H模。ITER是按照H模的约束性能来设计的,ITER建成初期的加热功率将刚刚达到按照经验定标率外推的阈值,能否顺利进入H模运行,仍然存在不确定性!较低的L-H模转换阈值功率将会大大缩减ITER以及未来聚变堆的工程造价,接近阈值功率条件下的L-H模转换是聚变界关注的焦点问题之一。本论文依托EAST托卡马克,主要利用高时空分辨的快速往复式探针诊断系统,针对L-H模转换动力学及其阈值功率进行系统的实验研究。EAST是第一个建成的具有和ITER类似先进磁场位形的全超导托卡马克,采用射频波加热为主以及ITER-like钨铜偏滤器,研究EAST上的L-H模转换的阈值功率对ITER具有重要指导意义。我们在EAST上开展了 L-H模转换动力学及其阈值功率的系统的实验研究,发现:(1)在中、高密度区(ne2.0 ×1019),单独低杂波、单独离子回旋波加热或单独中性束加热下L-H模转换功率阈值都接近国际托卡马克定标率。表明国际托卡马克L-H模转换功率阈值定标率适用于EAST,支持这一定标率的可靠性。在低密度区(ne2.0 × 1019),以及采用的低杂波和离子回旋波加热方式,首次观察到L-H模转换功率阈值对密度的U形依赖关系。这种现象与加热方式是否有关,需要在未来实验中验证。ITER主要以射频波加热为主的,在低密度下进入H模的难度可能进一步加大。(2)EAST采用了化学性质很活泼的锂来涂覆面对等离子体的器壁,在此之前很难取得H模。在较低的锂涂层注入量(500 g),使得进入等离子体的碳、氧等轻杂质含量减少,降低有效Z;在较多的锂涂层注入量后,使得器壁上的中性粒子的返流得到了显著的抑制,等离子体边界的中性粒子密度降低了接近一倍,从而显著降低了中性粒子对剪切流的阻尼。最终使得L-H模转换得阈值功率降低了接近一倍,低于国际定标率。ITER将采用铍作为第一壁材料,铍的特性和锂类似,EAST的实验结果为ITER在低加热功率下实现H模运行提供了重要的依据,具有重要意义。(3)研究了偏滤器位型和离子梯度漂移方向对L-H模转换行为及其阈值功率的影响。EAST在锂涂层壁处理下的全碳壁和钼/碳壁,都表现出在双零位型(DN)下功率阚值最低。对于单零位型(SN),当离子梯度漂移背离主X点时其功率阈值低于漂移指向主X点。通过静电探针观测到,低场侧刮削层平行流与Pfirsh-Schluter流的幅值及其时间演化具有相关性,并且依赖于纵场方向。对于W型偏滤器结构,当外中平面到偏滤器靶板的连接长度较短,即刮削层粒子输运沉积在外靶板时,粒子可以有效地被排出、屏蔽,进而有利于L-H模转换。在国际上率先提出了参与驱动刮削层等离子体平行流的Pfirsh Schuter流在决定L-H模转换的阈值功率上起到了关键性的作用,从而打破了刮削层参与L-H转换的物理机制的缺口。然而,当上偏滤器更换为ITER-like的钨铜偏滤器,上单零位型(USN)等离子体在离子梯度漂移指向主X点时其阈值功率较低。上下不对称的偏滤器结构使得低场侧刮削层流指向上偏滤器,对于不同的离子梯度漂移方向只是刮削层流的幅度不同,这和低场侧Pfirsh-Schluter流的指向具有相关性。此外,相比标准的偏滤器位形,EAST准雪花位形下的L-H模转换功率阈值高很多。EAST准雪花位形其等离子体控制的打击点偏离偏滤器靶板区域,不利于粒子的排出和屏蔽。实验观测到,边界中性粒子密度在准雪花位形下较高。L-H模转换行为和位型、偏滤器抽气效率及离子梯度漂移方向有一定的依赖关系,带有中间态的所谓'dithering L-H'模转换更加容易出现在双零位型且抽气效率低的情况,而且dithering的形态在不同位型下也表现出不同的行为特征。单步的转换'single-step L-H'经常出现在离子梯度漂移方向背离主X点的位型下。在DⅢ-D和AlcatorC-Mod的实验结果证明了在离子梯度漂移指向主X点的位形下L-H模转换的阈值功率更低。不同的是,JET发现离子梯度漂移背离有效偏滤器时L-H模转换的阈值功率相差不大或者较小。在ASDEX-U、MAST和NSTX也发现最小的阈值功率出现在DN位形。这意味着我们在研究L-H模转换时还忽略了其它重要因素,如中性粒子再循环和偏滤器几何位形等。EAST装置与众不同的表现,对L-H模转换的研究有着重要意义。(4)在加热功率接近L-H模转换功率阈值的条件下,研究了 L-H模转换的中间振荡状态(limit cycle oscillations,LCOs),发现了等离子体湍流驱动的剪切流、压力梯度主导的平衡流和湍流在L-H模转换过程中共同起作用,决定了 L-H模转换前的振荡态以及L-I-H、L-I-L模转换。(5)在一些实验中发现雷诺协强太小而不足以驱动流剪切,L-H模转换过程中并未观察到湍流驱动的剪切流。我们实验首次发现,E×B流剪切导致湍流径向谱移把湍流能量散射到强粘滞阻尼的高径向波数空间而被耗散,湍流幅度的减小导致压力梯度的增长进一步增强E × B流剪切的谱移,这样的正反馈使得湍流能够被快速抑制,为触发L-H模转换提供了一种新的思路。(6)分析了 EAST托卡马克L-H和H-L模转换前出现的振荡态,它们都表现出(m = 1,n = 0)的磁扰动结构。
[Abstract]:The mechanism of L-H mode conversion is a difficult problem that has not been completely solved for 35 years in the magnetic confinement fusion. Understanding its mechanism will help to realize and maintain the H mode.ITER in the steady state according to the constraints of the H mode. The heating power in the initial stage of ITER will just reach the threshold value extrapolated according to the empirical calibration rate, and whether it can enter the H mode and still exist, still there is no problem. Certainty! The lower L-H mode conversion threshold power will greatly reduce the cost of ITER and future fusion reactor, and the L-H mode conversion under the threshold power condition is one of the focus problems in the fusion field. This paper relies on the EAST Tokamak, mainly using high time space-time resolution fast reciprocating probe diagnosis system, for L-H mode conversion .EAST is the first complete superconducting tokamak with ITER similar advanced magnetic field configuration, using radio frequency wave heating and ITER-like tungsten copper bias filter. The study of the threshold power of L-H mode conversion on EAST has important guiding significance for ITER. We carry out on EAST The experimental study of the L-H mode conversion dynamics and the threshold power system shows that: (1) in the middle, high density region (ne2.0 x 1019), single low clutter, single ion cyclotron wave heating or single neutral beam heating, the L-H mode conversion power threshold is close to the international Tokamak calibration rate. It shows that the international Tokamak model conversion power threshold calibration The rate is suitable for EAST to support the reliability of this certain rate. In the low density zone (ne2.0 x 1019) and the low clutter and ion cyclotron heating mode, the U dependence of the L-H mode conversion power threshold on the density is observed for the first time. This phenomenon is related to the heating mode. It is necessary to verify the.ITER in the future experiment with the radio frequency wave. Heating mainly, the difficulty of entering the H mode at low density may further increase. (2) EAST uses a highly chemically active lithium to cover the plasma wall, and it is difficult to obtain the H mold before this. The lower injection amount of lithium coating (500 g) reduces the content of light impurities such as carbon and oxygen into the plasma, and reduces the effective Z; in comparison, the effective Z is reduced. After the injection of lithium, the reflux of neutral particles on the wall has been significantly suppressed. The neutral particle density of the plasma boundary is nearly doubled, thus significantly reducing the damping of the neutral particles on the shear flow. Finally, the threshold power of the L-H mode is reduced by nearly double the threshold power, which is lower than the international calibration rate of.ITER. By using beryllium as the first wall material, the properties of beryllium and lithium are similar. The experimental results of EAST provide important basis for the implementation of the H mode operation under low heating power. (3) the influence of the displacement of the filter and the direction of the ionic gradient drift on the conversion behavior of the L-H mode and the threshold power of the L-H are studied..EAST under the treatment of the lithium coating wall Both the full carbon wall and the molybdenum / carbon wall show the lowest power threshold at the two zero position type (DN). For the single zero position type (SN), the power threshold is lower than the drift pointing to the main X point when the ion gradient drift deviates from the main X point. The parallel flow in the low field scraping layer is correlated with the amplitude and time evolution of the Pfirsh-Schluter flow. For the structure of the W type filter, when the connection length of the outer middle plane to the filter target plate is short, that is, when the scraping layer particles transport and deposit on the outer target plate, the particles can be effectively discharged, shielded and thus beneficial to the L-H mode conversion. In the world, the first proposed Pfirsh Sc to participate in the parallel flow of the plasma of the scraping layer. The Huter flow plays a key role in determining the threshold power of the L-H mode conversion, thus breaking the gap in the physical mechanism of the scraping layer participating in the L-H conversion. However, when the upper partial filter is replaced by a tungsten copper bias filter of ITER-like, the upper single zero position (USN) plasma has a lower threshold power when the ion gradient drift points to the main X point. The symmetrical bias filter structure makes the low field side scraping laminar flow pointing to the upper partial filter. For the different direction of the ion gradient drift, the amplitude of the scraping laminar flow is different. This is related to the direction of the Pfirsh-Schluter flow on the low field side. In addition, the L-H mode conversion power threshold under the EAST quasi snowflake configuration is much higher than that of the standard filter position. The plasma controlled strike point of AST quasi snowflake deviates from the target plate area of the filter, which is not conducive to the discharge and shielding of the particles. The experimental observation shows that the neutral particle density at the boundary is higher in the.L-H mode conversion behavior and the position type under the quasi snowflake shape, and has a certain dependence on the suction efficiency of the filter and the direction of the ion gradient drift, with the intermediate state. The so-called'dithering L-H'mode conversion is more likely to occur in the case of double zero type and low pumping efficiency, and the morphology of dithering shows different behavior characteristics under different types. The single step transformation of'single-step L-H' often occurs in the position of the main X point in the direction of the ionic gradient drift. In D III -D and AlcatorC-Mod real. The test results show that the threshold power of the L-H mode conversion is lower under the position of the ion gradient drift pointing to the main X point. The difference is that the threshold power of the L-H mode conversion is small or small when JET finds the ion gradient drift deviates from the effective filter. In ASDEX-U, MAST and NSTX also find that the minimum threshold power appears in the DN bit. This means that I have the lowest threshold power in the DN bit. In the study of L-H mode conversion, other important factors, such as the unusual performance of the neutral particle recirculation and the geometric configuration of the bias filter, are important for the study of the L-H mode conversion. (4) the intermediate oscillation state of the L-H mode conversion (limit CYC) is studied under the condition that the heating power is close to the L-H mode conversion power threshold. Le oscillations, LCOs), the shear flow driven by the plasma turbulence is found, the pressure gradient dominated equilibrium flow and the turbulence play a role in the L-H mode conversion process. The oscillating state before the L-H mode conversion and the L-I-H, L-I-L mode conversion are determined. (5) in some experiments, it is found that the Reynolds co intensity is too small to drive the flow shear, and the L-H mode is converted over. The shear flow driven by turbulence is not observed in the process. In our experiment, it is found for the first time that the E x B flow shear leads to the turbulent radial spectral shift to scatter the turbulent energy into the high radial wave space with strong viscous damping and is dissipated. The decrease of the turbulence amplitude causes the increase of the pressure gradient to further enhance the spectral shift of the shear of the E x B flow, so that the positive feedback makes turbulence. The flow can be quickly suppressed to provide a new idea for triggering L-H mode conversion. (6) the oscillating states before the EAST Tokamak L-H and H-L mode conversion are analyzed. They all show the magnetic disturbance structure of (M = 1, n = 0).
【学位授予单位】：中国科学技术大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：TL631.24

【相似文献】