特征向量
物理
环面
不稳定性
磁场
数学分析
磁流体力学
旋转对称性
数学
等离子体
量子力学
机械
作者
J. W. Connor,R. J. Hastie,J. B. Taylor
出处
期刊:Proceedings of the Royal Society of London
[The Royal Society]
日期:1979-02-19
卷期号:365 (1720): 1-17
被引量:458
标识
DOI:10.1098/rspa.1979.0001
摘要
In the investigation of stability of a plasma confined by magnetic fields some of the most important modes of oscillation are those with long wavelength parallel to the magnetic field and short wavelength perpendicular to it. However, these characteristics conflict with the requirement of periodicity in a toroidal magnetic field with shear. This conflict can be resolved by transforming the calculation to one in an infinite domain without periodicity constraints. This transformation is the starting point for a full investigation of the magnetohydrodynamic stability of an axisymmetric plasma at large toroidal wave number n . (Small values of n can be studied by direct numerical computation but this fails when n is large.) For n > 1 there are two distinct length scales in the problem and a systematic approximation is developed around an eikonal representation, formally as an expansion in 1/ n . In lowest order the oscillations of each magnetic surface are decoupled and a local eigenvalue is obtained. However, the mode structure is not fully determined in this lowest order. In higher orders a second eigenvalue equation is obtained which completes the determination of the structure of the mode and relates the local eigenvalue of the lower order theory to the true eigenvalue for the problem. This higher order theory shows that unstable modes are localized in the vicinity of the surface with the smallest local eigenvalue, that the true eigenvalue is close to the lowest local eigenvalue and that the most unstable high n modes occur for n -> oo. Hence the local theory, which involves no more than the solution of an ordinary differential equation, is normally adequate for the determination of stability of any axisymmetric plasma to high mode number oscillations.
科研通智能强力驱动
Strongly Powered by AbleSci AI