Application of the Ito theory for Monte Carlo simulation of plasma diffusion

In this paper the full set of stochastic differential equations (SDEs) are presented describing the guiding centre motion of test charged particles in a plasma with an arbitrary inhomogeneous magnetic field, when the drift approximation is applicable. The derivation is based on the Ito formula which is used to determine stochastic differentials of functions of the non-gyro-averaged velocity diffusion in strict correspondence with the general kinetic equations involving Coulomb collision operators. The drift SDEs are obtained by calculating the Ito stochastic integrals within time intervals admitting the gyro-averaging procedure. The proposed SDEs reproduce the well-known Monte Carlo operators for orbital invariants, however additionally accounting for the spatial drift caused by the cross-field diffusion process with a classical diffusion coefficient. All SDE coefficients are explicitly expressed in terms of the Rosenbluth potentials in a gyro-tropic or isotropic background plasma. The SDEs are presented in particular for the case of an axisymmetric toroidal magnetic configuration to describe the spatial two-dimensional poloidal diffusion process providing a detailed description of neoclassical orbital effects.

SOME EVOLUTION EQUATIONS FOR AN ORNSTEIN–UHLENBECK PROCESS-DRIVEN DYNAMICAL SYSTEM

The statistical properties of the Ornstein–Uhlenbeck (OU) process, a colored noise process, confirm the real noise statistics, since the real noise process has finite, nonzero correlation time. For this reason, it seems worthwhile to develop the estimation-theoretic scenarios of dynamical systems embedded in the colored noise environment as well. Importantly, the application of the Itô theory is not straightforward to the dynamical system in which the OU variable is a driving input. The augmented solution vector approach coupled with the Itô stochastic differential rule plays the pivotal role to develop the theory of the OU process-driven Duffing–van der Pol (DvdP) system of this paper. Notably, the noise analysis of the Duffing–van der Pol system, especially from the estimation-theoretic viewpoint, under the colored noise influence is not available yet in literature. Numerical experimentations with three different sets of data are demonstrated to examine the efficacy of analytical findings of this paper. The results of this paper will be of interest to noise scientists, especially research communities in systems and control, looking for the estimation-theoretic scenarios of the colored noise-driven "vector" stochastic differential system.