James M. Bialek

Modeling of active control of external magnetohydrodynamic instabilities

A general circuit formulation of resistive wall mode (RWM) feedback stabilization developed by Boozer [Phys. Plasmas 5, 3350 (1998)] has been used as the basis for the VALEN computer code that calculates the performance of an active control system in arbitrary geometry. The code uses a finite element representation of a thin shell structure in an integral formulation to model arbitrary conducting walls. This is combined with a circuit representation of stable and unstable plasma modes. Benchmark comparisons of VALEN results with large aspect ratio analytic model of the current driven kink mode are in very good agreement. VALEN also models arbitrary sensors, control coils, and the feedback logic connecting these sensors and control coils to provide a complete simulation capability for feedback control of plasma instabilities. VALEN modeling is in good agreement with experimental results on DIII-D [Garofalo et al., Nucl. Fusion 40, 1491 (2000)] and HBT-EP [Cates et al., Phys. Plasmas 7, 3133 (2000)]. VALEN ...

High resolution detection and excitation of resonant magnetic perturbations in a wall-stabilized tokamaka)

We report high-resolution detection of the 3D plasma magnetic response of wall-stabilized tokamak discharges in the High Beta Tokamak-Extended Pulse [T. H. Ivers et al., Phys. Plasmas 3, 1926 (1996)] device. A new adjustable conducting wall has been installed on HBT-EP made up of 20 independent, movable, wall segments instrumented with three distinct sets of 40 modular coils that can be independently driven to generate a wide variety of magnetic perturbations. High-resolution detection of the plasma response is made with 216 poloidal and radial magnetic sensors that have been located and calibrated with high-accuracy. Static and dynamic plasma responses to resonant and non-resonant magnetic perturbations are observed through measurement of the step-response following a rapid change in the toroidal phase of the applied perturbations. Biorthogonal decomposition of the full set of magnetic sensors clearly defines the structures of naturally occurring external kinks as being composed of independent m/nu2009=u20093/1 and 6/2 modes. Resonant magnetic perturbations were applied to discharges with pre-existing, saturated m/nu2009=u20093/1 external kink mode activity. This m/nu2009=u20093/1 kink mode was observed to lock to the applied perturbation field. During this kink mode locked period, the plasma resonant response is characterized by a linear, a saturated, and a disruptive plasma regime dependent on the magnitude of the applied field and value of the edge safety factor and plasma rotation.We report high-resolution detection of the 3D plasma magnetic response of wall-stabilized tokamak discharges in the High Beta Tokamak-Extended Pulse [T. H. Ivers et al., Phys. Plasmas 3, 1926 (1996)] device. A new adjustable conducting wall has been installed on HBT-EP made up of 20 independent, movable, wall segments instrumented with three distinct sets of 40 modular coils that can be independently driven to generate a wide variety of magnetic perturbations. High-resolution detection of the plasma response is made with 216 poloidal and radial magnetic sensors that have been located and calibrated with high-accuracy. Static and dynamic plasma responses to resonant and non-resonant magnetic perturbations are observed through measurement of the step-response following a rapid change in the toroidal phase of the applied perturbations. Biorthogonal decomposition of the full set of magnetic sensors clearly defines the structures of naturally occurring external kinks as being composed of independent m/nu2009=u20093/1 ...

Validation of conducting wall models using magnetic measurements

The impact of conducting wall eddy currents on perturbed magnetic field measurements is a key issue for understanding the measurement and control of long-wavelength MHD stability in tokamak devices. As plasma response models have growth in sophistication, the need to understand and resolve small changes in these measurements has become more important, motivating increased fidelity in simulations of externally applied fields and the wall eddy current response. In this manuscript, we describe thorough validation studies of the wall models in the mars-f and valen stability codes, using coil–sensor vacuum coupling measurements from the DIII-D tokamak (Luxon et al 2005 Fusion Sci. Technol. 48 807). The valen formulation treats conducting structures with arbitrary three-dimensional geometries, while mars-f uses an axisymmetric wall model and a spectral decomposition of the problem geometry with a fixed toroidal harmonic n. The vacuum coupling measurements have a strong sensitivity to wall eddy currents induced by time-changing coil currents, owing to the close proximities of both the sensors and coils to the wall. Measurements from individual coil and sensor channels are directly compared with valen predictions. It is found that straightforward improvements to the valen model, such as refining the wall mesh and simulating the vertical extent of the DIII-D poloidal field sensors, lead to good agreement with the experimental measurements. In addition, couplings to multi-coil, n = 1 toroidal mode perturbations are calculated from the measurements and compared with predictions from both codes. The toroidal mode comparisons favor the fully three-dimensional simulation approach, likely because this approach naturally treats n > 1 sidebands generated by the coils and wall eddy currents, as well as the n = 1 fundamental.

Analysis of resistive wall mode LQG control in NSTX with mode rotation

Stabilization of the Resistive Wall Mode (RWM) in the NSTX tokamak is important to achieve high-beta plasmas. This paper numerically investigates state-space control algorithms for improved performance of RWM control using the existing six external control coils with off-midplane poloidal magnetic field sensors in NSTX. Experimentally βN = 5.6 was achieved with the present proportional gain controller. The proposed LQG controller is capable of reaching βN = 6.7 for slowly rotating plasma modes and the ideal wall limit N = 7:06 for plasma modes with higher natural rotation speed.

Controllability and Reduced State Space Models for Feedback Control of the Resistive Wall Kink Mode

Controllability conditions are derived for the resistive wall kink mode using analytic single circuit theory. Using feedback coils that couple more strongly to the plasma than the wall allows controllability up to the ideal wall limit of performance. Conversely, it is found that the controllability of the unstable resistive wall mode is lost at some value of instability growth rate between the no-wall and ideal wall limits when feedback coils are placed outside the passive stabilizing wall. The controllability criterion of the mode can be written in terms of the sign of a dimensionless coupling number that characterizes the magnitude of system inductances. Similar coupling numbers can be defined and used to characterize mode observability conditions as well as the interaction of sensor and feedback coils with the passive wall and plasma. A method is developed based upon contraction of a finite element inductance and resistance matrix model of the system conducting structures that allows quantitative calculation of these coupling numbers for a given system eigenmode. The construction of simple reduced order ordinary differential equation models based upon these coupling numbers accurately calculates mode eigenvalues and serves as a useful starting point for developing few mode approximations of the full eigenmode spectrum for use in advanced controller algorithm development

Nonlinear Continuum Mechanics for Finite Element Analysis

Nonlinear continuum mechanics of solids is a fascinating subject. All the assumptions inherited from an overexposure to linear behaviour and analysis must be re-examined. The standard definitions of strain designed for small deformation linear problems may be totally misleading when finite motion or large deformations are considered. Nonlinear behaviour includes phenomena like `snap-through, where bifurcation theory is applied to engineering design. Capabilities in this field are growing at a fantastic speed; for example, modern automobiles are presently being designed to crumple in the most energy absorbing manner in order to protect the occupants. The combination of nonlinear mechanics and the finite element method is a very important field. Most engineering designs encountered in the fusion effort are strictly limited to small deformation linear theory. In fact, fusion devices are usually kept in the low stress, long life regime that avoids large deformations, nonlinearity and any plastic behaviour. The only aspect of nonlinear continuum solid mechanics about which the fusion community now worries is that rare case where details of the metal forming process must be considered. This text is divided into nine sections: introduction, mathematical preliminaries, kinematics, stress and equilibrium, hyperelasticity, linearized equilibrium equations, discretization and solution, computer implementation and an appendix covering an introduction to large inelastic deformations. The authors have decided to use vector and tensor notation almost exclusively. This means that the usual maze of indicial equations is avoided, but most readers will therefore be stretched considerably to follow the presentation, which quickly proceeds to the heart of nonlinear behaviour in solids. With great speed the reader is led through the material (Lagrangian) and spatial (Eulerian) co-ordinates, the deformation gradient tensor (an example of a two point tensor), the right and left Cauchy-Green tensors, the Eulerian or Almansi strain tensor, distortional components, strain rate tensors, rate of deformation tensors, spin tensors and objectivity. The standard Cauchy stress tensor is mentioned in passing, and then virtual work and work conjugacy lead to alternative stress representations such as the Piola-Kirchoff representation. Chapter 5 concentrates on hyperelasticity (where stresses are derived from a stored energy function) and its subvarieties. Chapter 6 proceeds by linearizing the virtual work statement prior to discretization and Chapter 7 deals with approaches to solving the formulation. In Chapter 8 the FORTRAN finite element code written by Bonet (available via the world wide web) is described. In summary this book is written by experts, for future experts, and provides a very fast review of the field for people who already know the topic. The authors assume the reader is familiar with `elementary stress analysis and has had some exposure to `the principle of the finite element method. Their goals are summarized by the statement, `If the reader is prepared not to get too hung up on details, it is possible to use the book to obtain a reasonable overview of the subject. This is a very nice summary of what is going on in the field but as a stand-alone text it is much too terse. The total bibliography is a page and a half. It would be an improvement if there were that much reference material for each chapter.