GPU-based optimal control for RWM feedback in tokamaks

Clement, M.; Hanson, J.M.; Bialek, J.; Navratil, G.A.

doi:10.1016/j.conengprac.2017.08.002

Cited by 11 publications

(20 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RWM modelling codes, such as VALEN [11] and CarMa [12] are based on coupling the plasma MHD equations with the eddy current equations for the conductive wall. RWM control experiments were mostly performed with easily reproducible current-driven RWMs [6], or by simulating RWMs with a different set of control coils [13]. Pressure-driven RWMs [10] are more relevant for advanced tokamak scenarios, but they are less reproducible and are often accompanied by other types of MHD instabilities, including edge localized modes (ELM), neoclassical tearing modes (NTM), etc.…”

Section: Introductionmentioning

confidence: 99%

“…The initial RWM feedback control approaches were based on multiple local single-input single-output (SISO) proportional (P) or proportional-derivative (PD) control of actuator coils based on adjacent sensor measurements of the radial and/or poloidal magnetic field [9,4]. Later, several advanced model-based control approaches have been applied, including linear quadratic Gaussian (LQG) optimal control [11,10,7,15,16,13] and model predictive control (MPC) [17,18]. The control-oriented model required for these approaches can be obtained either with experimental or with firstprinciples modelling and model simplification/reduction procedures.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Finite-word-length FPGA implementation of model predictive control for ITER resistive wall mode control

Gerkšič

Pregelj

2021

Fusion Engineering and Design

View full text Add to dashboard Cite

In advanced tokamak scenarios, active feedback control of unstable resistive wall modes (RWM) may be required. A RWM is an instability due to plasma kink at higher plasma pressure, moderated by the presence of a resistive wall surrounding the plasma. We address the dominant kink instability associated with the main nonaxisymmetric (n = 1) RWM, described by the CarMa model. Model predictive control (MPC) is used, with the aim of enlarging the domain of attraction of the unstable RWM modes subject to power-supply voltage constraints. The implementation of MPC is challenging, because the related quadratic programming (QP) on-line optimization problems must be solved at a sub-ms sampling rate. Using complexity-reduction pre-processing techniques and a primal fast gradient method (FGM) QP solver, sufficiently short computation times for ITER are reachable using a standard personal computer (PC). In this work we explore even faster finite-word-length (FWL) implementation using field-programmable gate arrays (FPGA), which would facilitate experimental testing of such control algorithms on dynamically faster medium-sized tokamaks, and compare the computational accuracy and time with the PC implementation.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Finite-word-length FPGA implementation of model predictive control for ITER resistive wall mode control

Gerkšič

Pregelj

2021

Fusion Engineering and Design

View full text Add to dashboard Cite

show abstract

“…Extensive research of RWMs was carried out in reversed-field pinch (RFP) and tokamak experiments in RFX-mod [11,12,13,14,15], EXTRAP T2R [16,17,18,19,20], HBT-EP [21], DIII-D [4,22,23,8,24,15,25,26,27], NSTX [28,29,30], and JT-60 [6]. Modelling codes, such as VALEN [31,26,27], CarMa [32,33,34,35,36], CarMa0NL [37,38] are generally based on coupling MHD equations for the plasma with the equations describing the currents induced in the conductive wall. RWM control research for ITER [39,40,10] currently relies on modelling.…”

Section: Introductionmentioning

confidence: 99%

“…Setiadi et al [20] presented a LQG controller for EXTRAP T2R with a gray-box modelling approach combining a finite-element model for the conductive shell and black-box modelling (model identification) for the plasma. The volume of experimental results of LQG with high-βN H-mode tokamak plasmas is relatively limited in comparison to the conventional PD-control-based approaches, but it was demonstrated that LQG may facilitate longer pulse duration in H-mode high βN conditions in NSTX [30] and DIII-D [26,27].…”

Section: Introductionmentioning

confidence: 99%

Model predictive control of resistive wall mode for ITER

Gerkšič

Pregelj

Ariola

2020

Fusion Engineering and Design

View full text Add to dashboard Cite

Active feedback stabilization of the dominant resistive wall mode (RWM) for an ITER H-mode scenario at high plasma pressure using infinite-horizon model predictive control (MPC) is presented. The MPC approach is closely-related to linear-quadratic-Gaussian (LQG) control, improving the performance in the vicinity of constraints. The control-oriented model for MPC is obtained with model reduction from a high-dimensional model produced by CarMa code. Due to the limited time for on-line optimization, a suitable MPC formulation considering only input (coil voltage) constraints is chosen, and the primal fast gradient method is used for solving the associated quadratic programming problem. The performance is evaluated in simulation in comparison to LQG control. Sensitivity to noise, robustness to changes of unstable RWM dynamics, and size of the domain of attraction of the initial conditions of the unstable modes are examined.

show abstract

“…Accordingly, several methods have been developed to simulate the systems, to analyze their behaviours and to carry out the controller design in order to exhibit desired and/or imposed behaviours. In addition, as shown in [8] for optimal control, [24] for cascade control, [31] for robust control and [33] for minimum variance control, modelling the dynamics of complex systems by state-space models can guarantee high control system performance.…”

Section: Introductionmentioning

confidence: 99%

An Approach to the Design of Nonlinear State-Space Control Systems

Pozna¹,

Precup²

2018

STUD INFORM CONTROL

View full text Add to dashboard Cite

This paper proposes a cost-effective approach to the design of nonlinear state-space control systems. The approach is supported by a geometrical illustration of systems evolution in the state space, by the Lyapunov's direct method, the native behaviour of the controlled process, and the desired system matrix. The method is exemplified through the medium of a realworld process represented by the pendulum-cart system laboratory equipment.

show abstract

GPU-based optimal control for RWM feedback in tokamaks

Cited by 11 publications

References 28 publications

Finite-word-length FPGA implementation of model predictive control for ITER resistive wall mode control

Finite-word-length FPGA implementation of model predictive control for ITER resistive wall mode control

Model predictive control of resistive wall mode for ITER

An Approach to the Design of Nonlinear State-Space Control Systems

Contact Info

Product

Resources

About