Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV)
Plasma stability is one of the obstacles in the path to the successful operation of fusion devices. Numerical control-oriented codes as it is the case of the widely accepted RZIp may be used within Tokamak simulations. The novelty of this article relies in the hierarchical development of a dynamic c...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2016-08-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1996-1073/9/8/609 |
| _version_ | 1811263663548399616 |
|---|---|
| author | Izaskun Garrido Aitor J. Garrido Stefano Coda Hoang B. Le Jean Marc Moret |
| author_facet | Izaskun Garrido Aitor J. Garrido Stefano Coda Hoang B. Le Jean Marc Moret |
| author_sort | Izaskun Garrido |
| collection | DOAJ |
| description | Plasma stability is one of the obstacles in the path to the successful operation of fusion devices. Numerical control-oriented codes as it is the case of the widely accepted RZIp may be used within Tokamak simulations. The novelty of this article relies in the hierarchical development of a dynamic control loop. It is based on a current profile Model Predictive Control (MPC) algorithm within a multiloop structure, where a MPC is developed at each step so as to improve the Proportional Integral Derivative (PID) global scheme. The inner control loop is composed of a PID-based controller that acts over the Multiple Input Multiple Output (MIMO) system resulting from the RZIp plasma model of the Tokamak à Configuration Variable (TCV). The coefficients of this PID controller are initially tuned using an eigenmode reduction over the passive structure model. The control action corresponding to the state of interest is then optimized in the outer MPC loop. For the sake of comparison, both the traditionally used PID global controller as well as the multiloop enhanced MPC are applied to the same TCV shot. The results show that the proposed control algorithm presents a superior performance over the conventional PID algorithm in terms of convergence. Furthermore, this enhanced MPC algorithm contributes to extend the discharge length and to overcome the limited power availability restrictions that hinder the performance of advanced tokamaks. |
| first_indexed | 2024-04-12T19:49:24Z |
| format | Article |
| id | doaj.art-0694df742ba5439591c3663a561a45d6 |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-04-12T19:49:24Z |
| publishDate | 2016-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-0694df742ba5439591c3663a561a45d62022-12-22T03:18:52ZengMDPI AGEnergies1996-10732016-08-019860910.3390/en9080609en9080609Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV)Izaskun Garrido0Aitor J. Garrido1Stefano Coda2Hoang B. Le3Jean Marc Moret4Faculty of Engineering, University of the Basque Country (UPV/EHU), Paseo Rafael Moreno 3, Bilbao 48013, SpainFaculty of Engineering, University of the Basque Country (UPV/EHU), Paseo Rafael Moreno 3, Bilbao 48013, SpainCentre de Recherches en Physique des Plasmas, École Polytechnique Fédérale de Lausanne (CRPP-EPFL), CH-1015 Lausanne, SwitzerlandCentre de Recherches en Physique des Plasmas, École Polytechnique Fédérale de Lausanne (CRPP-EPFL), CH-1015 Lausanne, SwitzerlandCentre de Recherches en Physique des Plasmas, École Polytechnique Fédérale de Lausanne (CRPP-EPFL), CH-1015 Lausanne, SwitzerlandPlasma stability is one of the obstacles in the path to the successful operation of fusion devices. Numerical control-oriented codes as it is the case of the widely accepted RZIp may be used within Tokamak simulations. The novelty of this article relies in the hierarchical development of a dynamic control loop. It is based on a current profile Model Predictive Control (MPC) algorithm within a multiloop structure, where a MPC is developed at each step so as to improve the Proportional Integral Derivative (PID) global scheme. The inner control loop is composed of a PID-based controller that acts over the Multiple Input Multiple Output (MIMO) system resulting from the RZIp plasma model of the Tokamak à Configuration Variable (TCV). The coefficients of this PID controller are initially tuned using an eigenmode reduction over the passive structure model. The control action corresponding to the state of interest is then optimized in the outer MPC loop. For the sake of comparison, both the traditionally used PID global controller as well as the multiloop enhanced MPC are applied to the same TCV shot. The results show that the proposed control algorithm presents a superior performance over the conventional PID algorithm in terms of convergence. Furthermore, this enhanced MPC algorithm contributes to extend the discharge length and to overcome the limited power availability restrictions that hinder the performance of advanced tokamaks.http://www.mdpi.com/1996-1073/9/8/609model predictive controlmultiloop controlfusion reactorsplasma control |
| spellingShingle | Izaskun Garrido Aitor J. Garrido Stefano Coda Hoang B. Le Jean Marc Moret Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV) Energies model predictive control multiloop control fusion reactors plasma control |
| title | Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV) |
| title_full | Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV) |
| title_fullStr | Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV) |
| title_full_unstemmed | Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV) |
| title_short | Real Time Hybrid Model Predictive Control for the Current Profile of the Tokamak à Configuration Variable (TCV) |
| title_sort | real time hybrid model predictive control for the current profile of the tokamak a configuration variable tcv |
| topic | model predictive control multiloop control fusion reactors plasma control |
| url | http://www.mdpi.com/1996-1073/9/8/609 |
| work_keys_str_mv | AT izaskungarrido realtimehybridmodelpredictivecontrolforthecurrentprofileofthetokamakaconfigurationvariabletcv AT aitorjgarrido realtimehybridmodelpredictivecontrolforthecurrentprofileofthetokamakaconfigurationvariabletcv AT stefanocoda realtimehybridmodelpredictivecontrolforthecurrentprofileofthetokamakaconfigurationvariabletcv AT hoangble realtimehybridmodelpredictivecontrolforthecurrentprofileofthetokamakaconfigurationvariabletcv AT jeanmarcmoret realtimehybridmodelpredictivecontrolforthecurrentprofileofthetokamakaconfigurationvariabletcv |