The ITER Plasma Control Challenge Alfredo Portone Fusion for Energy Special thanks to: M Becoulet, DJ Campbell, JB Lister, A Loarte, G Saibene, H Zohm Workshop ‘Control for Nuclear Fusion’ May 7-8, 2008 Eindhoven University of Technology, The Netherlands
32
Embed
The ITER Plasma Control Challenge Alfredo Portone Fusion for Energy Special thanks to:
The ITER Plasma Control Challenge Alfredo Portone Fusion for Energy Special thanks to: M Becoulet, DJ Campbell, JB Lister, A Loarte, G Saibene, H Zohm Workshop ‘Control for Nuclear Fusion’ May 7-8, 2008 Eindhoven University of Technology, The Netherlands. Synopsis. What is ITER? - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The ITER Plasma Control Challenge
Alfredo PortoneFusion for Energy
Special thanks to: M Becoulet, DJ Campbell, JB Lister, A Loarte, G Saibene, H Zohm
Workshop ‘Control for Nuclear Fusion’May 7-8, 2008
Eindhoven University of Technology, The Netherlands
2
Synopsis
1) What is ITER?• Objectives• Reference parameters• Operation modes
2) Which are the ITER plasma control challenges?• ITER plasma control• Magnetic and kinetic subsystems• Key features of magnetic and kinetic control
3) Conclusions and outlook
3
ITER is ITER is the world’s largest S&T cooperation endeavor carried out under the auspices of carried out under the auspices of IAEA and involving EU, Japan, Russia, US IAEA and involving EU, Japan, Russia, US (founders Parties), China, South Korea & India(founders Parties), China, South Korea & India
ITER first plasma operation is expected in ITER first plasma operation is expected in 20182018
ITER is the experimental step between today’s machines (focused on ITER is the experimental step between today’s machines (focused on plasma physics studies) and tomorrow's fusion power plants. ITER is plasma physics studies) and tomorrow's fusion power plants. ITER is designed to achieve two key objectives:designed to achieve two key objectives:
confine DT plasmas for t > 300 s with α-particle heating >> aux. heatingconfine DT plasmas for t > 300 s with α-particle heating >> aux. heating (Q=Pfus/Paux~10, Paux~50 MW, Pfus~500 MW, P(Q=Pfus/Paux~10, Paux~50 MW, Pfus~500 MW, P~ 100 MW)~ 100 MW)
integrate all key technologies essential for a fusion reactor integrate all key technologies essential for a fusion reactor (superconducting magnets, remote (superconducting magnets, remote maintenancemaintenance, tritium breeding blanket,…), tritium breeding blanket,…)
It looks simple but everything is coupled!It looks simple but everything is coupled!
ITER Plasma Control
7
Courtesy of J ListerCourtesy of J Lister
MAGNETIC CONTROL KINETIC CONTROL
ITER Plasma Control
8
+
Reference Yref KFF
VFF
. .
-8
-6
-4
-2
0
2
4
6
8
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Z, m
R, m
CS
3UC
S2U
CS
1UC
S1L
CS
2LC
S3L
PF1PF2
PF3
PF4
PF5PF6
g1g2
g4
g3
g5
g6
Y VFBKFB
+
Diagnostics
+
-
Shape control requirements
Magnetic Control:Axi-symmetric (n=0) Control
9
Stabilized plant!Stabilized plant!
Control design issuesControl design issues
• Kz and Ky are decoupled in the Kz and Ky are decoupled in the frequency domain (Kz~ 10-30 Hz, frequency domain (Kz~ 10-30 Hz, Ky ~ 0.1-1 Hz)Ky ~ 0.1-1 Hz)
• Strong non-linear nature of power Strong non-linear nature of power supply (e.g. thyristors)supply (e.g. thyristors)
• Open-loop system has 1 pole & 1 Open-loop system has 1 pole & 1 zero in RHP (L* has 1 negative zero in RHP (L* has 1 negative eigen-value) (non min. phase)eigen-value) (non min. phase)
• Current saturation in Kz + Current saturation in Kz + unstable open loop= problems!unstable open loop= problems!
• Avoid loss of control (Kz-loop Avoid loss of control (Kz-loop bullet proof!)… or be prepared for bullet proof!)… or be prepared for a Vertical Displacement Event a Vertical Displacement Event (VDE)!!!(VDE)!!!
VFB
z
* 1ˆ ( )
ˆˆ
ˆˆ
y
z
I sL R V
y C I
z C I
Kz(s)
+
Ky(s)
y
+
Kz(s): typically a lead controller (PD)Kz(s): typically a lead controller (PD)Ky(s): constant gain matrix or LQG Ky(s): constant gain matrix or LQG
Plasma vertical position is open-loop unstable!Plasma vertical position is open-loop unstable!
~ 7000 ton!~ 7000 ton!
2~z pF I z
The vertical de-stabilization force scales asThe vertical de-stabilization force scales as
ITER VDE ~ 10 worse that JET ones!ITER VDE ~ 10 worse that JET ones!
Vz
+
+
-
Vz
Vz
Vz
- If these currents saturate….If these currents saturate….
Vz
z
* 1ˆ ( )
ˆˆ
ˆˆ
y
z
I sL R V
y C I
z C I
Kz(s)
y
Plasma Vertical Instabilities
11
CONTROL SPECSCONTROL SPECSStabilize as higher Stabilize as higher NN as possible as possible
Current limit ~ 200 kACurrent limit ~ 200 kA
CONTROL ACTUATORSCONTROL ACTUATORSOnly (brown) SIDE coils are used for feedback!Only (brown) SIDE coils are used for feedback!
Threshold level ~ 2 mTThreshold level ~ 2 mT
Magnetic Control: Resistive Wall Modes
12
Key design issuesKey design issues
• Plasma models resemble n=0 formalism. However, several complications are present (e.g. accurate modeling of plasma rotation effects). Considerable modeling effort ongoing world-wide (active research)
• Non-linear nature of power supply complicates again closed-loop (see n=0 case)
• For each unstable n the open-loop system has 1 pole. If more than 1 n-mode is unstable (e.g. for n=1 & n=2), enough control knobs are necessary
• Lower plasma current (~ 9 MA) and minor coupling to VDE results in less critical problems in case of loss of control…
• RWMs call for prompt control (f ~ 50 Hz)… superconducting coils do not like AC operation! Minimize control voltage derived from magnetic noise amplification !
* 1ˆ ( )
ˆn n n n
pn n n
I sL R V
B C I
Vn
Bn
KRWM(s)
KRWM must provide strong phase lead!Lead network, or LQG are designed
Magnetic Control: Resistive Wall Modes
13
Bp (mT)
ICC (MA)
VCC (V)
V. Amoskov et al., Plasma Devices and Operations, Vol. 12, No. 3, Sept. 04
Y. Liu et Al.: MARS-F simulation of n=1 RWM stabilization by CC & LQG control
Main actuator: ECCD. Main actuator: ECCD. ITER:4 steerable launchers in upper ports injecting 20 MW of ECCD power localized current drive inside magnetic island to suppress NTM
CommentsComments1. There is not a complete model of
the whole system! the coupling core+SOL is remarkably complicated!
2. 0D models are useful to get qualitatively analyses
3. 1.5 D models are based on computer codes such as ASTRA (core), B2 (SOL)
4. Sometime we try black/gray box approach (system identification)
Kinetic Control: Plasma Core & Divertor Control
18
Divertor temperature control by impurity seeding following a power stepDivertor temperature control by impurity seeding following a power step
Kinetic Control: Divertor Control
19
1. ITER demanding plasma performances (and costly consequences in case of failure!) call for an unprecedented level of sophistication in modeling and control techniques that MUST be both highly performing and fully reliable
2. Modern control competences are – especially at this point in time – of great help to the fusion community to improve the performances of tokamak “advanced mode” operation. The control problems that ITER face in this new physics realm are an outstanding challenge to modern control
3. Modern control areas that have been (and will be more and more) applied to ITER will likely include• Model-based, MIMO control (e.g. magnetic control)• Model reduction of large systems (e.g. eddy currents modeling)• MIMO, robust control (e.g. ITB control) • Non linear control (e.g. reactor kinetics)• System identification methods (e.g SOL modeling)
Conclusions and Outlook
20
The ingenuity and synergy of Physicists andThe ingenuity and synergy of Physicists andControl Engineers is the key to success!Control Engineers is the key to success!
Conclusions and Outlook
21
The ITER challengesITER will provide first test of major fusion technologies… many complex
systems & new problems to be solved in a nuclear environment
Superconducting magnetsUnprecedented size of super-conducting magnet and structures High field performance ~12TPower plant size and field 40GJ
Plasma facing componentsPlasma facing components>10 MW/m2 steady heat flux>10 MW/m2 steady heat flux>10000 cycles/ severe damage>10000 cycles/ severe damage
Diagnostic systemsDiagnostic systems40 different diagnostic systems40 different diagnostic systems
Heating and current drivesHeating and current drives>50 MW continuous>50 MW continuous~1 MeV neutral atoms~1 MeV neutral atomsIon cyclotron, electron cyclotronIon cyclotron, electron cyclotron
Tritium systemsTritium systemsActive recycling of tritiumActive recycling of tritiumTest of lithium blanketsTest of lithium blankets
MHD stability and plasma control -limitsControl of NTMs.Stabilization of RWMs.Disruptions control.