The performance of improved H-modes at ASDEX Upgrade and projection to ITER

Transcription

1 EX/1-1 The performance of improved H-modes at ASDEX Upgrade and projection to George Sips MPI für Plasmaphysik, EURATOM-Association, D-85748, Germany G. Tardini 1, C. Forest 2, O. Gruber 1, P. Mc Carthy 3, A. Gude 1, L.D. Horton 1, V. Igochine 1, O. Kardaun 1, C.F. Maggi 1, M. Maraschek 1, V. Mertens 1, R. Neu 1, A. Peeters 1, G. V. Pereverzev 1, A. Stäbler 1, J. Stober 1, W. Suttrop 1 and the ASDEX Upgrade Team. 1 Max-Planck-Institut für Plasmaphysik, EURATOM-Association, D-85748, Germany. 2 The University of Wisconsin, Madison, USA. 3 Dep. of Physics, University College Cork, Association EURATOM-DCU, Cork, Ireland. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 1

2 Motivation: performance 8 :15 MA, <n e >/n GW = P aux = 4MW P fus 6 primary goal: Q=1, P fus ~4MW. Q 4 β N =1.8 4 Energy confinement: IPB98(y,2) scaling expression. 2 2 size, 15MA/5.3T, P aux 73 MW, <n e >=.85, H 98 (y,2) Mukhovatov V. et al Nucl. Fusion 43 (23) 942 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 2

3 Motivation: performance Q :15 MA, <n e >/n GW =.85 P aux = 4MW β N =1.8 β N >2 P fus Integrated simulation codes: envelope of performance. Strong dependence, of Q and P fus with H 98 (y,2). Significant increase in performance for small increase in energy confinement ±1 tech. st. dev log non-linear int H 98 (y,2) Mukhovatov V. et al Nucl. Fusion 43 (23) 942 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 3

4 Motivation: performance Q :15 MA, <n e >/n GW =.85 P aux = 4MW β N =1.8 β N >2 P fus Integrated simulation codes: envelope of performance. Strong dependence, of Q and P fus with H 98 (y,2). Significant increase in performance for small increase in energy confinement ±1 tech. st. dev log non-linear int Mukhovatov V. et al Nucl. Fusion 43 (23) H 98 (y,2) H 98 (y,2) > 1, and β N > 2: Higher performance at 15 MA. Long pulse operation, I p =11 to 14 MA, with Q=5-1 (Hybrid). DEMO: H 98 (y,2) 1.2, β N 3.5. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 4

5 Motivation Outline ASDEX Upgrade: Improved H-modes (H 98 (y,2) > 1, and β N > 2). Operational range: - wide range in density, T i /T e and ν*. - at q high beta: Limits and control of NTMs. Scaling to using experimental data. Conclusions. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 5

6 ASDEX Upgrade: H-mode data 1.6 ASDEX Upgrade H-mode data 1.4 H-mode operation for a wide range of plasma conditions: H 98 (y,2) I p =.6-1.4MA, B T =1.6-3.T. All type I ELMy H-modes:.8 q 95 < 5.5, stationary >.2 s. (values shown are averaged).6 H-modes with Type I ELMs β N George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 6

7 Current rise phase Low magnetic shear z(m) I p =3 ka z(m) I p =54 ka z(m) I p =7 ka z(m) I p = MA W C R(m) -1. W C R(m) -1. W C R(m) -1. W C R(m) heating >1. time (s) Current rise: 1. Early divertor configuration t~.5s. 2. low <n e > ~ 3x1 19 m -3, control of impurities. 3. Additional heating: Slow down the current penetration. Low magnetic shear in the centre. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 7

8 Low magnetic shear Improved H-mode 5 4 ASDEX Upgrade: Equilibrated q-profiles ~ zero magnetic shear q q 95 = shear/q ρ tor 1 MSE data available in 26. Timing of preheating: control of q-profile H 98 (y,2). The q-profile remains flat in the core. McCarthy P. et al, EX/P3-7 Stober J. et al, EX/P1-7 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 8

9 Low magnetic shear Improved H-mode ASDEX Upgrade: Equilibrated q-profiles ~ zero magnetic shear (3,2) NTM activity Fishbone activity q q 95 = shear/q ρ tor 1 MSE data available in 26. Timing of preheating: control of q-profile H 98 (y,2). The q-profile remains flat in the core. MHD modes main candidate. Typically (3,2) NTMs and higher (m,n) activity or fishbone activity. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 9

10 ASDEX Upgrade: Improved H-mode data ASDEX Upgrade, all improved H-modes A second dataset includes all improved H-mode discharges: H 98 (y,2) Minimum values required for DEMO Early heating (some selected cases with late heating). Stationary for >.5 s..8 NOT just a selection of the best discharges..6 Improved H-modes β N George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 1

11 ASDEX Upgrade: Improved H-mode data 1.6 ASDEX Upgrade Improved H-modes do not exclusively occupy the domain 1.4 H 98 (y,2) > 1 at β N = 2-3. H 98 (y,2) Any H-mode, when it manages to achieve high beta is likely to develop low central shear in the.8 centre, due to bootstrap current (overlap of grey and red data)..6 Improved H-modes H-modes (Type I) Physics criteria for improved H β N modes would depend on the details of the current density profile (not routinely available). George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 11

12 ASDEX Upgrade: Improved H-mode data Average values q 95 H 98 (y,2) β N Type I ELMy H-modes (944) Improved H-modes (259) Various physics studies on the reason for the high stability/ confinement in improved H-modes. Core: Stober J. et al, EX/P1-7 Pedestal: Suttrop W. et al, EX/8-5 Maggi C. et al, IT/P1-6 (various experiments) l i Similar results have been obtained in other experiments (DIII-D, JET, JT-6U, NSTX..), for a range of conditions (See this conference!!). George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 12

13 Operational range: Density ASDEX Upgrade Improved H-modes: Typically operate at low density, <n e >/n GW = H 98 (y,2) After formation of q(r), the density can be increased together with an increase in.8 heating power and δ..6 Improved H-modes H-modes (Type I) <n e >/n GW <n e >/n GW =.85, H 98 (y,2) ~ 1.2 are obtained at high δ=.4. At 1MA: <n e >=1.1x 1 2 m -3. Tolerable ELMs. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 13

14 Operational range: T i /T e ASDEX Upgrade improved H-modes 1. Operation at high <n e >. 2. ICRH and NBI at low <n e >..7 < T i /T e < 2.5 T i /T e T i /T e ~ 1 for a substantial subset of the data, while maintaining high confinement..5. NBI only NBI+ICRH <n e >/n GW Moreover, some of these discharges have low plasma rotation (low momentum input). George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 14

15 Operational range: ν* H 98 (y,2) ASDEX Upgrade Improved H-modes H-modes (Type I) ν*/ν* Strongest correlation of maximum H 98 (y,2) with plasma collisionality (ν*/ν* ). Also density peaking (n e /<n e >) can be higher at lower ν*/ν*. Weisen H. et al, EX/8-4 β and v* dependence of the IPB98(y,2) scaling expression are under investigation. McDonald D. et al, EX/P3-5 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 15

16 Operational range: low q 95 = I p (MA) P NBI even n odd n D α # 2449 P ICH 1.2MA/2.T q 95 = 3.17, NBI used with beta feedback, 5% of NBI is off-axis! Central ICRF heating. <n e >=6.4x1 19 m -3, T i /T e =1.4 <n e >/n GW =.42, ν*/ν* =2. H 98 (y,2) rises to 1.4 at β N =2.9. C W,core = 2.5x1-5 (< 1-4 ). 3 β N 1 H 98 (y,2) time (s) George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 16

17 Operational range: low q 95 = I p (MA) P NBI even n odd n D α # 2449 P ICH 1.2MA/2.T q 95 = 3.17, NBI used with beta feedback, 5% of NBI is off-axis. Central ICRF heating. <n e >=6.4x1 19 m -3, T i /T e =1.4 <n e >/n GW =.42, ν*/ν* =2. H 98 (y,2) rises to 1.4 at β N =2.9. C W,core = 2.5x1-5 (< 1-4 ). 3 1 β N H 98 (y,2) time (s) Core MHD: (1,1) fishbones. (4,3) NTMs. NO sawteeth. Early (3,2) NTM when β N ~2. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 17

18 Operational range: ECCD, stabilise (3,2) NTM # (1MW ECRH) β N H 98 (y,2) P ECRH # Time (s) 3,2 activity Killer gas disruption at q=1.5 position 1. ECCD: Stabilise this (3,2) NTM at q 95 ~3.1 After ECCD, discharge continues as improved H-mode (shown before). Zohm H. et al, EX/4-1Rb 2. Otherwise (2,1) NTM may develop, and disruption recognition algorithms will act: (mode-lock, radiation peaking, regime identification, neural net): React by issuing a soft stop or triggering killer gas injection. Pautasso G. et al, EX/P Stabilisation of (3,2) NTM is not required for improved H-modes with q 95 > 4. Stober J. et al, EX/P1-7 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 18

19 Scaling to using experimental data Motivation: Integrated simulation codes rely on use of transport models and models for actuators, require benchmarking on current experiments. Scaling the measured kinetic profiles to : 1% fit to experimental data, gradient length preserved, pedestal. Use experimental profile shapes. Same β N, H-factor's. Use geometry. Impurities: Be 2 %, Ar.12 %, He 4.3 % (n D +n T )/n e ~.8. Luce T. et al 24 Phys. Plasmas Assemble set of discharges with good profile measurements. predictions: P fus, P aux, Q (+ other parameters using ASTRA). Tardini G. et al 33rd EPS Conference, Rome June 26, P1.112 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 19

20 Scaling to using experimental data Density: Use shape of n e profile from ASDEX Upgrade, set <n e > =.85n GW. Temperatures: Use shape of T i profile or use shape of T e profile from ASDEX Upgrade (select highest), set T i =T e and β N =β N AUG. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 2

21 Scaling to : q 95 =3., β N =2.1, high <n e > n e (x1 19 m -3 ) n e r min (m) T i,e (kev) β N,th = β N,th AUG T i = T e r min (m) Density: Use shape of n e profile from ASDEX Upgrade, set <n e > =.85n GW. Temperatures: Use shape of T i profile or use shape of T e profile from ASDEX Upgrade (select highest), set T i =T e and β N =β N AUG. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 21

22 Scaling to : q 95 =3., β N =2.1, high <n e > n e (x1 19 m -3 ) n e Scenario 2 T i,e (kev) Scenario 2 β N,th = β N,th AUG T i = T e r min (m) r min (m) Comparison with integrated scenario modelling (Scenario 2): Density: Assumed flat. Temperatures: Set pedestal, models for core transport & heating systems. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 22

23 Scaling to : q 95 =3.17, β N =2.9 n e (x1 19 m -3 ) n e n D +n T Scenario 2 T i,e (kev) Scenario 2 β N,th = β N,th AUG T i = T e r min (m) r min (m) At higher β N (Improved H-mode): Temperature profiles rise over whole plasma region. Consistent with observations of increased edge pedestal. Suttrop W. et al, EX/8-5 Maggi C. et al, IT/P1-6 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 23

24 Prediction of fusion performance in P fus Q=6.5 Q=11.4 predictions at <n>/n GW =.85 Q= Q=8 15 MA 4 typical cases: q 95 β N H 98 (y,2) Open symbols P aux > 73 MW* Closed symbols P aux 73 MW* I p (MA) *P aux based on using IPB98(y,2) scaling. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 24

25 Prediction of fusion performance in 12 predictions at <n>/n GW =.85 Operation at β N,th =2-3: P fus β N,th = 2-3 β N,th MA Significant fusion power for I p =9.5MA to I p =12MA. With bootstrap current fraction f BS ~.4, and Q=6-15. Long pulse lengths (> 4s). 2 Open symbols P aux > 73 MW* Closed symbols P aux 73 MW* I p (MA) At low q 95 ~3 (I p =14-15 MA) would in principle be able to reach P fus ~1GW, Q=. *P aux based on using IPB98(y,2) scaling. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 25

26 Prediction of fusion performance in 4 Physics Basis 1999 Nucl. Fusion IPB98 scaling Petty G. et al 23 Fusion Sci. Technol GyroBohm scaling Cordey J. G. et al 25 Nucl. Fusion Cordey5 scaling P aux 2 P aux = 73 MW Q = β N,th - Q = β N,th - Q = β N,th - Q and input power (P aux ) required (ASDEX Upgrade profile data): Calculated for three different energy confinement scaling expressions. Using IPB98(y,2): In some high β cases : P aux > 73MW. I p <11MA, despite H 98 (y,2)~1.4. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 26

27 Conclusions ASDEX Upgrade: H-modes with low magnetic shear in the centre: H 98 (y,2) > 1 and β N = Operation at high density and T i /T e ~ 1 is demonstrated. Highest H 98 (y,2) values are achieved at relevant ν*. At q 95 =3.1: H 98 (y,2)=1.4, β N =2.9, fishbone activity keeps q(r) stationary. ECCD can be used to stabilise (3,2) NTM activity (q 95 ~3). predictions, scaling kinetic profile shapes : <n e >=.85n GW, keep β N,th. At q 95 =3.1: P fus =17 MW, Q=. At I p =9.5-12MA: P fus 3-6 MW, Q=6-15 (some cases P aux >73 MW). George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 27

28 Back-up slides George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 28

29 Operation: Tungsten coverage 85% W CONC, in the core 1e-2 1e-3 1e-4 1e-5 ASDEX Upgrade Improved H-modes All AUG discharges 1e-6 Gruber O. et al, OV/2-3 Dux R. et al, EX/3-3Ra 1e Discharge # All H-modes require central ICRH or ECRH to avoid impurity accumulation. Operation at <n e > ~ 4-6x1 19 m -3 requires a boronisation to minimise tungsten influxes. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 29

30 Operational range: Density, ν* Contours for H 98 (y,2) values Strongest correlation of maximum H 98 (y,2) with plasma collisionality (ν*/ν* ). ν*/ν* Also density peaking (n e /<n e >) can be higher at lower ν*/ν*. 2 1 β and v* dependence of the IPB98(y,2) scaling expression are under investigation β N McDonald D. et al, EX/P3-5 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 3

31 Operational range: low q 95 = F (khz) Core MHD activity ELMs Soft X-ray data 6 4 (4,3) NTMs 2 fishbone activity time (s) George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 31

32 Use of ECCD to stabilise (3,2) NTM # (1MW ECRH) # T e _Fit (kev) # ECRH deposition 4 T i _Fit (kev) # ECRH deposition 3,2 activity β N H 98 (y,2) Killer gas disruption 1 NTM t=1.9s With ECRH t=2.3s rho_pol # n e _Fit (x1 19 m -3 ) ECRH deposition 1 NTM t=1.9s With ECRH t=2.3s rho_pol Zohm H. et al, EX/4-1Rb P ECRH Time (s) 2 NTM t=1.9s With ECRH t=2.3s rho_pol George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 32

33 Scenario 2 Physics Basis 1999 Nucl. Fusion n e (x1 19 m -3 ) n e n D +n T T i,e (kev) T i, T e (T i +T e )/ r min (m) r min (m) <n e >/n GW =.85 q 95 I p (MA) β N P fus H 98 (y,2) Q P aux <n e > (1 19 m -3 ) n e (1 19 m -3 ) T e /T i (kev) Sc / 21.2 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 33

34 Scaling to : q 95 =3., β N =2.1, high <n e > n e (x1 19 m -3 ) n e n D +n T Scenario 2 T i,e (kev) Scenario 2 β N,th = β N,th AUG T i = T e r min (m) r min (m) <n e >/n GW =.85 q 95 I p (MA) β N P fus H 98 (y,2) Q P aux <n e > (1 19 m -3 ) n e (1 19 m -3 ) T e /T i (kev) AUG # (78) George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 34

35 Scaling to : q 95 =3.17, β N =2.9 n e (x1 19 m -3 ) n e n D +n T Scenario 2 T i,e (kev) Scenario 2 β N,th = β N,th AUG T i = T e r min (m) r min (m) <n e >/n GW =.85 q 95 I p (MA) β N P fus H 98 (y,2) Q P aux <n e > (1 19 m -3 ) n e (1 19 m -3 ) T e /T i (kev) AUG # George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 35

36 Scaling to : q 95 =3.8, β N =2.6 n e (x1 19 m -3 ) n e n D +n T Scenario r min (m) T i,e (kev) Scenario 2 β N,th = β N,th AUG T i = T e r min (m) <n e >/n GW =.85 q 95 I p (MA) β N P fus H 98 (y,2) Q P aux <n e > (1 19 m -3 ) n e (1 19 m -3 ) T e /T i (kev) AUG # George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 36

37 Scaling to : q 95 =4.6, β N =2.4 n e (x1 19 m -3 ) n D +n T n e Scenario r min (m) T i,e (kev) Scenario 2 β N,th = β N,th AUG T i = T e r min (m) <n e >/n GW =.85 q 95 I p (MA) β N P fus H 98 (y,2) Q P aux <n e > (1 19 m -3 ) n e (1 19 m -3 ) T e /T i (kev) AUG # George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 37

38 Prediction of fusion performance in <n e >/n GW =.85 q 95 I p (MA) β N P fus H 98 (y,2) Q P aux <n e > (1 19 m -3 ) n e (1 19 m -3 ) T e /T i (kev) profile shapes Sc / 21.2 AUG # (78) AUG # AUG # AUG # George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 38

39 Prediction of fusion performance in 12 I BS /I p predictions at <n>/n GW =.85 Calculated bootstrap fraction f BS increases at lower current. P fus =.25 =.3 =.35 =.4 (15MA) ~.18. Maximum of cases presented here: f BS ~ MA A current ramp to lower plasma 2 currents compared to reference design AND the I p (MA) ability to operate at same stored energy. longer pulse length. George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 39

40 Prediction of fusion performance in P fus ASDEX Upgrade Improved H-modes Open symbols P aux > 73 MW Closed symbols P aux 73 MW 5.3 kev ± 23% T e,ped _ (kev) Most of the scaled edge pedestal temperatures are within range of what is expected for typical H-modes in. 5.3keV ± 23% Sugihara M. et al 23 Plasma Phys. Control. Fusion 45 L55 George Sips 21 st IAEA Fusion Energy Conference, Chengdu, China, October 26 4