Optimization of the magnetic field Optimization of the magnetic field configuration for JET breakdown configuration for JET breakdown F. Maviglia F. Maviglia with contribution from R. Albanese, P.J. Lomas Lomas, A. , A. Manzanares Manzanares, M. , M. Mattei Mattei, A. , A. Neto Neto, F.G. , F.G. R. Albanese, P.J. Rimini, P.C. de Rimini , P.C. de Vries Vries and JET EFDA Contributors and JET EFDA Contributors F.Maviglia 1 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Outline Outline Introduction Modelling activity Static and dynamic simulations Intensified Visible Fast Camera KL8A Experimental results Conclusions F.Maviglia 2 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Introduction Introduction Static simulation Breakdown optimization parameters: IPRIM −15 (kA) 4 IP4T 180 (A) filling pressure; P2RU IERFA 0(A) P3MU 3 toroidal electric field (E 0 ≈ 1V/m in P3RU present Tokamaks, E 0 ≈ 0.33V/m 2 5 6 4 ITER); 7 3 P4U 2 1 8 1 9 magnetic field configuration (null Z (m) P1 0 10 position and extension of the low 18 11 17 field region); −1 P4L 12 16 15 13 Circuits used for breakdown at JET: 14 −2 P3RL IPRIM transformer (open loop): −3 P3ML • mode D,C initial value(= premag) [-7,-40] kA, iron fully saturated; P2RL ∆Ψ = 0.01 Vs −4 • mode B premag ≈ 0, iron with residual magnetization; 1 2 3 4 5 6 IP4 Vertical field (open loop) R (m) Hexapolar field IERFA radial field (feedback) F.Maviglia 3 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Introduction Introduction Magnetic field configuration •Ionization length : λ i 3 3 2 10 IP1 −15 [kA] P 4 exp 1 . 3 10 ; IP4 180 [A] i P E 2 P = pressure; with: 1 E = electric field; 1 a B •Connection length L c [1]: null ; Z [m] L 0 c 4 B z a = size of the field min. null −1 B with: = toroidal field; B = avg. stray field over the null; −2 z Avalanche successful if L c / λ i >> 1 ∆Ψ = 0.01 Vs iso B= [0.5;1;1.5;2;2.5] mT −3 minimize < δ B z >, maximize a null 1 2 3 4 5 R [m] [1] B. LLoyd, et al., Nucl. Fus. 31 (1991) 2031. F.Maviglia 4 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Introduction Introduction JET poloidal field (PF) coils JET iron P2 P3 P4 P1 Collar (Shoes not Limbs Central 8 Coil sets, named P1 to shown) column P4,D1 to D4 (D not shown). F.Maviglia 5 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Modelling Modelling Passive structures • CREATE Model [1] based on 2D FEM: • Passive structures: MS, VV, RR, MK2. Resistivity value of MS has been refined via best fit of the simulations with the experimental data. Resistivity of VV, RR, MK2 in agreement with [2-3]. Dyn. Sim. v.s. exp. estimations of mk2 current #78021 (I pre = -15kA) References: [1] R. Albanese et al., Nucl. Fusion, 38, 1998, pp. 723–738. [2] R. Albanese, et.al. Nucl. Fusion 44 (2004) 999–1007. [3] S.Gerasimov, ‘ JET_PassiveSimpleModel.pdf ‘. F.Maviglia 6 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Modelling Modelling JET Iron Model 3 mm gap present in the magnetic circuit at z=4.5m, and not at z=-4.5m, present at JET and Outer limb pad considered in the model: up-down asymmetry. CREATE iron model CREATE−NL/CREATE−L model 6 Centre limb pad 4 -Outer limb pad: thickness 3mm 2 Z [m] 0 −2 -Centre limb pad: thickness 3mm −4 −6 0 2 4 6 R [m] F.Maviglia 7 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Static simulations Static simulations Mode D breakdown Static simulation using as input primary and vertical field current (other noisy E curr. set to 0). -Hexapole null splits in two quadrupole nulls in the |Br| ≈ 0.35mT direction of the radial field given by the upper iron gaps. -Inboard-low null expected to be preferred for plasma formation for the higher electrical field. Critical points for JET breakdown magnetic reconstruction: Residual iron magnetization, not included in present exp. measurements; Perturbing effect of vessel and in-vessel passive currents. F.Maviglia 8 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Intensified Visible Fast Camera KL8A Intensified Visible Fast Camera KL8A Configurations used for breakdown: Configuration 1 Configuration 2 Improved 2 Specifications Specifications 1.5 Region Region 272 x 1 176 x of of 384 px 256 px interest interest 0.5 Z [m] Z [m] Freq. 1.0 kHz Freq. 7.5 kHz 0 Exposur 142.9 Exposur 125.0 −0.5 s s e Time e Time −1 Filter none Filter none −1.5 Intensifi Intensifi 700 v 750 v 2 2.5 3 3.5 er Gain er Gain R [m] F.Maviglia 9 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Standard non optimized breakdown dynamic evolution time increasing H Pulse No: 78369 Pulse No: 78369 Pulse No: 78369 Pulse No: 78369 Pulse No: 78369 t = 0.014028s t = 0.028028s t = 0.038028s t = 0.006028s CREATE L flux-map 2.5 2.0 2.0 2.0 6ms IPRIM = - 19.4(kA) IP4T = 123(A) 2.0 IFRFA = 138(A) Other 0(A) 1.5 1.5 1.5 1.5 Force 1.0 1.0 1.0 1.0 0.5 0.5 0.5 0.5 Z (m) Z (m) Z (m) Z (m) 0 0 0 0 - 0.5 - 0.5 - 0.5 - 0.5 - 1.0 - 1.5 - 1.0 - 1.0 - 1.0 JG10.248-4c JG10.248-5c - 2.0 L - 1.5 - 1.5 - 1.5 ∆ψ = 0.02Vs - 2.5 2.0 2.5 3.0 3.5 2.0 2.5 3.0 3.5 2.0 2.5 3.0 3.5 1.5 2.0 2.5 3.0 3.5 4.0 4.5 R (m) R (m) R (m) R (m) Comparison between dynamic simulation and fast camera [1] ( carbon wall ): -Inner field null preferred for plasma formation. -Plasma pushed against the inboard and included in the region between the two lines where B tg = 0. -The ionization cloud appears to be pushed down in the divertor region. -For a successful breakdown the plasma is pushed up toward the outer wall. [1] F.Mavilgia et al.,Fus. Eng. and Des. 86 (2011) 675–679. F.Maviglia 10 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Static simulations Static simulations Multi-pole field (2D plane geometry): Symmetric multipolar field: np=6 Perturbed hexapolar field, represents iron-gaps filaments added represent JET irongaps - Hexapole null . Hexapole splits in 2 quadrupole nulls in the direction of the perturbation field with vertical and radial components. F.Maviglia 11 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Static simulations Static simulations Radial field added to rotate perturbed field: Zoom in * The two quadrupole nulls are rotated in the direction given by the correction * radial field. * filaments added represent radial field correction F.Maviglia 12 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Static simulations Static simulations Optimization of field null position during breakdown Aims: Place the null field position far from the inner divertor region. Solution: Rotate the 2 quadrupole field nulls by applying an offset radial field bias. Static sim all at 40s: Increasing radial field bias → quadrupoles rotation Standard radial bias = 0A radial bias =+ 120A radial bias =+ 240A radial bias =+ 360A 2.5 2.5 2.5 2.5 0 ms IP1 −20.0 [kA] IP1 −20.0 [kA] IP1 −20.0 [kA] IP1 −20.0 [kA] 0 ms 0 ms 0 ms IP4 258 [A] IP4 258 [A] IP4 258 [A] IP4 258 [A] 2 2 2 2 IFRFA −2 [A] IFRFA 118 [A] IFRFA 238 [A] IFRFA 358 [A] Other PF 0 [A] Other PF 0 [A] Other PF 0 [A] Other PF 0 [A] 1.5 1.5 1.5 1.5 1 1 1 1 0.5 0.5 0.5 0.5 Z [m] 0 0 0 0 −0.5 −0.5 −0.5 −0.5 −1 −1 −1 −1 −1.5 −1.5 −1.5 −1.5 Flux Flux −2 ∆Ψ = 0.01 Vs Flux Flux −2 −2 −2 ∆Ψ = 0.01 Vs ∆Ψ = 0.01 Vs ∆Ψ = 0.01 Vs iso B= [0.5;1;1.5;2;2.5;] mT iso B= [0.5;1;1.5;2;2.5;] mT iso B= [0.5;1;1.5;2;2.5;] mT iso B= [0.5;1;1.5;2;2.5;] mT −2.5 −2.5 −2.5 −2.5 2 3 4 2 3 4 2 3 4 2 3 4 R [m] R [m] R [m] R [m] -In the cases with radial bias ≠ 0 the upper inner field null expected to be preferred for plasma formation due to higher electrical field. F.Maviglia 13 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Experimental results mode D - -20kA 20kA premag premag Experimental results mode D Pulse #80400, std. radial bias =0 Pulse #80402, radial bias =+240A kl1-o4wb-raw @40.04: kl1-o4wb-raw @40.03: kl1-o4wb slow camera (40ms) first visible frame: plasma starts in an upper inner wall position with radial bias correction, far from divertor region. F.Maviglia 14 (27) 7th Workshop on Fusion Data Processing Validation and Analysis. Frascati 26 - 28 March 2012
Recommend
More recommend
