Status of H-mode confinement Database by K. Thomsen, Lausanne, May 2007

Status of H-mode confinement Databaseby K. Thomsen, Lausanne, May 2007

New preliminary version DB4v4 (12100 obs) - New data from AUG, DIII-D, JET, JT-60U, NSTX- Hybrid data marked with new variable HYBRID (=yes)- Other new variables: OJK2006, RHOINV, TAUCR, FBS, WROT, RHOQ2, ITB, ITBTYPE

Corrections have been received from AUG, JET, JT-60U. Only JT-60U has released FBS and RHOQ2 data Other problems has been identified by A.Chudnovskiy. Hybrid datasets from AUG, DIII-D and JET are

preliminary versions. Notice PHASE=HGELM? OJK2006 will be set in the next release NF paper on DB3 has been published

Nucl. Fusion 47 (2007) 147-174. Well done Darren!!!

Analyses

New scaling expression(s) from Otto Kardaunas presented yesterday

We need to establish an action plan for the paper based on Otto’s analyses

Analyses of the Hybrid data in DB4v4- KT no time due to other commitments- Any other?- Can we do some analyses at this meeting?- Action plan

Version DB4 compared to DB3

New data from NSTX New data from JET New data from DIII-D New data from MAST New data from AUG New data from JT-60U

Data descriptions outstanding on the new data Review Sheets

Pedestal data?

Web page with documentation and access to different formats of the Db4v4 database http://efdasql.ipp.mpg.de/igd/ITER/DataInfo/Db4v3/DB4v4.htm

DB4v3 is combined w. Pedestal DB DB3v3 (HDB4v3_PDB3v3)

Review sheets for all data in DB4v3

New List of Variables

Action List from Chengdu

Combine hybrid data with DB4v3 + new AUG dataAUG, DIII-D and JET data before 1st DecemberJT-60U and NSTX data before 1st January – Done(late)

Data values for the new ‘hybrid’ variables to be collectedr/a(q=2), taucr, fBS, Rinv - (taucr and Rinv calc. Only JT-60U has provided r/a(q=2) and FBS)

Analyse Hybrid data H-mode WS contribution (KT) – not yet

Work on OK’s analysis paper ( and EPS?) – in progress

Work together with pedestal group to improve scalar pedestal db with hybrid data new two term scaling analysis – no progress

Comments on Issue Cards as required – none so far

Data submission List as promised in Chengdu

AUG: Hybrid dataset - DONE FR C-Mod: ?? DIII-D: Hybrid dataset - DONE JD JET: Hybrid dataset - DONE

DM JFT-2M: ?? JT-60U: Hybrid dataset - DONE HT

Beta scaling data - ? TT

TCV: ?? MAST: ?? NSTX: IP, BT and beta scan data - DONE SK Others ??

Comments from Alexander Chudnovskiy on DB4v4

Comments on results of comparison1. There is TYP-I-II among ELMTYP types.

This type is not described in the List of variables. The variable list includes TYPE-II but it is absent in DB4v4.

2. There is O1 among ECHMOD types. It is not described in the List of variables. The variable list includes O but it is absent in DB4v4.

Remarks on parameters values.1. The PGASA for 2 JET shots 46004 and 46005 is equal to 14.01 in

both DB4v4 and DB3v13.2. The BETMHD values for NSTX shots from 115495 to 117995 are

in the range from 5 to 35. They are probably given in percents.3. A lot of AUG shots have to high NE0 values.

Hybrid definition

1. Define baseline ITER scenario with aims

2. Create candidate modes on present toks

3. See if they scale to baseline

Definition:

• Steady, low central (x=0-0.5) shear q-profile (tdur > 1-2 tauCR)

• steady state w. ELMs for tdur >> tauE,

Hybrid characterisation

• Confinement scaling v tau98(y,2)…

• Residuals v. H98(y,2) for new fraction

• Correlation study

• Bootstrap fraction

Extra variables: fBS, rho(q=2), TauCR, inversion radius, ITB flag

Analysis of Hybrid data

Tauth against tau98y2 ☺, rhostar ☺, nustar ☺ Li vs betaN Nustar vs rhostar ☺ Get db2p8 of db4v3 ☺ Fbs vs q95 Ip vs Bt ☺ Tauth vs nel ☺ Wth vs plth H98y2 vs tauth ☺,nel ☺, plth ☺, rhostar ☺, nustar ☺ JT-60U hybrid data ☺

Correlations in Hybrid dataset

B,I: AUG has good coverage, JET has fixed q95 but a I range, DIII-D at fixed B,I

nu*,rho*: good rho* spread, less so in nu*. AUG has some big outliers due to high density discharges

AUG, DIII-D, JET

Fit of confinement to IPB98(y,2)

JET seems to follow IPB98(y,2) with H=1.

Both AUG and DIII-D have improved confinement with the bulk of the data with H=1-1.7. H clearly increases with tauE, so has a different scaling

AUG, DIII-D, JET

Residual trend with density and power

Density and power scalings on DIII-D are very different to 98(y,2). JET shows no residual, AUG may follow either trend

One concern is the narrow spread of data in DIII-D

AUG, DIII-D, JET

Trend with rho* and nu*

•Strongest correlation seems to be residual H scaling with rho*

•This would not be good for ITER

•May be due to strong rho*-BetaNth correlation, but may be real (q95~4 had more Bohm-like scaling in ELMy H-mode rho* scan)

AUG, DIII-D, JET