Benchmark AUG testcase: MEGA, ORB5, HYMAGYC (2019)

In this page a set of files and parameters used to benchmark MEGA, ORB5, HYMAGYC is collected. The so-called NLED AUG testcase is used (as originally proposed by Ph. Lauber, see NLED_AUG_benchmark_case.pdf). The aim is to cross check these three codes using a set of input and file parameters as closest as possible.

For this purpose, we have modified slightly the original set of parameters in order to have the largest overlapping set among the three codes. In particular:

• The AUG original shaped cross section geqdsk (g031213.00003, updated G-file) file provided by Ph. Lauber, has been used as input to the CHEASE code in order to:
• Smooth the last closed magnetic surface in order to avoid a too sharp surface at the separatrix (CHEASE ang HYMAGYC cannot deal with a separatrix plasma);
• the new equilibrium has been generated from the original one by reading TT' and p', and imposing a scaling such that the safety factor on the magnetic axis matches the value given in the original geqdsk one (q_0 = 2.39895701);
• a new, high resolution EQDSK file, EQDSK_COCOS_02.OUT, has been generated on output, with a (R, Z) box of 361 x 361 mesh points;
• the CHEASE namelist input file is given here (CHEASE_datain.txt); a graphic output (please, ignore the "poloidal beta" plots which are wrong!) of the new equilibrium is given here (pchease.pdf).



• Regarding the density profiles to be used as inputs we agree to define the bulk-ion density profile n_i(s) according to what ORB5 requires, in terms of the electron density profile n_e(s) and of the Energetic Particle (EP) density profile n_H(s), i.e. (for deuteron bulk ions and deuteron energetic particles):
• n_i(s)=n_e(s)-n_H(s), s being the square root of the normalized poloidal flux function (s=0 being the magnetic axis, s=1 being the edge of the plasma. The n_e, n_H profiles are taken as in Ph. Lauber, see p.17-21 NLED_AUG_benchmark_case.pdf).
• To this purpose, an Excel spreadsheet to compute the density profiles for the peaked on-axis and for the peaked off-axis energetic particle density profiles is given here ( profiles_AUG_testcase_benchmark_HYMAGYC_MEGA_ORB5.xlsx). See also the following Picture1.png and Picture2.png, where the different radial profiles are plotted vs. s.
• Update 2022.12.22: New spreedsheet to compute the density profiles for the peaked on-axis and for the peaked off-axis energetic particle density profiles with scaled Energetic Particle density is given here ( profiles_AUG_testcase_benchmark_HYMAGYC_MEGA_ORB5-scaled.xlsx). See also the following Picture1.png and Picture2.png, where the different radial profiles are plotted vs. s

Picture1.png Picture2.png

In the following Table 1, some physical values used in the benchmark for the peaked on-axis and off-axis EP density profile cases are reported, togheter with the origin of the data itself. Note that some of the normalized quantities used, e.g., by CHEASE and HYMAGYC, depend on how the EQDSK file has been written: in particular, the quantities used to scale the lengths and the magnetic field, here indicated, respectively, with R0 and B0. For the orginal EQDSK file g031213.00003 those quantities correspond, respectively, to the magnetic axis major radius R_mag and magnetic field B_mag.

In the following figures, the Shear Alfvén wave continuous spectra alone (obtained using Gamma=0) and the shear Alfven wave + the Ion Sound wave continuous spectra (obtained using Gamma=5/3) as obtained by the linear MHD code MARS are shown:

n=-1_SAW_Gamma0_monotonic_n_H.png, n=-1_SAW_Gamma0_monotonic_n_H.pdf. The data used for this figure can be found here (the txt file contains 3 columns: the first is the poloidal mode number where the singularity characterising the continua is observed, the second one is the "s" coordinate of the singularity, and the third one is the corresponding omega_cont value normalized to omega_A0).

n=-1_SAW+ISW_Gamma5/3_monotonic_n_H.png, n=-1_SAW+ISW_Gamma5/3_monotonic_n_H.pdf. The data used for this figure can be found here .

n=-1_SAW_Gamma0_nonmonotonic_n_H.png, n=-1_SAW_Gamma0_nonmonotonic_n_H.pdf. The data used for this figure can be found here .

n=-1_SAW+ISW_Gamma53_nonmonotonic_n_H.png, n=-1_SAW+ISW_Gamma53_nonmonotonic_n_H.pdf. The data used for this figure can be found here .

Update 2022.12.22: In the following figure, the Shear Alfvén wave continuous spectra as obtained by the FALCON code is shown for the nonmonotonic, Gamma=5/3 case:

continuum_from_Falcon_n-1_Gamma53_AUG-off-axis-kaleida.png, continuum_from_Falcon_n-1_Gamma53_AUG-off-axis-kaleida.pdf. The data used for this figure can be found here .

Update 2022.12.22: In the following figure, a comparison between the Shear Alfvén wave continuous spectra as obtained by the FALCON code and by the linear MHD code MARS is shown for the nonmonotonic, Gamma=5/3 case:

Falcon-MARS-n=-1_Gamma53_AUG-on-axis-comparison.png, Falcon-MARS-n=-1_Gamma53_AUG-on-axis-comparison.pdf.

In the following, some remarks on the simulations are given:

• HYMAGYC and MEGA will use the adiabatic index Gamma=5/3.
• As a first benchmark, only Finite Orbit Width, FOW, effects will be considered (i.e., no Full Larmor Radius, FLR, effects will be used).
• MEGA usually runs with finite resistivity: for this benchmark, a value of S^(-1)=5 x 10^(-7) constant on radius will be used on MEGA for normalized resistivity (S=tau_R/tau_A0 is the Lundquist number (with "tau_R" the resistive time as defined in each code, and "tau_A0" the on-axis Alfvén time). MEGA defines the Lundquist number as S_MEGA=mu_0 R_0 v_A0/eta, "R_0" the on-axis major radius, "v_A0" the on-axis Alfvén velocity, "mu_0" the magnetic permeability and "eta" the resistivity; in the input for HYMAGYC, instead, the Lundquist number is defined as S_HYMAGYC= mu_0 (R_0^2/a) v_A0/eta, where "a" is the minor radius. Note that MEGA also uses a viscosity term, while HYMAGYC does not.

Update 2022.12.22: Nuclear Fusion paper on linear Benchmark: vlad21nf.pdf.

Update 2022.12.22: Slides presented at the zoom meeting of Dec. 21st, 2022: HYMAGYC-nonlinear-benchmark-nonlinear-material-for-IAEA23.pdf.

Update 2022.12.22: Slides presented at the zoom meeting of Dec. 21st, 2022: AUGNLED MEGA simulations_off-axis_nonlinear_simulations-2021.10.22-Noted-GV.pdf.

Update 2022.12.22: Slides presented at the zoom meeting of Dec. 21st, 2022: Francesco-Vannini-mail-2021.08.26_Re_next_EPPI_meeting-and-2021.10.26-GV-notes.pdf.

Update 2023.07.10: Adding the possibility of computing the Energetic Particle pressure terms in HYMAGYC also using (R,Z) mesh. Slides presented at the zoom meeting of July. 10th, 2023: HYMAGYC-nonlinear-benchmark-new-material-for-IAEA23-2023.07.05.pdf.