Skip to content

Fix integration weights for Chodura condition in moment-kinetic runs #456

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
johnomotani merged 1 commit into from
Jun 22, 2026
Merged

Fix integration weights for Chodura condition in moment-kinetic runs #456

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
90 changes: 55 additions & 35 deletions moment_kinetics/src/analysis.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -187,7 +187,7 @@ function check_Chodura_condition(r, z, vperp, vpa, dens, upar, vth, temp_e, comp

v_parallel = vpagrid_to_vpa(vpa.grid, vth[1,ir,is,it], upar[1,ir,is,it], evolve_p,
evolve_upar)
vpabar = @. v_parallel - 0.5 * geometry.rhostar * Er[1,ir,it] / geometry.bzed[1,ir]
vpabar = @. v_parallel - geometry.rhostar * Er[1,ir,it] / geometry.bzed[1,ir]

# Get rid of a zero if it is there to avoid a blow up - f should be zero at that
# point anyway
Expand All @@ -197,27 +197,42 @@ function check_Chodura_condition(r, z, vperp, vpa, dens, upar, vth, temp_e, comp
end
end

@views lower_result[ir,it] =
integral(f_lower[:,:,ir,is,it], vpabar, -2, vpa.wgts, vperp.grid, 0,
vperp.wgts)
if it == ntime
println("check vpabar lower", vpabar)
println("result lower ", lower_result[ir,it])
if evolve_p
vpa_weights_lower = @. vth[1,ir,is,it] * vpa.wgts
vpa_weights_upper = @. vth[end,ir,is,it] * vpa.wgts
else
vpa_weights_lower = vpa.wgts
vpa_weights_upper = vpa.wgts
end
if evolve_p && vperp.n > 1
vperp_weights_lower = @. vth[1,ir,is,it]^2 * vperp.wgts
vperp_weights_upper = @. vth[end,ir,is,it]^2 * vperp.wgts
else
vperp_weights_lower = vperp.wgts
vperp_weights_upper = vperp.wgts
end

lower_result[ir,it] *= 0.5 * temp_e[1,ir,it] / dens[1,ir,is,it]
@views lower_result[ir,it] =
integral(f_lower[:,:,ir,is,it], vpabar, -2, vpa_weights_lower, vperp.grid, 0,
vperp_weights_lower)
#if it == ntime
# println("check vpabar lower", vpabar)
# println("result lower ", lower_result[ir,it])
#end

lower_result[ir,it] *= temp_e[1,ir,it] / dens[1,ir,is,it]

if find_extra_offset
if lower_result[ir,it] ≤ 1.0
extra_offset_lower[ir,it] = 0
else
integrand = f_lower[:,:,ir,is,it]
for ivperp ∈ 1:nvperp
@. integrand[:,ivperp] *= vpabar^(-2) * vpa.wgts * vperp.wgts[ivperp]
@. integrand[:,ivperp] *= vpabar^(-2) * vpa_weights_lower * vperp_weights_lower[ivperp]
end
vperp_integral = @view sum(integrand; dims=2)[:,1]
cumulative_vpa_integral = cumsum(vperp_integral)
cutoff_index = searchsortedfirst(cumulative_vpa_integral, 2.0 * dens[1,ir,is,it] / temp_e[1,ir,it]) - 1
cutoff_index = searchsortedfirst(cumulative_vpa_integral, dens[1,ir,is,it] / temp_e[1,ir,it]) - 1
cutoff_lower[ir,it] = mean(vpabar[cutoff_index:cutoff_index+1])
vpa_before_zero_index = searchsortedfirst(vpabar, -zero) - 1
extra_offset_lower[ir,it] = vpa_before_zero_index - cutoff_index
Expand All @@ -229,7 +244,7 @@ function check_Chodura_condition(r, z, vperp, vpa, dens, upar, vth, temp_e, comp

v_parallel = vpagrid_to_vpa(vpa.grid, vth[end,ir,is,it], upar[end,ir,is,it],
evolve_p, evolve_upar)
vpabar = @. v_parallel - 0.5 * geometry.rhostar * Er[end,ir,it] / geometry.bzed[end,ir]
vpabar = @. v_parallel - geometry.rhostar * Er[end,ir,it] / geometry.bzed[end,ir]

# Get rid of a zero if it is there to avoid a blow up - f should be zero at that
# point anyway
Expand All @@ -240,26 +255,26 @@ function check_Chodura_condition(r, z, vperp, vpa, dens, upar, vth, temp_e, comp
end

@views upper_result[ir,it] =
integral(f_upper[:,:,ir,is,it], vpabar, -2, vpa.wgts, vperp.grid, 0,
vperp.wgts)
if it == ntime
println("check vpabar upper ", vpabar)
println("result upper ", upper_result[ir,it])
end
integral(f_upper[:,:,ir,is,it], vpabar, -2, vpa_weights_upper, vperp.grid, 0,
vperp_weights_upper)
#if it == ntime
# println("check vpabar upper ", vpabar)
# println("result upper ", upper_result[ir,it])
#end

upper_result[ir,it] *= 0.5 * temp_e[end,ir,it] / dens[end,ir,is,it]
upper_result[ir,it] *= temp_e[end,ir,it] / dens[end,ir,is,it]

if find_extra_offset
if upper_result[ir,it] ≤ 1.0
extra_offset_upper[ir,it] = 0
else
integrand = f_upper[:,:,ir,is,it]
for ivperp ∈ 1:nvperp
@. integrand[:,ivperp] *= vpabar^(-2) * vpa.wgts * vperp.wgts[ivperp]
@. integrand[:,ivperp] *= vpabar^(-2) * vpa_weights_upper * vperp_weights_upper[ivperp]
end
vperp_integral = @view sum(integrand; dims=2)[:,1]
cumulative_vpa_integral = reverse(cumsum(reverse(vperp_integral)))
cutoff_index = searchsortedfirst(cumulative_vpa_integral, 2.0 * dens[end,ir,is,it] / temp_e[end,ir,it]; rev=true)
cutoff_index = searchsortedfirst(cumulative_vpa_integral, dens[end,ir,is,it] / temp_e[end,ir,it]; rev=true)
cutoff_upper[ir,it] = mean(vpabar[cutoff_index-1:cutoff_index])
vpa_after_zero_index = searchsortedlast(vpabar, zero) + 1
extra_offset_upper[ir,it] = cutoff_index - vpa_after_zero_index
Expand Down Expand Up @@ -368,7 +383,7 @@ end

"""
"""
function analyze_pdf_data(ff, n_species, ntime, z, vpa, vth, evolve_ppar)
function analyze_pdf_data(ff, n_species, ntime, z, vpa, vth, evolve_p)
print("Analyzing distribution function data...")
f_fldline_avg = allocate_float(vpa.n,n_species,ntime)
for i ∈ 1:ntime
Expand All @@ -395,7 +410,7 @@ function analyze_pdf_data(ff, n_species, ntime, z, vpa, vth, evolve_ppar)
end
end
end
if evolve_ppar
if evolve_p
@. dens_moment *= vth
@. upar_moment *= vth^2
@. ppar_moment *= vth^3
Expand Down Expand Up @@ -1078,15 +1093,15 @@ coordinate at a point in space.
Inputs should depend only on vpa.
"""
function get_unnormalised_f_dzdt_1d(f, vpa_grid, density, upar, vth, evolve_density,
evolve_upar, evolve_ppar)
evolve_upar, evolve_p)

dzdt = vpagrid_to_vpa(vpa_grid, vth, upar, evolve_ppar, evolve_upar)
dzdt = vpagrid_to_vpa(vpa_grid, vth, upar, evolve_p, evolve_upar)

f_unnorm = get_unnormalised_f_1d(f, density, vth, evolve_density, evolve_ppar)
f_unnorm = get_unnormalised_f_1d(f, density, vth, evolve_density, evolve_p)

return f_unnorm, dzdt
end
function get_unnormalised_f_1d(f, density, vth, evolve_density, evolve_ppar)
function get_unnormalised_f_1d(f, density, vth, evolve_density, evolve_p)
n_v_dims = ndims(f) - ndims(density)
function add_v_dims(x)
if ndims(x) == 0
Expand All @@ -1097,8 +1112,13 @@ function get_unnormalised_f_1d(f, density, vth, evolve_density, evolve_ppar)
return result
end
end
if evolve_ppar
f_unnorm = f .* add_v_dims(density) ./ add_v_dims(vth)
if evolve_p
if size(f, 2) == 1
# 1V case
f_unnorm = f .* add_v_dims(density) ./ add_v_dims(vth)
else
f_unnorm = f .* add_v_dims(density) ./ add_v_dims(vth).^3
end
elseif evolve_density
f_unnorm = f .* add_v_dims(density)
else
Expand All @@ -1113,34 +1133,34 @@ Get the unnormalised distribution function and unnormalised ('lab space') coordi
Inputs should depend only on z and vpa.
"""
function get_unnormalised_f_coords_2d(f, z_grid, vpa_grid, density, upar, vth,
evolve_density, evolve_upar, evolve_ppar)
evolve_density, evolve_upar, evolve_p)

nvpa, nz = size(f)
z2d = zeros(nvpa, nz)
for iz ∈ 1:nz
z2d[:,iz] .= z_grid[iz]
end
v_parallel_2d = vpagrid_to_v_parallel_2d(vpa_grid, vth, upar, evolve_ppar, evolve_upar)
f_unnorm = get_unnormalised_f_2d(f, density, vth, evolve_density, evolve_ppar)
v_parallel_2d = vpagrid_to_v_parallel_2d(vpa_grid, vth, upar, evolve_p, evolve_upar)
f_unnorm = get_unnormalised_f_2d(f, density, vth, evolve_density, evolve_p)

return f_unnorm, z2d, v_parallel_2d
end
function vpagrid_to_v_parallel_2d(vpa_grid, vth, upar, evolve_ppar, evolve_upar)
function vpagrid_to_v_parallel_2d(vpa_grid, vth, upar, evolve_p, evolve_upar)
nvpa = length(vpa_grid)
nz = length(vth)
v_parallel_2d = zeros(nvpa, nz)
for iz ∈ 1:nz
@views v_parallel_2d[:,iz] .= vpagrid_to_vpa(vpa_grid, vth[iz], upar[iz],
evolve_ppar, evolve_upar)
evolve_p, evolve_upar)
end
return v_parallel_2d
end
function get_unnormalised_f_2d(f, density, vth, evolve_density, evolve_ppar)
function get_unnormalised_f_2d(f, density, vth, evolve_density, evolve_p)
f_unnorm = similar(f)
nz = size(f, 2)
for iz ∈ 1:nz
@views f_unnorm[:,iz] .= get_unnormalised_f_1d(f[:,iz], density[iz], vth[iz],
evolve_density, evolve_ppar)
evolve_density, evolve_p)
end
return f_unnorm
end
Expand Down
Loading