Output correct CRM values for THERM plants with maintenance #589
Conversation
| eff_cap = thermal_plant_effective_capacity.(Ref(EP), Ref(inputs), resources, Ref(capres_zone), Ref(timesteps)) | ||
| return reduce(hcat, eff_cap) |
There was a problem hiding this comment.
Ref is a part of Julia base which has not before been used in GenX. It allows vectorizing a function over arguments other than the Ref'd ones.
| function thermal_plant_effective_capacity(EP::Model, inputs::Dict, y, capres_zone::Int) | ||
| T = inputs["T"] | ||
| timesteps = collect(1:T) | ||
| return thermal_plant_effective_capacity(EP, inputs, y, capres_zone, timesteps) | ||
| end | ||
|
|
There was a problem hiding this comment.
This function can be used for y::Int or y::Vector{Int}.
In the first case it outputs a Vector,{Float64} in the second case it outputs a Matrix{Float64}.
This a convenience function to automatically get the effective capacity over every time period.
There was a problem hiding this comment.
This logic is very neat.
| if has_maintenance(inputs) && y in resources_with_maintenance(dfGen) | ||
| resource_component = dfGen[y, :Resource] | ||
| cap_size = dfGen[y, :Cap_Size] | ||
| down_var = EP[Symbol(maintenance_down_name(resource_component))] | ||
| vDOWN = value.(down_var[timesteps]) | ||
| effective_capacity = effective_capacity .- capresfactor(y, capres_zone) * vDOWN * cap_size | ||
| end |
There was a problem hiding this comment.
The actual accounting logic is all specified here, in a function that deals with one plant at a time.
Later, this will allow more detailed logic for FUSION plants, for example.
| @@ -52,9 +54,7 @@ function write_reserve_margin_revenue(path::AbstractString, inputs::Dict, setup: | |||
| tempresrev[DC_CHARGE] .-= dfVRE_STOR[(dfVRE_STOR.STOR_DC_CHARGE.!=0), sym_vs] .* ((value.(EP[:vCAPRES_DC_CHARGE][DC_CHARGE, :]).data ./ dfVRE_STOR[(dfVRE_STOR.STOR_DC_CHARGE.!=0), :EtaInverter]) * (dual.(EP[:cCapacityResMargin][i, :]))) | |||
There was a problem hiding this comment.
We could similarly use weighted_price to replace calls to dual.(EP[:cCapResMargin][i, :]) on other lines but I felt that would be better as a separate "refactor" PR.
|
The new Ideally all the mathematical logic would be also the same as gets put into the CRM expressions during model generation! But all the expressions for output need |
cfe316
force-pushed
the
feature-output-correct-crm-for-maintenance
branch
from
9676021 to
6a4f821
Compare
| function thermal_plant_effective_capacity(EP::Model, inputs::Dict, y, capres_zone::Int) | ||
| T = inputs["T"] | ||
| timesteps = collect(1:T) | ||
| return thermal_plant_effective_capacity(EP, inputs, y, capres_zone, timesteps) | ||
| end | ||
|
|
There was a problem hiding this comment.
This logic is very neat.
| capresfactor(y, capres) = dfGen[y, Symbol("CapRes_$capres")] | ||
| eTotalCap = value.(EP[:eTotalCap][y]) | ||
|
|
||
| effective_capacity = capresfactor(y, capres_zone) * eTotalCap * one.(timesteps) |
There was a problem hiding this comment.
you could also use the fill method to create a vector of identical values.
cfe316
force-pushed
the
feature-output-correct-crm-for-maintenance
branch
from
e843554 to
f8fa6da
Compare
Account for the fact that THERM plants have provide no "capacity" while they are under maintenance.