[Bug]: vre_stor resources not respecting availability profiles in multistage?

[joshdr83]

Bug description

It looks like vre_stor resources are not respecting their availability profiles when running in multistage? I took the vre_stor resources from the 7_three_zones_w_colocated_VRE_storage example system and put them in each stage of the 6b_three_zones_w_multistage_VRE_storage example system to get the attached 6b... system and I seem to be getting the below behavior, which looks like the solar component of the vre_stor system is just putting out a constant output?

6b_three_zones_w_multistage_VRE_storage.zip

Environment and Version

MacOS, Julia 1.11.7, GenX 0.4.5

Relevant error messages

Additional context

No response

[NLaws]

From what I can tell, there is no constraint on vP_SOLAR, the power output of the co-located solar resources, like the curtailable_variable_renewable constraint:

GenX.jl/src/model/resources/curtailable_variable_renewable/curtailable_variable_renewable.jl

Lines 69 to 76 in c1000b9

	# Maximum power generated per hour by renewable generators must be less than
	# sum of product of hourly capacity factor for each bin times its the bin installed capacity
	# Note: inequality constraint allows curtailment of output below maximum level.
	@constraint(EP,
	[t = 1:T],
	EP[:vP][y,t]<=sum(inputs["pP_Max"][yy, t] * EP[:eTotalCap][yy]
	for yy in VRE_BINS))
	end

Am I missing something @lbonaldo @sambuddhac ?

[lbonaldo]

Hi @joshdr83, hi @NLaws, thanks for opening this issue and for sharing the input data.

I reviewed the code and the inputs, and it looks like the variabilities for solar resources in Vre_and_stor_solar_variability.csv have all 1s. This is most likely due to the TDR clustering. When running TDR with vre_stor resources, all variabilities should also be in the Generators_variability.csv file, as TDR will use this file to run the clustering and then create the Vre_and_stor_solar_variability.csv and Vre_and_stor_wind_variability.csv files.

The code then uses that input data to set the following constraints:

# Constraint 4: PV Generation
@constraint(EP,
    cSolarGenMaxS[y in SOLAR, t = 1:T],
    EP[:eSolarGenMaxS][y, t]<=inputs["pP_Max_Solar"][y, t] * EP[:eTotalCap_SOLAR][y]
)

# Constraint 5: Wind Generation
@constraint(EP,
    cWindGenMaxW[y in WIND, t = 1:T],
    EP[:eWindGenMaxW][y, t]<=inputs["pP_Max_Wind"][y, t] * EP[:eTotalCap_WIND][y]
)

I will update the documentation to make this clearer. In the meantime, please let me know if this fixes the issue. Thank you! And happy to iterate further if needed.

[NLaws]

Thanks @lbonaldo ! Does this mean that we do not need to define inputs/system/Vre_and_stor_solar_variability.csv? (We just need to put the values in the inputs/system/Generators_variability.csv).

[lbonaldo]

If you're running the TDR module (e.g., by setting TimeDomainReduction: 1 in genx_settings.yml), then yes, those files will be created by the TDR module for you. On the other hand, if the input data is already clustered (and you're running GenX with TimeDomainReduction: 0), then it should go in inputs/system/Vre_and_stor_solar_variability.csv and inputs/system/Vre_and_stor_wind_variability.csv.

[lbonaldo]

Would it be okay if I close this issue and open another one to update the documentation?

[joshdr83]

That would be fine with me, I think I was able to get it to work.Should the change be to just include all the availability files in the same place instead of having the `vre_stor` ones in a different place sometimes and sometimes not?Thanks!Joshua Rhodes Sent from my iPadOn Nov 12, 2025, at 8:56 AM, Luca Bonaldo ***@***.***> wrote:lbonaldo left a comment (GenXProject/GenX.jl#879) Would it be okay if I close this issue and open another one to update the documentation? —Reply to this email directly, view it on GitHub, or unsubscribe.You are receiving this because you were mentioned.Message ID: ***@***.***>

[lbonaldo]

When building the model, GenX will always use also the vre_stor ones together with the "default" Generators_variability.
However, when running TDR before GenX, the current TDR module is expecting to receive all availabilities in the same file. It will then cluster all the time series and create separate files for vre_stor. I understand this may not be ideal, but just a heads-up that there’s a plan to migrate soon to a new version of the TDR module, and I will make sure to include all availailibites as inputs to the updated version of TDR.

[lbonaldo]

Thank you again, @joshdr83 and @NLaws, for opening this issue. We have updated the documentation following your suggestions, and clarified how TDR handles VreStorage variability data. I’m closing this issue for now, but we’ll work on integrating GenX with the new version of the TDR module to make the code more straightforward. Thank you for your patience.