DC coupled PV and ElectricStorage #476

hdunham · 2025-02-05T16:07:04Z

TODO:

bit more documentation
move test into runtests.jl
check on consistency of using grid_charge_efficiency vs rectifier_efficiency_fraction

Added

Add option to model DC-coupled PV and ElectricStorage
- Add PV input dc_coupled_with_storage and ElectricStorage input dc_coupled, which both default to false
- ElectricStorage inputs inverter_efficiency_fraction and rectifier_efficiency_fraction define the DC-coupled inverter and PV input inf_eff is ignored
- Add sets in Techs struct (dc_coupled_with_storage and ac_coupled_with_storage) and StorageTypes struct (dc_coupled and ac_coupled)
- Modify constraints to account for AC and DC coupling

Removed

Don't expose ElectricStorage fields charge_efficiency, discharge_efficiency, and grid_charge_efficiency. It can lead to unexpected behavior when users provide these instead of the inputs they are calculated from (inverter_efficiency_fraction, rectifier_efficiency_fraction, and internal_efficiency_fraction).

This reverts commit a5072f5.

This reverts commit 78dab57.

was subtracting prod from load wrong; shortened outage

This reverts commit a5072f5.

adfarth · 2025-04-01T17:28:04Z

src/constraints/outage_constraints.jl

+        @constraint(m, [s in p.s.electric_utility.scenarios, tz in p.s.electric_utility.outage_start_time_steps, ts in p.s.electric_utility.outage_time_steps],
+            m[:dvStoragePower]["ElectricStorage"] >= m[:dvMGDischargeFromStorage][s, tz, ts]
+                + sum(m[:dvMGProductionToStorage][t, s, tz, ts] for t in p.techs.ac_coupled_with_storage)
+                + sum(m[:dvMGProductionToStorage][t, s, tz, ts] for t in p.techs.dc_coupled_with_storage) * p.s.storage.attr["ElectricStorage"].inverter_efficiency_fraction #TODO: remove term or leave for future flexibility?


If "ElectricStorage" in not in p.s.storage.types.dc_coupled, why would we need this here?

This is flow from dc coupled techs out inverter to ac coupled storage, which is currently is not allowed. With multiple storage and adding a bool input to say whether this flow is allowed, this term would be relevant, which is why I wrote that TODO.

adfarth · 2025-04-01T17:30:45Z

src/constraints/outage_constraints.jl

+    # Dispatch to and from MG electrical storage is no greater than power capacity
+    if "ElectricStorage" in p.s.storage.types.dc_coupled
+        @constraint(m, [s in p.s.electric_utility.scenarios, tz in p.s.electric_utility.outage_start_time_steps, ts in p.s.electric_utility.outage_time_steps],
+            m[:dvStoragePower]["ElectricStorage"] >= m[:dvMGDischargeFromStorage][s, tz, ts] / p.s.storage.attr["ElectricStorage"].inverter_efficiency_fraction


I think I'm confused about why the if statement is needed here, whereas in many of the other constraints you were able to avoid if statements by adding flows indexed on ac_coupled to flows index on dc_coupled. And, why does the discharge from storage need to be divided by inverter_efficiency_fraction for the dc-coupled case? Shouldn't the discharge from storage be treated the same between ac- and dc-coupled cases, because in both cases the discharge passes through an inverter (and isn't that already captured in dvMGDischargeFromStorage)?

I could be misunderstanding some of this, but just checking!

src/constraints/outage_constraints.jl

adfarth · 2025-04-01T17:34:59Z

src/constraints/outage_constraints.jl

+    # Future development could make this an option by adding bool inputs an updating load balance/battery dispatch constraints to include these flows
+    if "ElectricStorage" in p.s.storage.types.dc_coupled
+        # Don't let AC coupled elec techs charge DC coupled battery. 
+        for ts in 1:p.s.site.min_resil_time_steps


So, if you have a run with wind (AC production) and PV that is dc-coupled with storage, the wind cannot charge the battery? Do you have some test cases in this realm in runtests?

Yes that's the current set up. Could add an input bool to allow this though. Noting this as a test to add.

src/constraints/outage_constraints.jl

adfarth · 2025-04-01T18:18:05Z

src/constraints/outage_constraints.jl

+            - m[:dvMGCurtail][t, s, tz, ts]
+            for t in p.techs.dc_coupled_with_storage
+        )
+        # (rectifier direction, though currently dvMGProductionToStorage for AC coupled techs must be 0)


Is the constraint above the only thing limiting this? Is this typical operational behavior for a system with DC-coupled PV + BESS and an AC-producing tech like wind or CHP?

src/constraints/storage_constraints.jl

adfarth · 2025-04-01T19:05:36Z

src/constraints/storage_constraints.jl

    end
+
+    # Future development could make this an option by adding bool inputs an updating load balance/battery dispatch constraints to include these flows
+    if b in p.s.storage.types.dc_coupled


Same comment from me here as in "outage_constraints.jl"

…rage=true

adfarth · 2025-04-01T21:58:41Z

General comment on things to document:

AC-producing techs cannot currently charge a BESS that is DC coupled with PV. Maybe make a note on what dev would be needed to change this?
Notes in code on whether dv's are DC or AC units
Description of overall configuration as it compares to current config (PV and BESS have separate inverter systems)
Notes on how to consider modifying costs (lower PV costs, raise ElectricStorage costs when DC-Coupled)

src/core/pv.jl

src/constraints/operating_reserve_constraints.jl

adfarth · 2025-11-19T19:11:22Z

src/results/electric_storage.jl

    r["size_kwh"] = round(value(m[Symbol("dvStorageEnergy"*_n)][b]), digits=2)
    r["size_kw"] = round(value(m[Symbol("dvStoragePower"*_n)][b]), digits=2)
-
+    if b in p.s.storage.types.dc_coupled


Could you add dc_coupled_inverter_size_kw to the results markdown section above?

adfarth · 2025-11-19T19:14:21Z

src/results/electric_storage.jl

@@ -23,7 +23,10 @@ function add_electric_storage_results(m::JuMP.AbstractModel, p::REoptInputs, d::
    r = Dict{String, Any}()


@hdunham is it accurate to say that the ElectricStorage size_kw (and dvStoragePower) is AC if dc_coupled is false and is DC if dc_coupled is true? Or is it always AC? Could you add whichever is true to the help text above?

adfarth · 2025-11-19T19:41:47Z

src/constraints/operating_reserve_constraints.jl

    @constraint(m, [b in p.s.storage.types.elec, ts in p.time_steps_without_grid],
-        m[Symbol("dvOpResFromBatt"*_n)][b,ts] <= m[Symbol("dvStoragePower"*_n)][b] - m[Symbol("dvDischargeFromStorage"*_n)][b,ts] / p.s.storage.attr[b].discharge_efficiency
+        m[Symbol("dvOpResFromBatt"*_n)][b,ts] <= 
+        m[Symbol("dvStoragePower"*_n)][b] * (b in p.s.storage.types.dc_coupled ? p.s.storage.attr[b].inverter_efficiency_fraction : 1.0)


Just to confirm my understanding here, dvStoragePower is AC if dc_coupled is false but is DC otherwise and thus needs to be converted from DC to AC with inverter_efficiency_fraction?

adfarth · 2025-11-19T19:46:49Z

src/constraints/operating_reserve_constraints.jl

-                        m[:ProductionToLoadOR][t,ts]) * (1 - p.techs_operating_reserve_req_fraction[t])
+        m[Symbol("dvOpResFromTechs"*_n)][t,ts] <= (
+            p.production_factor[t, ts] * p.levelization_factor[t] * m[Symbol("dvSize"*_n)][t] 
+                * (t in p.techs.dc_coupled_with_storage ? p.s.storage.attr["ElectricStorage"].rectifier_efficiency_fraction : 1.0) 


I think this should also be inverter_ and not rectifier_ efficiency?

hdunham added 30 commits January 14, 2025 11:18

set up testing

3931a81

add results to compare to

41af908

Update debugging_tests.jl

f6afbd0

add dc_coupled_with_battery option to PV model

94192fe

rename result file

001a6ee

update testing

9e8f5e4

warn that PV inv_eff ignored if dc_coupled_with_electric_storage

f79e4a6

create dc and ac coupled tech sets

584bc3f

store dc_(dis)charge_efficiency in ElectricStorage

1d655ce

test checkpoint

195bd5e

return dc prod if dc_coupled_with_battery in call_pvwatts_api

1be7e8c

correct call_pvwatts_api arg issues

8b4ad0b

REVERT LATER temp return ac prod for dc coupled

5bdd53d

create dvDCCoupledTechStorageInverterSize

a741853

create DV for dc coupled inverter size

caa574c

add dc_couple_inverter_size_kw to ElectricStorage results

8ee32a6

rename to dc_coupled_with_storage

42b00b3

make dc coupled constraints compatible with multinode

28f36a7

call dc coupled constraints if dc coupled tech not not empty

a61629e

update consrtaint on dvDCCoupledTechStorageInverterSize

ee0e7d0

don't let AC coupled techs charge dc coupled bess

a5072f5

update test scenarios

adc276d

update testing

7b6c0fe

update testing

ce59f4f

Revert "don't let AC coupled techs charge dc coupled bess"

78dab57

This reverts commit a5072f5.

Update debugging_tests.jl

68cda37

testing checkpoint

266f472

Reapply "don't let AC coupled techs charge dc coupled bess"

3f25088

This reverts commit 78dab57.

update testing

878f864

was subtracting prod from load wrong; shortened outage

Revert "don't let AC coupled techs charge dc coupled bess"

5eab825

This reverts commit a5072f5.

adfarth reviewed Apr 1, 2025

View reviewed changes

src/constraints/outage_constraints.jl Outdated Show resolved Hide resolved

adfarth reviewed Apr 1, 2025

View reviewed changes

src/constraints/outage_constraints.jl Show resolved Hide resolved

hdunham added 2 commits April 1, 2025 11:49

rename to dvHybridInverterSizeAC

344c521

todo comment

9a95fda

adfarth reviewed Apr 1, 2025

View reviewed changes

src/constraints/storage_constraints.jl Show resolved Hide resolved

adfarth reviewed Apr 1, 2025

View reviewed changes

hdunham added 4 commits April 1, 2025 13:07

default ElectricStorage.dc_coupled=true if any PV.dc_coupled_with_sto…

8536f9c

…rage=true

fix syntax

c2c1bfd

rm dc_coupled input from tests to test default behavior

3bf2cd9

test no changes

72fa189

hdunham and others added 8 commits June 10, 2025 10:52

update constraint numbering

4a2a579

update constraint numbering/comment

89c432a

use time_step_wrap_around more places where needed

b99841f

update comment

526654e

Apply rectifier efficiency when dc coupled in OR constraints 1,4

1034c6f

handle dc coupled stor in OR constr 3

72dc85c

docstring notes about cost defaults being ac coupled

ec3dea1

small helptext update

542f67a

adfarth reviewed Nov 19, 2025

View reviewed changes

src/core/pv.jl Show resolved Hide resolved

adfarth reviewed Nov 19, 2025

View reviewed changes

src/constraints/operating_reserve_constraints.jl Show resolved Hide resolved

adfarth reviewed Nov 19, 2025

View reviewed changes

minor comment mods

7eb1b32

		@@ -23,7 +23,10 @@ function add_electric_storage_results(m::JuMP.AbstractModel, p::REoptInputs, d::
		r = Dict{String, Any}()

DC coupled PV and ElectricStorage #476

Are you sure you want to change the base?

DC coupled PV and ElectricStorage #476

Uh oh!

Conversation

hdunham commented Feb 5, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Added

Removed

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

adfarth commented Apr 1, 2025

Uh oh!

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

adfarth Nov 19, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

hdunham commented Feb 5, 2025 •

edited

Loading

adfarth Nov 19, 2025 •

edited

Loading