Improve agg_p_nom_limits configuration

config/config.default.yaml

@@ -98,7 +98,6 @@ electricity:
   co2limit_enable: false
   co2limit: 7.75e+7
   co2base: 1.487e+9
-  agg_p_nom_limits: data/agg_p_nom_minmax.csv
 
   operational_reserve:
     activate: false
@@ -758,6 +757,14 @@ solving:
     linearized_unit_commitment: true
     horizon: 365
 
+  agg_p_nom_limits:
+    agg_offwind: false
+    include_existing: false
+    years:
+      2030: data/agg_p_nom_minmax.csv
+      2040: data/agg_p_nom_minmax.csv
+      2050: data/agg_p_nom_minmax.csv
+
   constraints:
     CCL: false
     EQ: false

doc/configtables/electricity.csv

@@ -5,7 +5,6 @@ gaslimit,MWhth,float or false,Global gas usage limit
 co2limit_enable,bool,true or false,Add an overall absolute carbon-dioxide emissions limit configured in ``electricity: co2limit``.
 co2limit,:math:`t_{CO_2-eq}/a`,float,Cap on total annual system carbon dioxide emissions
 co2base,:math:`t_{CO_2-eq}/a`,float,Reference value of total annual system carbon dioxide emissions if relative emission reduction target is specified in ``{opts}`` wildcard.
-agg_p_nom_limits,file,path,Reference to ``.csv`` file specifying per carrier generator nominal capacity constraints for individual countries if ``'CCL'`` is in ``{opts}`` wildcard. Defaults to ``data/agg_p_nom_minmax.csv``.
 operational_reserve,,,Settings for reserve requirements following `GenX <https://genxproject.github.io/GenX/dev/core/#Reserves>`_
 ,,,
 -- activate,bool,true or false,Whether to take operational reserve requirements into account during optimisation

doc/configtables/solving.csv

@@ -14,6 +14,10 @@ options,,,
 -- transmission_losses,int,[0-9],"Add piecewise linear approximation of transmission losses based on n tangents. Defaults to 0, which means losses are ignored."
 -- linearized_unit_commitment,bool,"{'true','false'}",Whether to optimise using the linearized unit commitment formulation.
 -- horizon,--,int,Number of snapshots to consider in each iteration. Defaults to 100.
+agg_p_nom_limits,,,Configure per carrier generator nominal capacity constraints for individual countries if ``'CCL'`` is in ``{opts}`` wildcard.
+-- agg_offwind,bool,"{'true','false'}",Aggregate together all the types of offwind when writing the constraint. Default is false.
+-- include_existing,bool,"{'true','false'}",Take existing capacities into account when writing the constraint. Default is false.
+-- years,-,Dictionary with planning horizons as key and path as value,Reference to ``.csv`` file for each planning horizon. Defaults to ``data/agg_p_nom_minmax.csv``.
 constraints ,,,
 -- CCL,bool,"{'true','false'}",Add minimum and maximum levels of generator nominal capacity per carrier for individual countries. These can be specified in the file linked at ``electricity: agg_p_nom_limits`` in the configuration. File defaults to ``data/agg_p_nom_minmax.csv``.
 -- EQ,bool/string,"{'false',`n(c| )``; i.e. ``0.5``-``0.7c``}",Require each country or node to on average produce a minimal share of its total consumption itself. Example: ``EQ0.5c`` demands each country to produce on average at least 50% of its consumption; ``EQ0.5`` demands each node to produce on average at least 50% of its consumption.

doc/release_notes.rst

@@ -181,6 +181,16 @@ Upcoming Release
 
 * Fix custom busmap read in `cluster_network`.
 
+* Improved the behaviour of `agg_p_nom_limits`:
+
+  - Moved the associated configuration to `solving`. This allows *Snakemake* to correctly decide which rules to run when the configuration changes.
+
+  - Added the ability to enable aggregation of all *offwind* types (*offwind-ac* and *offwind-dc*) when writing the constraint.
+
+  - Added the possibility to take existing capacities into account when writing the constraint.
+
+  - Added the possibility to have a different file for each planning horizon.
+
 PyPSA-Eur 0.10.0 (19th February 2024)
 =====================================
 

scripts/solve_network.py

@@ -422,23 +422,54 @@ def add_CCL_constraints(n, config):
         agg_p_nom_limits: data/agg_p_nom_minmax.csv
     """
     agg_p_nom_minmax = pd.read_csv(
-        config["electricity"]["agg_p_nom_limits"], index_col=[0, 1]
+        config["solving"]["agg_p_nom_limits"]["years"][int(snakemake.wildcards.planning_horizons)], index_col=[0, 1]
     )
     logger.info("Adding generation capacity constraints per carrier and country")
     p_nom = n.model["Generator-p_nom"]
 
     gens = n.generators.query("p_nom_extendable").rename_axis(index="Generator-ext")
-    grouper = pd.concat([gens.bus.map(n.buses.country), gens.carrier])
+    if config["solving"]["agg_p_nom_limits"]["agg_offwind"]:
+        rename_offwind = {"offwind-ac": "offwind-all", "offwind-dc": "offwind-all", "offwind": "offwind-all"}
+        gens = gens.replace(rename_offwind)
+    grouper = pd.concat([gens.bus.map(n.buses.country), gens.carrier], axis=1)
     lhs = p_nom.groupby(grouper).sum().rename(bus="country")
 
-    minimum = xr.DataArray(agg_p_nom_minmax["min"].dropna()).rename(dim_0="group")
+    if config["solving"]["agg_p_nom_limits"]["include_existing"]:
+        gens_cst = n.generators.query("~p_nom_extendable").rename_axis(index="Generator-cst")
+        gens_cst = gens_cst[
+            (gens_cst["build_year"] + gens_cst["lifetime"]) >= int(snakemake.wildcards.planning_horizons)]
+        if config["solving"]["agg_p_nom_limits"]["agg_offwind"]:
+            gens_cst = gens_cst.replace(rename_offwind)
+        rhs_cst = (
+            pd.concat([gens_cst.bus.map(n.buses.country), gens_cst[["carrier", "p_nom"]]], axis=1)
+            .groupby(["bus", "carrier"]).sum()
+        )
+        rhs_cst.index = rhs_cst.index.rename({"bus": "country"})
+        rhs_min = agg_p_nom_minmax["min"].dropna()
+        idx_min = rhs_min.index.join(rhs_cst.index, how="left")
+        rhs_min = rhs_min.reindex(idx_min).fillna(0)
+        rhs = (rhs_min - rhs_cst.reindex(idx_min).fillna(0).p_nom).dropna()
+        rhs[rhs < 0] = 0
+        minimum = xr.DataArray(rhs).rename(dim_0="group")
+    else:
+        minimum = xr.DataArray(agg_p_nom_minmax["min"].dropna()).rename(dim_0="group")
+
     index = minimum.indexes["group"].intersection(lhs.indexes["group"])
     if not index.empty:
         n.model.add_constraints(
             lhs.sel(group=index) >= minimum.loc[index], name="agg_p_nom_min"
         )
 
-    maximum = xr.DataArray(agg_p_nom_minmax["max"].dropna()).rename(dim_0="group")
+    if config["solving"]["agg_p_nom_limits"]["include_existing"]:
+        rhs_max = agg_p_nom_minmax["max"].dropna()
+        idx_max = rhs_max.index.join(rhs_cst.index, how="left")
+        rhs_max = rhs_max.reindex(idx_max).fillna(0)
+        rhs = (rhs_max - rhs_cst.reindex(idx_max).fillna(0).p_nom).dropna()
+        rhs[rhs < 0] = 0
+        maximum = xr.DataArray(rhs).rename(dim_0="group")
+    else:
+        maximum = xr.DataArray(agg_p_nom_minmax["max"].dropna()).rename(dim_0="group")
+
     index = maximum.indexes["group"].intersection(lhs.indexes["group"])
     if not index.empty:
         n.model.add_constraints(