diff --git a/pyproject.toml b/pyproject.toml index 9e778e694..0a34847c0 100644 --- a/pyproject.toml +++ b/pyproject.toml @@ -35,7 +35,11 @@ dependencies = [ "xlrd", "mypy-extensions", "pypubsub", - "tomlkit" + "tomlkit", + "duckdb", + "fsspec", + "sqlalchemy", + "duckdb-engine" ] dynamic = ["version"] diff --git a/src/muse/data/example/default_new_input/agent_objectives.csv b/src/muse/data/example/default_new_input/agent_objectives.csv new file mode 100644 index 000000000..331c649c9 --- /dev/null +++ b/src/muse/data/example/default_new_input/agent_objectives.csv @@ -0,0 +1,3 @@ +agent_id,objective,objective_data,objective_sort +Agent1,LCOE,1,TRUE +Agent2,LCOE,1,TRUE diff --git a/src/muse/data/example/default_new_input/agent_pairs.csv b/src/muse/data/example/default_new_input/agent_pairs.csv new file mode 100644 index 000000000..172632275 --- /dev/null +++ b/src/muse/data/example/default_new_input/agent_pairs.csv @@ -0,0 +1,2 @@ +id,new_agent_id,retrofit_agent_id,quantity +A1,Agent1,Agent2,1 diff --git a/src/muse/data/example/default_new_input/agent_regions.csv b/src/muse/data/example/default_new_input/agent_regions.csv new file mode 100644 index 000000000..6a39852ea --- /dev/null +++ b/src/muse/data/example/default_new_input/agent_regions.csv @@ -0,0 +1,2 @@ +agent_pair_id,region_id +A1,R1 diff --git a/src/muse/data/example/default_new_input/agents.csv b/src/muse/data/example/default_new_input/agents.csv new file mode 100644 index 000000000..a95e5b9dd --- /dev/null +++ b/src/muse/data/example/default_new_input/agents.csv @@ -0,0 +1,3 @@ +id,description,type,maturity_threshold,annual_cost_limit,search_rule,decision_rule +Agent1,New agent for A1,new,-1,inf,all,single +Agent2,Retrofit agent for A1,retrofit,-1,inf,all,single diff --git a/src/muse/data/example/default_new_input/assets.csv b/src/muse/data/example/default_new_input/assets.csv new file mode 100644 index 000000000..8648bb891 --- /dev/null +++ b/src/muse/data/example/default_new_input/assets.csv @@ -0,0 +1,8 @@ +agent_id,process_id,region_id,year,capacity +Agent2,gassupply1,R1,2020,15 +Agent2,gassupply1,R1,2025,15 +Agent2,gassupply1,R1,2030,7.5 +Agent2,gasCCGT,R1,2020,1 +Agent2,gasCCGT,R1,2025,1 +Agent2,gasboiler,R1,2020,10 +Agent2,gasboiler,R1,2025,5 diff --git a/src/muse/data/example/default_new_input/commodities.csv b/src/muse/data/example/default_new_input/commodities.csv new file mode 100644 index 000000000..09857851f --- /dev/null +++ b/src/muse/data/example/default_new_input/commodities.csv @@ -0,0 +1,6 @@ +name,description,type,unit +electricity,Electricity,energy,PJ +gas,Gas,energy,PJ +heat,Heat,energy,PJ +wind,Wind,energy,PJ +CO2f,Carbon dioxide,energy,kt diff --git a/src/muse/data/example/default_new_input/commodity_costs.csv b/src/muse/data/example/default_new_input/commodity_costs.csv new file mode 100644 index 000000000..0a64542b3 --- /dev/null +++ b/src/muse/data/example/default_new_input/commodity_costs.csv @@ -0,0 +1,58 @@ +commodity_id,region_id,year,value +electricity,R1,2010,14.81481472 +electricity,R1,2015,17.89814806 +electricity,R1,2020,19.5 +electricity,R1,2025,21.93518528 +electricity,R1,2030,26.50925917 +electricity,R1,2035,26.51851861 +electricity,R1,2040,23.85185194 +electricity,R1,2045,23.97222222 +electricity,R1,2050,24.06481472 +electricity,R1,2055,25.3425925 +electricity,R1,2060,25.53703694 +electricity,R1,2065,25.32407417 +electricity,R1,2070,23.36111111 +electricity,R1,2075,22.27777778 +electricity,R1,2080,22.25925917 +electricity,R1,2085,22.17592583 +electricity,R1,2090,22.03703694 +electricity,R1,2095,21.94444444 +electricity,R1,2100,21.39814806 +gas,R1,2010,6.6759 +gas,R1,2015,6.914325 +gas,R1,2020,7.15275 +gas,R1,2025,8.10645 +gas,R1,2030,9.06015 +gas,R1,2035,9.2191 +gas,R1,2040,9.37805 +gas,R1,2045,9.193829337 +gas,R1,2050,9.009608674 +gas,R1,2055,8.832625604 +gas,R1,2060,8.655642534 +gas,R1,2065,8.485612708 +gas,R1,2070,8.315582883 +gas,R1,2075,8.152233126 +gas,R1,2080,7.988883368 +gas,R1,2085,7.831951236 +gas,R1,2090,7.675019103 +gas,R1,2095,7.524252461 +gas,R1,2100,7.373485819 +CO2f,R1,2010,0 +CO2f,R1,2015,0.052913851 +CO2f,R1,2020,0.08314119 +CO2f,R1,2025,0.120069795 +CO2f,R1,2030,0.156998399 +CO2f,R1,2035,0.214877567 +CO2f,R1,2040,0.272756734 +CO2f,R1,2045,0.35394801 +CO2f,R1,2050,0.435139285 +CO2f,R1,2055,0.542365578 +CO2f,R1,2060,0.649591871 +CO2f,R1,2065,0.780892624 +CO2f,R1,2070,0.912193378 +CO2f,R1,2075,1.078321687 +CO2f,R1,2080,1.244449995 +CO2f,R1,2085,1.4253503 +CO2f,R1,2090,1.606250604 +CO2f,R1,2095,1.73877515 +CO2f,R1,2100,1.871299697 diff --git a/src/muse/data/example/default_new_input/commodity_trade.csv b/src/muse/data/example/default_new_input/commodity_trade.csv new file mode 100644 index 000000000..092dd1559 --- /dev/null +++ b/src/muse/data/example/default_new_input/commodity_trade.csv @@ -0,0 +1 @@ +commodity_id,region_id,year,import,export diff --git a/src/muse/data/example/default_new_input/demand.csv b/src/muse/data/example/default_new_input/demand.csv new file mode 100644 index 000000000..b26c1b54d --- /dev/null +++ b/src/muse/data/example/default_new_input/demand.csv @@ -0,0 +1,3 @@ +commodity_id,region_id,year,demand +heat,R1,2020,10 +heat,R1,2050,30 diff --git a/src/muse/data/example/default_new_input/demand_slicing.csv b/src/muse/data/example/default_new_input/demand_slicing.csv new file mode 100644 index 000000000..6877d5663 --- /dev/null +++ b/src/muse/data/example/default_new_input/demand_slicing.csv @@ -0,0 +1,7 @@ +commodity_id,region_id,year,timeslice,fraction +heat,R1,,night,0.1 +heat,R1,,morning,0.15 +heat,R1,,afternoon,0.1 +heat,R1,,early-peak,0.15 +heat,R1,,late-peak,0.3 +heat,R1,,evening,0.2 diff --git a/src/muse/data/example/default_new_input/process_availabilities.csv b/src/muse/data/example/default_new_input/process_availabilities.csv new file mode 100644 index 000000000..1386d9db6 --- /dev/null +++ b/src/muse/data/example/default_new_input/process_availabilities.csv @@ -0,0 +1,6 @@ +process_id,region_id,year,timeslice,lim_type,value +gassupply1,,,ALL,UP,0.9 +gasCCGT,,,ALL,UP,0.9 +windturbine,,,ALL,UP,0.4 +gasboiler,,,ALL,UP,1 +heatpump,,,ALL,UP,1 diff --git a/src/muse/data/example/default_new_input/process_flows.csv b/src/muse/data/example/default_new_input/process_flows.csv new file mode 100644 index 000000000..1c602f2d1 --- /dev/null +++ b/src/muse/data/example/default_new_input/process_flows.csv @@ -0,0 +1,12 @@ +process_id,commodity_id,region_id,year,flow +gassupply1,gas,,,1 +gasCCGT,gas,,,-1.67 +gasCCGT,electricity,,,1 +gasCCGT,CO2f,,,91.67 +windturbine,wind,,,-1 +windturbine,electricity,,,1 +gasboiler,gas,,,-1.16 +gasboiler,heat,,,1 +gasboiler,CO2f,,,64.71 +heatpump,electricity,,,-0.4 +heatpump,heat,,,1 diff --git a/src/muse/data/example/default_new_input/process_parameters.csv b/src/muse/data/example/default_new_input/process_parameters.csv new file mode 100644 index 000000000..5f162962a --- /dev/null +++ b/src/muse/data/example/default_new_input/process_parameters.csv @@ -0,0 +1,6 @@ +process_id,region_id,year,cap_par,cap_exp,fix_par,fix_exp,var_par,var_exp,max_capacity_addition,max_capacity_growth,total_capacity_limit,life,scaling_size,efficiency,discount_rate +gassupply1,,,0,1,0,1,2.55,1,5,1,60,35,0.00000189,86,0.1 +gasCCGT,,,23.78234399,1,0,1,0,1,2,1,60,35,0.00000189,86,0.1 +windturbine,,,36.30771182,1,0,1,0,1,2,1,60,25,0.00000189,86,0.1 +gasboiler,,,3.8,1,0,1,0,1,10,0.02,60,10,0.00000189,86,0.1 +heatpump,,,8.866667,1,0,1,0,1,10,0.02,60,10,0.00000189,86,0.1 diff --git a/src/muse/data/example/default_new_input/process_regions.csv b/src/muse/data/example/default_new_input/process_regions.csv new file mode 100644 index 000000000..8700ece5a --- /dev/null +++ b/src/muse/data/example/default_new_input/process_regions.csv @@ -0,0 +1,6 @@ +process_id,region_id +gassupply1,R1 +gasCCGT,R1 +windturbine,R1 +gasboiler,R1 +heatpump,R1 diff --git a/src/muse/data/example/default_new_input/processes.csv b/src/muse/data/example/default_new_input/processes.csv new file mode 100644 index 000000000..e653e3ed6 --- /dev/null +++ b/src/muse/data/example/default_new_input/processes.csv @@ -0,0 +1,6 @@ +id,sector_id,type,fuel,end_use,level +gassupply1,gas,energy,gas,gas,fixed +gasCCGT,power,energy,gas,electricity,fixed +windturbine,power,energy,wind,electricity,fixed +gasboiler,residential,energy,gas,heat,fixed +heatpump,residential,energy,electricity,heat,fixed diff --git a/src/muse/data/example/default_new_input/regions.csv b/src/muse/data/example/default_new_input/regions.csv new file mode 100644 index 000000000..1ce17d1ce --- /dev/null +++ b/src/muse/data/example/default_new_input/regions.csv @@ -0,0 +1,2 @@ +id,description +R1,Region 1 diff --git a/src/muse/data/example/default_new_input/sectors.csv b/src/muse/data/example/default_new_input/sectors.csv new file mode 100644 index 000000000..7488adac9 --- /dev/null +++ b/src/muse/data/example/default_new_input/sectors.csv @@ -0,0 +1,4 @@ +id,description +gas,Gas sector +power,Power sector +residential,Residential sector diff --git a/src/muse/data/example/default_new_input/time_slices.csv b/src/muse/data/example/default_new_input/time_slices.csv new file mode 100644 index 000000000..376022d96 --- /dev/null +++ b/src/muse/data/example/default_new_input/time_slices.csv @@ -0,0 +1,7 @@ +season,day,time_of_day,fraction +all,all,night,0.1667 +all,all,morning,0.1667 +all,all,afternoon,0.1667 +all,all,early-peak,0.1667 +all,all,late-peak,0.1667 +all,all,evening,0.1667 diff --git a/src/muse/examples.py b/src/muse/examples.py index 189a82c85..6bfaa6e7e 100644 --- a/src/muse/examples.py +++ b/src/muse/examples.py @@ -52,6 +52,7 @@ "multiple_agents", "minimum_service", "trade", + "default_new_input", ] @@ -137,6 +138,8 @@ def copy_model( _copy_minimum_service(path) elif name.lower() == "trade": _copy_trade(path) + elif name.lower() == "default_new_input": + _copy_default_new_input(path) return path @@ -316,6 +319,12 @@ def update_lpsolver(data): modify_toml(path / "settings.toml", update_lpsolver) +def _copy_default_new_input(path: Path): + from shutil import copytree + + copytree(example_data_dir() / "default_new_input", path) + + def _copy_default_timeslice(path: Path): copytree(example_data_dir() / "default_timeslice", path) diff --git a/src/muse/new_input/readers.py b/src/muse/new_input/readers.py new file mode 100644 index 000000000..5c2dddde2 --- /dev/null +++ b/src/muse/new_input/readers.py @@ -0,0 +1,89 @@ +import numpy as np +import xarray as xr +from sqlalchemy import CheckConstraint, ForeignKey +from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column + + +class TableBase(DeclarativeBase): + pass + + +class Regions(TableBase): + __tablename__ = "regions" + + name: Mapped[str] = mapped_column(primary_key=True) + + +class Commodities(TableBase): + __tablename__ = "commodities" + + name: Mapped[str] = mapped_column(primary_key=True) + type: Mapped[str] = mapped_column( + CheckConstraint("type IN ('energy', 'service', 'material', 'environmental')") + ) + unit: Mapped[str] + + +class Demand(TableBase): + __tablename__ = "demand" + + year: Mapped[int] = mapped_column(primary_key=True, autoincrement=False) + commodity: Mapped[Commodities] = mapped_column( + ForeignKey("commodities.name"), + primary_key=True, + info=dict(header="commodity_name"), + ) + region: Mapped[Regions] = mapped_column( + ForeignKey("regions.name"), primary_key=True + ) + demand: Mapped[float] + + +def read_inputs(data_dir): + from sqlalchemy import create_engine + + engine = create_engine("duckdb:///:memory:") + TableBase.metadata.create_all(engine) + con = engine.raw_connection().driver_connection + + with open(data_dir / "regions.csv") as f: + regions = read_csv(f, Regions, con) # noqa: F841 + + with open(data_dir / "commodities.csv") as f: + commodities = read_csv(f, Commodities, con) + + with open(data_dir / "demand.csv") as f: + demand = read_csv(f, Demand, con) # noqa: F841 + + data = {} + data["global_commodities"] = calculate_global_commodities(commodities) + return data + + +def read_csv(buffer_, table_class, con): + table_name = table_class.__tablename__ + columns = ", ".join( + column.info.get("header", column.name) + for column in table_class.__table__.columns + ) + + rel = con.read_csv(buffer_, header=True, delimiter=",") # noqa: F841 + con.execute(f"INSERT INTO {table_name} SELECT {columns} FROM rel") + return con.execute(f"SELECT * from {table_name}").fetchnumpy() + + +def calculate_global_commodities(commodities): + names = commodities["name"].astype(np.dtype("str")) + types = commodities["type"].astype(np.dtype("str")) + units = commodities["unit"].astype(np.dtype("str")) + + type_array = xr.DataArray( + data=types, dims=["commodity"], coords=dict(commodity=names) + ) + + unit_array = xr.DataArray( + data=units, dims=["commodity"], coords=dict(commodity=names) + ) + + data = xr.Dataset(data_vars=dict(type=type_array, unit=unit_array)) + return data diff --git a/tests/test_fullsim_regression.py b/tests/test_fullsim_regression.py index 21897c660..49197e704 100644 --- a/tests/test_fullsim_regression.py +++ b/tests/test_fullsim_regression.py @@ -5,6 +5,9 @@ from muse.examples import AVAILABLE_EXAMPLES +# temporary skip for default_new_input as this is not yet working +AVAILABLE_EXAMPLES.remove("default_new_input") + @mark.regression @mark.example diff --git a/tests/test_readers.py b/tests/test_readers.py index 924dcacff..9148969b1 100644 --- a/tests/test_readers.py +++ b/tests/test_readers.py @@ -1,6 +1,9 @@ +from io import StringIO from itertools import chain, permutations from pathlib import Path +import duckdb +import numpy as np import pandas as pd import toml import xarray as xr @@ -314,3 +317,199 @@ def test_get_nan_coordinates(): dataset3 = xr.Dataset.from_dataframe(df3.set_index(["region", "year"])) nan_coords3 = get_nan_coordinates(dataset3) assert nan_coords3 == [] + + +@fixture +def default_new_input(tmp_path): + from muse.examples import copy_model + + copy_model("default_new_input", tmp_path) + return tmp_path / "model" + + +@fixture +def con(): + from sqlalchemy import create_engine + from sqlalchemy.orm import Session + + from muse.new_input.readers import TableBase + + engine = create_engine("duckdb:///:memory:") + session = Session(engine) + TableBase.metadata.create_all(engine) + return session.connection().connection + + +@fixture +def populate_regions(default_new_input, con): + from muse.new_input.readers import Regions, read_csv + + with open(default_new_input / "regions.csv") as f: + return read_csv(f, Regions, con) + + +@fixture +def populate_commodities(default_new_input, con): + from muse.new_input.readers import Commodities, read_csv + + with open(default_new_input / "commodities.csv") as f: + return read_csv(f, Commodities, con) + + +@fixture +def populate_demand(default_new_input, con, populate_regions, populate_commodities): + from muse.new_input.readers import Demand, read_csv + + with open(default_new_input / "demand.csv") as f: + return read_csv(f, Demand, con) + + +def test_read_regions(populate_regions): + assert populate_regions["name"] == np.array(["R1"]) + + +def test_read_regions_primary_key_constraint(default_new_input, con): + from muse.new_input.readers import Regions, read_csv + + csv = StringIO("name\nR1\nR1\n") + with raises(duckdb.ConstraintException, match=".*duplicate key.*"): + read_csv(csv, Regions, con) + + +def test_read_new_commodities(populate_commodities): + data = populate_commodities + assert list(data["name"]) == ["electricity", "gas", "heat", "wind", "CO2f"] + assert list(data["type"]) == ["energy"] * 5 + assert list(data["unit"]) == ["PJ"] * 4 + ["kt"] + + +def test_calculate_global_commodities(populate_commodities): + from muse.new_input.readers import calculate_global_commodities + + data = calculate_global_commodities(populate_commodities) + + assert isinstance(data, xr.Dataset) + assert set(data.dims) == {"commodity"} + for dt in data.dtypes.values(): + assert np.issubdtype(dt, np.dtype("str")) + + assert list(data.coords["commodity"].values) == list(populate_commodities["name"]) + assert list(data.data_vars["type"].values) == list(populate_commodities["type"]) + assert list(data.data_vars["unit"].values) == list(populate_commodities["unit"]) + + +def test_read_new_commodities_primary_key_constraint(default_new_input, con): + from muse.new_input.readers import Commodities, read_csv + + csv = StringIO("name,type,unit\nfoo,energy,bar\nfoo,energy,bar\n") + with raises(duckdb.ConstraintException, match=".*duplicate key.*"): + read_csv(csv, Commodities, con) + + +def test_read_new_commodities_type_constraint(default_new_input, con): + from muse.new_input.readers import Commodities, read_csv + + csv = StringIO("name,type,unit\nfoo,invalid,bar\n") + with raises(duckdb.ConstraintException): + read_csv(csv, Commodities, con) + + +def test_new_read_demand_csv(populate_demand): + data = populate_demand + assert np.all(data["year"] == np.array([2020, 2050])) + assert np.all(data["commodity"] == np.array(["heat", "heat"])) + assert np.all(data["region"] == np.array(["R1", "R1"])) + assert np.all(data["demand"] == np.array([10, 30])) + + +def test_new_read_demand_csv_commodity_constraint( + default_new_input, con, populate_commodities, populate_regions +): + from muse.new_input.readers import Demand, read_csv + + csv = StringIO("year,commodity_name,region,demand\n2020,invalid,R1,0\n") + with raises(duckdb.ConstraintException, match=".*foreign key.*"): + read_csv(csv, Demand, con) + + +def test_new_read_demand_csv_region_constraint( + default_new_input, con, populate_commodities, populate_regions +): + from muse.new_input.readers import Demand, read_csv + + csv = StringIO("year,commodity_name,region,demand\n2020,heat,invalid,0\n") + with raises(duckdb.ConstraintException, match=".*foreign key.*"): + read_csv(csv, Demand, con) + + +def test_new_read_demand_csv_primary_key_constraint( + default_new_input, con, populate_commodities, populate_regions +): + from muse.new_input.readers import Demand, Regions, read_csv + + # Add another region so we can test varying it as a primary key + csv = StringIO("name\nR2\n") + read_csv(csv, Regions, con) + + # all fine so long as one primary key column differs + csv = StringIO( + """year,commodity_name,region,demand +2020,gas,R1,0 +2021,gas,R1,0 +2020,heat,R1,0 +2020,gas,R2,0 +""" + ) + read_csv(csv, Demand, con) + + # no good if all primary key columns match a previous entry + csv = StringIO("year,commodity_name,region,demand\n2020,gas,R1,0") + with raises(duckdb.ConstraintException, match=".*duplicate key.*"): + read_csv(csv, Demand, con) + + +@mark.xfail +def test_demand_dataset(default_new_input): + import duckdb + + from muse.new_input.readers import read_commodities, read_demand, read_regions + + con = duckdb.connect(":memory:") + + read_regions(default_new_input, con) + read_commodities(default_new_input, con) + data = read_demand(default_new_input, con) + + assert isinstance(data, xr.DataArray) + assert data.dtype == np.float64 + + assert set(data.dims) == {"year", "commodity", "region", "timeslice"} + assert list(data.coords["region"].values) == ["R1"] + assert list(data.coords["timeslice"].values) == list(range(1, 7)) + assert list(data.coords["year"].values) == [2020, 2050] + assert set(data.coords["commodity"].values) == { + "electricity", + "gas", + "heat", + "wind", + "CO2f", + } + + assert data.sel(year=2020, commodity="electricity", region="R1", timeslice=0) == 1 + + +@mark.xfail +def test_new_read_initial_market(default_new_input): + from muse.new_input.readers import read_inputs + + all_data = read_inputs(default_new_input) + data = all_data["initial_market"] + + assert isinstance(data, xr.Dataset) + assert set(data.dims) == {"region", "year", "commodity", "timeslice"} + assert dict(data.dtypes) == dict( + prices=np.float64, + exports=np.float64, + imports=np.float64, + static_trade=np.float64, + )