Merge pull request #40 from switchbox-data/27-add-scattershot-electrification

27 add scattershot electrification

Author: alxsmith

run_example.py

@@ -68,7 +68,7 @@
     "peak_kw_winter_headroom": 10.0,
     "peak_kw_summer_headroom": 10.0,
     "aircon_percent_adoption_pre_npa": 0.8,
-    "non_npa_scattershot_electrifiction_users_per_year": 5,
+    "scattershot_electrification_users_per_year": 5,
     "gas_fixed_overhead_costs": 100.0,
     "electric_fixed_overhead_costs": 100.0,
     "gas_bau_lpp_costs_per_year": 100.0,

src/npa_howtopay/data/sample.yaml

@@ -44,13 +44,16 @@ shared:
 time_series:
   npa_projects:
     project_year: [2025, 2025, 2026, 2026, 2027, 2028, 2029, 2030, 2031, 2032, 2033, 2034, 2035, 2036, 2037, 2038, 2039, 2040, 2041, 2042, 2043, 2044, 2045, 2046, 2047, 2048, 2049, 2050]
-    num_converts: [1.0e3, 0, 0, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3, 1.0e3]
+    num_converts: [1000, 0, 0, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000]
     pipe_value_per_user: [30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3, 30.0e3]
     pipe_decomm_cost_per_user: [10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3, 10.0e3]
     peak_kw_winter_headroom: [10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0]
     peak_kw_summer_headroom: [10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0, 10.0]
     aircon_percent_adoption_pre_npa: [0.8, 0.9, 1, 0.8, 0.9, 1, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.8]
     is_scattershot: [false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false, false]
+  scattershot_electrification_users_per_year:
+    project_year: [2025, 2026, 2027, 2028, 2029, 2030, 2031, 2032, 2033, 2034, 2035, 2036, 2037, 2038, 2039, 2040, 2041, 2042, 2043, 2044, 2045, 2046, 2047, 2048, 2049, 2050]
+    num_converts: [1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000, 1000]
   gas_fixed_overhead_costs:
     year: [2025, 2026, 2027, 2028, 2029, 2030, 2031, 2032, 2033, 2034, 2035, 2036, 2037, 2038, 2039, 2040, 2041, 2042, 2043, 2044, 2045, 2046, 2047, 2048, 2049, 2050]
     cost: [5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6, 5.0e6]

src/npa_howtopay/model.py

@@ -508,6 +508,7 @@ def compute_bill_costs(
 
 
 def run_model(scenario_params: ScenarioParams, input_params: InputParams, ts_params: TimeSeriesParams) -> pl.DataFrame:
+    # in the business-as-usual scenario, we don't have any npa projects. We maintain the scattershot electrification which will still reduce the number of gas customers and total gas usage but will not trigger grid upgrade or capex/opex for either utility.
     if scenario_params.bau:
         ts_params = evolve(ts_params, npa_projects=npa.return_empty_npa_df())
 

src/npa_howtopay/npa_project.py

@@ -44,78 +44,23 @@ def to_df(self) -> pl.DataFrame:
         })
 
 
-def generate_npa_projects(
-    start_year: int,
-    end_year: int,
-    total_num_projects: int,
-    num_converts_per_project: int,
-    pipe_value_per_user: float,
-    pipe_decomm_cost_per_user: float,
-    peak_kw_winter_headroom_per_project: float,
-    peak_kw_summer_headroom_per_project: float,
-    aircon_percent_adoption_pre_npa: float,
-    pipe_decomm_cost_inflation_rate: float = 0.0,
+def append_scattershot_electrification_df(
+    npa_projects_df: pl.DataFrame,
+    scattershot_electrification_df: pl.DataFrame,
 ) -> pl.DataFrame:
     """
-    Generate a dataframe of NPA projects of length total_num_projects.
-    The projects are distributed evenly across the years, with remainders added
-    to earlier years. Note that there can be multiple project rows per year
-    in the result
+    Append a dataframe of scattershot electrification projects to the npa projects df.
+    Scattershot electrification projects match the schema for npa projects, but will only affect the number of users and total electric usage, not anything related to pipe value or grid upgrades. It is meant to capture customers leaving the gas network independent of NPA projects.
     """
-    years = range(start_year, end_year + 1)
-    num_years = len(years)
-    base = total_num_projects // num_years
-    remainder = total_num_projects % num_years
-
-    projects_per_year = np.full(num_years, base, dtype=int)
-    projects_per_year[:remainder] += 1
-
-    project_years = [b for a in [[y] * r for y, r in zip(years, projects_per_year)] for b in a]
-    pipe_decomm_costs = [
-        pipe_decomm_cost_per_user * (1.0 + pipe_decomm_cost_inflation_rate) ** (y - start_year) for y in project_years
-    ]
-
-    return pl.DataFrame({
-        "project_year": project_years,
-        "num_converts": [num_converts_per_project] * total_num_projects,
-        "pipe_value_per_user": [float(pipe_value_per_user)] * total_num_projects,
-        "pipe_decomm_cost_per_user": pipe_decomm_costs,
-        "peak_kw_winter_headroom": [float(peak_kw_winter_headroom_per_project)] * total_num_projects,
-        "peak_kw_summer_headroom": [float(peak_kw_summer_headroom_per_project)] * total_num_projects,
-        "aircon_percent_adoption_pre_npa": [float(aircon_percent_adoption_pre_npa)] * total_num_projects,
-        "is_scattershot": [False] * total_num_projects,
-    })
-
-
-def generate_scattershot_electrification_projects(
-    start_year: int,
-    end_year: int,
-    total_num_converts: int,
-) -> pl.DataFrame:
-    """
-    Generate a dataframe of scattershot electrification projects, one per year.
-    The projects are distributed evenly across the years, with remainders added
-    to earlier years. These match the schema for npa projects, but will only
-    affect the number of users, not anything related to pipe value or grid upgrades
-    """
-    years = range(start_year, end_year + 1)
-    num_years = len(years)
-    base = total_num_converts // num_years
-    remainder = total_num_converts % num_years
-
-    converts_per_year = np.full(num_years, base, dtype=int)
-    converts_per_year[:remainder] += 1
-
-    return pl.DataFrame({
-        "project_year": years,
-        "num_converts": converts_per_year,
-        "pipe_value_per_user": [0.0] * num_years,
-        "pipe_decomm_cost_per_user": [0.0] * num_years,
-        "peak_kw_winter_headroom": [np.inf] * num_years,
-        "peak_kw_summer_headroom": [np.inf] * num_years,
-        "aircon_percent_adoption_pre_npa": [0.0] * num_years,
-        "is_scattershot": [True] * num_years,
-    })
+    scattershot_with_npa_cols = scattershot_electrification_df.with_columns(
+        pl.lit(0.0).alias("pipe_value_per_user"),
+        pl.lit(0.0).alias("pipe_decomm_cost_per_user"),
+        pl.lit(np.inf).alias("peak_kw_winter_headroom"),
+        pl.lit(np.inf).alias("peak_kw_summer_headroom"),
+        pl.lit(0.0).alias("aircon_percent_adoption_pre_npa"),
+        pl.lit(True).alias("is_scattershot"),
+    )
+    return pl.concat([npa_projects_df, scattershot_with_npa_cols])
 
 
 def compute_hp_converts_from_df(year: int, df: pl.DataFrame, cumulative: bool = False, npa_only: bool = False) -> int:
@@ -182,12 +127,12 @@ def compute_pipe_decomm_cost_from_df(year: int, df: pl.DataFrame) -> float:
 
 def return_empty_npa_df() -> pl.DataFrame:
     return pl.DataFrame({
-        "project_year": [],
-        "num_converts": [],
-        "pipe_value_per_user": [],
-        "pipe_decomm_cost_per_user": [],
-        "peak_kw_winter_headroom": [],
-        "peak_kw_summer_headroom": [],
-        "aircon_percent_adoption_pre_npa": [],
-        "is_scattershot": [],
+        "project_year": pl.Series([], dtype=pl.Int64),
+        "num_converts": pl.Series([], dtype=pl.Int64),
+        "pipe_value_per_user": pl.Series([], dtype=pl.Float64),
+        "pipe_decomm_cost_per_user": pl.Series([], dtype=pl.Float64),
+        "peak_kw_winter_headroom": pl.Series([], dtype=pl.Float64),
+        "peak_kw_summer_headroom": pl.Series([], dtype=pl.Float64),
+        "aircon_percent_adoption_pre_npa": pl.Series([], dtype=pl.Float64),
+        "is_scattershot": pl.Series([], dtype=pl.Boolean),
     })

src/npa_howtopay/params.py

@@ -3,6 +3,7 @@
 from ruamel.yaml import YAML
 import polars as pl
 from npa_howtopay.web_params import create_time_series_from_web_params, WebParams
+from npa_howtopay.npa_project import append_scattershot_electrification_df
 
 # from npa_project import NpaProject
 import os
@@ -123,10 +124,16 @@ def __attrs_post_init__(self) -> None:
 @define
 class TimeSeriesParams:
     npa_projects: pl.DataFrame
+    scattershot_electrification: pl.DataFrame
     gas_fixed_overhead_costs: pl.DataFrame
     electric_fixed_overhead_costs: pl.DataFrame
     gas_bau_lpp_costs_per_year: pl.DataFrame
 
+    def __attrs_post_init__(self) -> None:
+        """Automatically append scattershot electrification to npa projects. In the BAU scenario, this will only return the scattershot electrification dataframe."""
+
+        self.npa_projects = append_scattershot_electrification_df(self.npa_projects, self.scattershot_electrification)
+
 
 @define
 class ScenarioParams:
@@ -183,6 +190,7 @@ def _load_time_series_params_from_yaml(yaml_path: str) -> TimeSeriesParams:
 
     return TimeSeriesParams(
         npa_projects=pl.DataFrame(config["time_series"]["npa_projects"]),
+        scattershot_electrification=pl.DataFrame(config["time_series"]["scattershot_electrification_users_per_year"]),
         gas_fixed_overhead_costs=pl.DataFrame(config["time_series"]["gas_fixed_overhead_costs"]),
         electric_fixed_overhead_costs=pl.DataFrame(config["time_series"]["electric_fixed_overhead_costs"]),
         gas_bau_lpp_costs_per_year=pl.DataFrame(config["time_series"]["gas_bau_lpp_costs_per_year"]),
@@ -230,6 +238,7 @@ def load_time_series_params_from_web_params(
 
     return TimeSeriesParams(
         npa_projects=generated_data["npa_projects"],
+        scattershot_electrification=generated_data["scattershot_electrification_users_per_year"],
         gas_fixed_overhead_costs=generated_data["gas_fixed_overhead_costs"],
         electric_fixed_overhead_costs=generated_data["electric_fixed_overhead_costs"],
         gas_bau_lpp_costs_per_year=generated_data["gas_bau_lpp_costs_per_year"],

src/npa_howtopay/web_params.py

@@ -12,7 +12,7 @@ class WebParams:
     peak_kw_winter_headroom: float
     peak_kw_summer_headroom: float
     aircon_percent_adoption_pre_npa: float
-    non_npa_scattershot_electrifiction_users_per_year: int
+    scattershot_electrification_users_per_year: int
     gas_fixed_overhead_costs: float
     electric_fixed_overhead_costs: float
     gas_bau_lpp_costs_per_year: float
@@ -33,6 +33,25 @@ def create_npa_projects(web_params: WebParams, start_year: int, end_year: int) -
     })
 
 
+def create_scattershot_electrification_df(
+    web_params: WebParams,
+    start_year: int,
+    end_year: int,
+) -> pl.DataFrame:
+    """
+    Generate a dataframe of scattershot electrification projects, one per year.
+    The projects are distributed evenly across the years. These match the schema for npa projects, but will only
+    affect the number of users, not anything related to pipe value or grid upgrades
+    """
+    years = range(start_year, end_year + 1)
+    num_years = len(years)
+
+    return pl.DataFrame({
+        "project_year": list(years),
+        "num_converts": [web_params.scattershot_electrification_users_per_year] * num_years,
+    })
+
+
 def create_gas_fixed_overhead_costs(
     web_params: WebParams, start_year: int, end_year: int, cost_inflation_rate: float = 0.0
 ) -> pl.DataFrame:
@@ -84,6 +103,9 @@ def create_time_series_from_web_params(
     """Create all time series DataFrames from web parameters"""
     return {
         "npa_projects": create_npa_projects(web_params, start_year, end_year),
+        "scattershot_electrification_users_per_year": create_scattershot_electrification_df(
+            web_params, start_year, end_year
+        ),
         "gas_fixed_overhead_costs": create_gas_fixed_overhead_costs(
             web_params, start_year, end_year, cost_inflation_rate
         ),

tests/test_input_load.py

@@ -1,3 +1,4 @@
+import numpy as np
 import polars as pl
 import pytest
 from polars.testing import assert_frame_equal
@@ -20,7 +21,7 @@ def web_params():
         "peak_kw_winter_headroom": 10.0,
         "peak_kw_summer_headroom": 10.0,
         "aircon_percent_adoption_pre_npa": 0.8,
-        "non_npa_scattershot_electrifiction_users_per_year": 5,
+        "scattershot_electrification_users_per_year": 5,
         "gas_fixed_overhead_costs": 100.0,
         "electric_fixed_overhead_costs": 100.0,
         "gas_bau_lpp_costs_per_year": 100.0,
@@ -86,14 +87,14 @@ def expected_electric_fixed_overhead_costs_with_inflation():
 def expected_npa_projects():
     """Expected NPA projects for 2025-2030"""
     return pl.DataFrame({
-        "project_year": [2025, 2026, 2027, 2028, 2029, 2030],
-        "num_converts": [100, 100, 100, 100, 100, 100],
-        "pipe_value_per_user": [1000.0, 1000.0, 1000.0, 1000.0, 1000.0, 1000.0],
-        "pipe_decomm_cost_per_user": [100.0, 100.0, 100.0, 100.0, 100.0, 100.0],
-        "peak_kw_winter_headroom": [10.0, 10.0, 10.0, 10.0, 10.0, 10.0],
-        "peak_kw_summer_headroom": [10.0, 10.0, 10.0, 10.0, 10.0, 10.0],
-        "aircon_percent_adoption_pre_npa": [0.8, 0.8, 0.8, 0.8, 0.8, 0.8],
-        "is_scattershot": [False, False, False, False, False, False],
+        "project_year": [2025, 2026, 2027, 2028, 2029, 2030, 2025, 2026, 2027, 2028, 2029, 2030],
+        "num_converts": [100, 100, 100, 100, 100, 100, 5, 5, 5, 5, 5, 5],
+        "pipe_value_per_user": [1000.0, 1000.0, 1000.0, 1000.0, 1000.0, 1000.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
+        "pipe_decomm_cost_per_user": [100.0, 100.0, 100.0, 100.0, 100.0, 100.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
+        "peak_kw_winter_headroom": [10.0, 10.0, 10.0, 10.0, 10.0, 10.0, np.inf, np.inf, np.inf, np.inf, np.inf, np.inf],
+        "peak_kw_summer_headroom": [10.0, 10.0, 10.0, 10.0, 10.0, 10.0, np.inf, np.inf, np.inf, np.inf, np.inf, np.inf],
+        "aircon_percent_adoption_pre_npa": [0.8, 0.8, 0.8, 0.8, 0.8, 0.8, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0],
+        "is_scattershot": [False, False, False, False, False, False, True, True, True, True, True, True],
     })
 
 
@@ -174,7 +175,8 @@ def test_load_time_series_params_from_yaml():
     assert set(params.electric_fixed_overhead_costs.columns) == {"year", "cost"}
 
     # Verify npa_projects df has correct shape and columns
-    assert params.npa_projects.shape == (28, 8)
+    # 28 rows from npa_projects + 26 rows from scattershot_electrification = 54 total rows
+    assert params.npa_projects.shape == (54, 8)
     assert set(params.npa_projects.columns) == {
         "project_year",
         "num_converts",