# This plotting module has a corresponding demo notebook in
# PostREISE/postreise/plot/demo: energy_emissions_stack_bar_demo.ipynb
import matplotlib.pyplot as plt
import numpy as np
from powersimdata.network.model import ModelImmutables
from powersimdata.scenario.scenario import Scenario
from postreise.analyze.generation.emissions import generate_emissions_stats
# Define common classifications
possible_renewable = ["solar", "wind", "wind_offshore"]
possible_clean = possible_renewable + ["geothermal", "hydro", "nuclear"]
possible_carbon = ["coal", "dfo", "ng"]
all_possible = possible_carbon + ["other"] + possible_clean
[docs]def plot_n_scenarios(*args):
"""For 1-to-n scenarios, plot stacked energy and carbon side-by-side.
:param powersimdata.scenario.scenario.Scenario args: scenario instances.
:raises ValueError: if arguments are not scenario instances.
"""
if not all([isinstance(a, Scenario) for a in args]):
raise ValueError("all inputs must be Scenario objects")
# Build dictionaries of calculated data for each scenario
scenario_numbers = [a.info["id"] for a in args]
first_id = scenario_numbers[0]
scenarios = {id: scen for (id, scen) in zip(scenario_numbers, args)}
grid = {id: scenario.get_grid() for id, scenario in scenarios.items()}
plant = {k: v.plant for k, v in grid.items()}
# First scenario is chosen to set fuel colors
type2color = ModelImmutables(args[0].info["grid_model"]).plants["type2color"]
carbon_by_type, energy_by_type = {}, {}
for id, scenario in scenarios.items():
# Calculate raw numbers
annual_plant_energy = scenario.get_pg().sum()
raw_energy_by_type = annual_plant_energy.groupby(plant[id].type).sum()
annual_plant_carbon = generate_emissions_stats(scenario).sum()
raw_carbon_by_type = annual_plant_carbon.groupby(plant[id].type).sum()
# Drop fuels with zero energy (e.g. all offshore_wind scaled to 0 MW)
energy_by_type[id] = raw_energy_by_type[raw_energy_by_type != 0]
carbon_by_type[id] = raw_carbon_by_type[raw_energy_by_type != 0]
# Carbon multiplier is inverse of carbon intensity, to scale bar heights
carbon_multiplier = energy_by_type[first_id].sum() / carbon_by_type[first_id].sum()
# Determine the fuel types with generation in either scenario
fuels = list(set.union(*[set(v.index) for v in energy_by_type.values()]))
# Fill zeros in scenarios without a fuel when it's present in another
for id in scenario_numbers:
energy_by_type[id] = energy_by_type[id].reindex(fuels, fill_value=0)
carbon_by_type[id] = carbon_by_type[id].reindex(fuels, fill_value=0)
# Re-order according to plotting preferences
fuels = [f for f in all_possible if f in fuels]
# Re-assess subsets
renewable_fuels = [f for f in possible_renewable if f in fuels]
clean_fuels = [f for f in possible_clean if f in fuels]
carbon_fuels = [f for f in possible_carbon if f in fuels]
dropped_fuels = (set(possible_clean) | set(possible_carbon)) - (
set(clean_fuels) | set(carbon_fuels)
)
print("no energy in any grid(s), dropping: %s" % ", ".join(dropped_fuels))
fuel_series = {
f: sum(
[
[energy_by_type[s][f], carbon_by_type[s][f] * carbon_multiplier]
for s in scenario_numbers
],
[],
)
for f in fuels
}
num_bars = 2 * len(scenario_numbers)
ind = np.arange(num_bars)
width = 0.5
line_alpha = 0.25
line_offsets = np.array((0.25, 0.75))
# Strart plotting
fig = plt.figure(figsize=(10, 8))
ax = fig.add_subplot(111)
# Plot bars
patches = {}
bottom = np.zeros(num_bars)
for i, f in enumerate(fuels):
patches[i] = plt.bar(
ind, fuel_series[f], width, bottom=bottom, color=type2color[f]
)
bottom += fuel_series[f]
# Plot lines
for i, fuel in enumerate(carbon_fuels):
for j, s in enumerate(scenario_numbers):
cumulative_energy = sum(
[energy_by_type[s][carbon_fuels[k]] for k in range(i + 1)]
)
cumulative_scaled_carbon = carbon_multiplier * sum(
[carbon_by_type[s][carbon_fuels[k]] for k in range(i + 1)]
)
ys = (cumulative_energy, cumulative_scaled_carbon)
xs = line_offsets + 2 * j
plt.plot(xs, ys, "k-", alpha=line_alpha)
# Labeling
energy_fmt = "{0:,.0f} TWh\n{1:.0f}% renewable\n{2:.0f}% carbon-free\n"
carbon_fmt = "{0:,.0f} million\ntons CO2\n"
energy_total = {s: energy_by_type[s].sum() for s in scenarios}
for j, s in enumerate(scenario_numbers):
carbon_total = carbon_by_type[s].sum()
renewable_energy = sum([energy_by_type[s][f] for f in renewable_fuels])
clean_energy = sum([energy_by_type[s][f] for f in clean_fuels])
renewable_share = renewable_energy / energy_total[s] * 100
clean_share = clean_energy / energy_total[s] * 100
annotation_str = energy_fmt.format(
energy_total[s] / 1e6, renewable_share, clean_share
)
annotation_x = 2 * (j - 0.08)
plt.annotate(xy=(annotation_x, energy_total[s]), text=annotation_str)
annotation_str = carbon_fmt.format(carbon_total / 1e6)
annotation_x = 2 * (j - 0.08) + 1
annotation_y = carbon_total * carbon_multiplier
plt.annotate(xy=(annotation_x, annotation_y), text=annotation_str)
# Plot formatting
plt.ylim((0, max(energy_total.values()) * 1.2))
labels = sum([["%s Energy" % id, "%s Carbon" % id] for id in scenario_numbers], [])
plt.xticks(ind, labels)
plt.yticks([], [])
ax.tick_params(axis="x", labelsize=12)
# Invert default legend order to match stack ordering
plt.legend(
handles=(patches[i] for i in range(len(fuels) - 1, -1, -1)),
labels=fuels[::-1],
fontsize=12,
bbox_to_anchor=(1.02, 1),
loc="upper left",
)
plt.tight_layout()
plt.show()
