#!/usr/bin/env python
# -*- coding: utf-8 -*-
import networkx as nx
import numpy as np
import geopandas as gpd
import matplotlib.pyplot as plt
from dhd.utils import get_moment
[docs]def plot_district_heating_network(pipelines, plot, loc="upper center"):
"""
Plot the geometries of the district heating network on the plot interface.
Parameters
----------
pipeplines: GeoDataFrame
Geodataframe of the district heating network.
plot: dhd.plot.Plot
Plot interface of the considered city.
loc: tuple, optional
Location of the legend box. Default is *loc* = 'upper center'.
"""
label = "District Heating Network"
kwargs = {"color": "black", "linewidth": 1.5}
plot.add_geodataframe(pipelines, kwargs)
plot.add_legend([kwargs["color"]], [label], "line", loc=loc)
[docs]def plot_terminals(terminals, plot, indices=None, loc="lower left"):
"""
Plot the geometries of the sinks on the plot interface.
The connected sinks are green and the disconnected sinks are red.
Depending on the type of *indices* all geometries, a series of geometries or
only one geometry are plotted.
Parameters
----------
terminals: GeoDataFrame
Geodataframe of the terminals.
plot: dhd.plot.Plot
Plot interface of the considered city.
indices: None or list or string, optional
Indices of the sinks to plot. It may be a list or only one index. If
None all geometries are plotted. Default is None.
loc: tuple, optional
Location of the legend box. Default is *loc* = 'lower left'.
"""
if indices is None:
gdf = gpd.GeoDataFrame(terminals)
elif type(indices) == list:
gdf = gpd.GeoDataFrame(terminals.loc[indices])
else:
gdf = gpd.GeoDataFrame(terminals.loc[[indices]])
gdf = gdf.loc[gdf["kind"] == "sink"]
gdf1 = gdf.loc[gdf["connected"] == True]
gdf2 = gdf.loc[gdf["connected"] == False]
colors = ["green", "red"]
labels = ["connected", "disconnected"]
kwargs1 = {"color": colors[0]}
kwargs2 = {"color": colors[1]}
plot.add_geodataframe(gdf1, kwargs1)
plot.add_geodataframe(gdf2, kwargs2)
plot.add_legend(colors, labels, "patch", loc=loc)
[docs]def get_connection_path(pipelines, id1, id2):
"""
Find the path between two terminals and return the selection of the dataframe
*pipeplines* referring to it.
Parameters
----------
pipelines: GeoDataFrame
Geodataframe of the pipelines.
id1: string
Index of the first terminal.
id2: string
Index of the second terminal.
Returns
-------
GeoDataFrame
Selection of the dataframe *pipelines* referring to the connection path.
"""
pipelines_ = pipelines.reset_index()
graph = nx.from_pandas_edgelist(
pipelines_, source="idA", target="idB", edge_attr=["weight", "index"]
)
path = nx.shortest_path(graph, id1, id2, weight="weight")
indices = list()
for i in range(len(path) - 1):
idA = path[i]
idB = path[i + 1]
indices.append(graph[idA][idB]["index"])
connection_path = pipelines.loc[indices]
return connection_path
[docs]def get_column_statistics(df, key):
"""
Compute the mean and the standard deviation of the data stored in the column
*key* of the dataframe *df*.
Parameters
----------
df: DataFrame
key: string
Key of the column of interest.
Returns
-------
float
Mean of the data.
float
Standard deviation of the data.
"""
data = list(df[key])
m1 = get_moment(data, 1)
m2 = get_moment(data, 2)
mean = m1
stddev = np.sqrt(m2 - m1 ** 2)
return mean, stddev
[docs]def get_statistics(pipelines, keys=None):
"""
Print the statistics (mean and standard deviation) of the different columns
of the dataframe *pipelines*.
Parameters
----------
pipelines: GeoDataFrame
Geodataframe of the pipelines.
keys: list, optional
List of the columns statistics to print. If None all the numeric columns
are considered. Default is None.
"""
if keys is None:
keys = ["length", "weight", "load", "n_sink", "diameter"]
print("Statistics of the pipelines:\n")
for key in keys:
mean, stddev = get_column_statistics(pipelines, key)
text = "{}: mean = {} / standard deviation = {}.\n"
print(text.format(key, mean, stddev))
[docs]def plot_distribution(pipelines, key, bins=20, color="C1"):
"""
Plot the distribution of the data of the column *key* of the dataframe
*pipelines* on a histogram.
Parameters
----------
pipelines: GeoDataFrame
Geodataframe of the pipelines.
key: string
Key of the column of interest.
bins: int, optional
Number of bars in the histogram. Default is 20.
color: string
Color of the histogram. Default is 'C1'.
"""
fig, ax = plt.subplots(figsize=(6, 4))
data = [x for x in pipelines[key]]
plt.hist(data, bins=bins, color=color)
ax.set(xlabel=key, ylabel="# of occurences")
plt.show()
[docs]def get_intersection_number(pipelines, terminals):
"""
Compute the number of intersections in the district heating network.
Parameters
----------
pipelines: DataFrame
Dataframe of the district heating network.
terminals: DataFrame
Dataframe of the terminals.
Returns
-------
int
Number of intersections.
"""
sA = set(pipelines["idA"])
sB = set(pipelines["idB"])
s = sA.union(sB)
n1 = len(s)
n2 = len(terminals.loc[terminals["connected"] == True])
n = n1 - n2
return n
