Recife: Simple Simulation Example¶
In [1]:
from smum.microsim.run import run_calibrated_model
from smum.microsim.table import TableModel
/usr/lib/python3.6/site-packages/h5py-2.7.1-py3.6-linux-x86_64.egg/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
1. Define model¶
In [52]:
census_file = 'data/benchmarks_br_rec_year_bias.csv'
In [3]:
tm = TableModel(census_file = census_file)
In [4]:
tm.from_excel("data/tableModel_Water_rec.xlsx", "Water")
formula_water = "w_Intercept +\
c_w_ban * w_ban + \
c_w_connectio * w_connectio + \
c_w_age * w_age + \
c_w_dutyp * w_dutyp + \
c_w_urban * w_urban + \
c_w_sex * w_sex"
tm.add_formula(formula_water, 'Water')
In [5]:
tm.from_excel("data/tableModel_Electricity_rec.xlsx", "Electricity")
formula_electricity ="e_Intercept + \
c_e_edu * e_edu + \
c_e_sqm * e_sqm + \
c_e_income * e_income + \
c_e_hhsize * e_hhsize + \
c_e_dutyp * w_dutyp + \
c_e_urban * w_urban + \
c_e_sex * w_sex + \
e_cdd"
tm.add_formula(formula_electricity, 'Electricity')
In [168]:
table_model = tm.make_model()
2. Define model variables¶
In [7]:
labels_age = [
'age_20a24', 'age_25a29', 'age_30a34',
'age_35a39', 'age_40a44', 'age_45a49',
'age_50a54', 'age_55a59', 'age_60a64',
'age_65a69', 'age_70a74', 'age_75a79',
'age_80anosou'
]
cut_age = [19,
24, 29, 34,
39, 44, 49,
54, 59, 64,
69, 74, 79,
120]
labels_hh = ['hhsize_{}'.format(i) for i in range(1, 8)]
cut_hh = [0,1.55,2.55,3.55,4.55,5.55,6.55,20]
to_cat = {'w_age':[cut_age, labels_age], 'e_hhsize':[cut_hh, labels_hh]}
3. Run Simulation¶
In [ ]:
fw = run_calibrated_model(
table_model,
census_file = census_file,
year = 2016,
name = "Recife_simple",
drop_col_survey = ['e_income', 'e_cdd'],
to_cat = to_cat,
)
4. Validate Simulation¶
In [9]:
from smum.microsim.util_plot import plot_error
In [10]:
REC = plot_error(
'data/survey_Recife_simple.csv',
'data/benchmarks_br_rec_year_bias.csv',
'Recife_simple (1000 iterations)',
year = 2016,
skip = ['e_sqm'],
fit_cols = ['Electricity', 'Water'],
)
In [12]:
import pandas as pd
survey = pd.read_csv('data/survey_Recife_simple.csv', index_col=0)
In [22]:
iss = survey.shape[0]
rss = survey.loc[survey.wf > 0].shape[0]
print("""
The valid synthetic sample size is small!
Initial sample size: {:0.0f}, representative sample size: {:0.0f}""".format(iss, rss))
print("Only {:0.2%} of the synthetic sample is valid for this population".format(rss/iss))
The valid synthetic sample size is small!
Initial sample size: 956, representative sample size: 307
Only 32.11% of the synthetic sample is valid for this population
5. Define transition scenarios¶
In [1]:
from smum.microsim.run import transition_rate
from smum.microsim.util_plot import plot_transition_rate
from smum.microsim.run import reduce_consumption
/usr/lib/python3.6/site-packages/h5py-2.7.1-py3.6-linux-x86_64.egg/h5py/__init__.py:36: FutureWarning: Conversion of the second argument of issubdtype from `float` to `np.floating` is deprecated. In future, it will be treated as `np.float64 == np.dtype(float).type`.
from ._conv import register_converters as _register_converters
In [2]:
Elec = transition_rate(0, 0.4, start=2016)
Water = transition_rate(0, 0.2, start=2016)
pr = transition_rate(0, 0.3, start=2016)
In [3]:
plot_transition_rate(
{"Technology penetration rate": pr,
"Electricity efficiency increase rate": Elec,
"Water efficiency increase rate": Water},
"simple scenario")
In [4]:
sampling_rules = {
"e_edu == 'edu_master'": 30,
"e_hhsize == 'hhsize_1'": 20,
"e_hhsize == 'hhsize_2'": 10,
"w_dutyp == 'dutyp_Casa'": 5,
}
file_name = "data/survey_Recife_simple.csv"
scenario_name="simple_scenario"
reduction = {'Electricity':Elec, 'Water':Water}
In [5]:
reduce_consumption(
file_name, pr, sampling_rules, reduction, scenario_name=scenario_name)
00.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2010
-0.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2010
00.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2011
-0.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2011
00.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2012
-0.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2012
-0.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2013
00.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2013
00.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2014
00.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2014
00.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2015
-0.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2015
-0.00% Electricity reduction; efficiency 00.00%; penetration 00.00; year 2016
00.00% Water reduction; efficiency 00.00%; penetration 00.00; year 2016
00.06% Electricity reduction; efficiency 02.86%; penetration 00.02; year 2017
00.03% Water reduction; efficiency 01.43%; penetration 00.02; year 2017
00.24% Electricity reduction; efficiency 05.71%; penetration 00.04; year 2018
00.12% Water reduction; efficiency 02.86%; penetration 00.04; year 2018
00.53% Electricity reduction; efficiency 08.57%; penetration 00.06; year 2019
00.27% Water reduction; efficiency 04.29%; penetration 00.06; year 2019
00.94% Electricity reduction; efficiency 11.43%; penetration 00.09; year 2020
00.48% Water reduction; efficiency 05.71%; penetration 00.09; year 2020
01.47% Electricity reduction; efficiency 14.29%; penetration 00.11; year 2021
00.75% Water reduction; efficiency 07.14%; penetration 00.11; year 2021
02.12% Electricity reduction; efficiency 17.14%; penetration 00.13; year 2022
01.08% Water reduction; efficiency 08.57%; penetration 00.13; year 2022
02.89% Electricity reduction; efficiency 20.00%; penetration 00.15; year 2023
01.48% Water reduction; efficiency 10.00%; penetration 00.15; year 2023
03.78% Electricity reduction; efficiency 22.86%; penetration 00.17; year 2024
01.93% Water reduction; efficiency 11.43%; penetration 00.17; year 2024
04.78% Electricity reduction; efficiency 25.71%; penetration 00.19; year 2025
02.45% Water reduction; efficiency 12.86%; penetration 00.19; year 2025
05.90% Electricity reduction; efficiency 28.57%; penetration 00.21; year 2026
03.02% Water reduction; efficiency 14.29%; penetration 00.21; year 2026
07.15% Electricity reduction; efficiency 31.43%; penetration 00.24; year 2027
03.66% Water reduction; efficiency 15.71%; penetration 00.24; year 2027
08.50% Electricity reduction; efficiency 34.29%; penetration 00.26; year 2028
04.35% Water reduction; efficiency 17.14%; penetration 00.26; year 2028
09.98% Electricity reduction; efficiency 37.14%; penetration 00.28; year 2029
05.10% Water reduction; efficiency 18.57%; penetration 00.28; year 2029
11.58% Electricity reduction; efficiency 40.00%; penetration 00.30; year 2030
05.92% Water reduction; efficiency 20.00%; penetration 00.30; year 2030
6. Visualize transition scenarios¶
In [6]:
from smum.microsim.util_plot import plot_data_projection
In [7]:
iterations = 1000
typ = 'reweighted'
reweighted_survey = 'data/survey_Recife_simple'
In [9]:
var = ['Water', 'Electricity']
data, cap = plot_data_projection(
reweighted_survey, var, "{}; Recife".format(iterations),
benchmark_year=2016,
)
