Skip to content

Commit

Permalink
Merge pull request #153 from nismod/feature/yield_enduses
Browse files Browse the repository at this point in the history 
Feature/yield enduses
  • Loading branch information
eggimasv committed Jan 31, 2019
2 parents fee6c1b + 9ab56fe commit c4e4bae
Show file tree
Hide file tree
Showing 11 changed files with 97 additions and 216 deletions.
5 changes: 1 addition & 4 deletions .environment.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,16 +13,13 @@ dependencies:
- matplotlib
- numpy
- palettable
- pyinstrument
- pyproj
- pysal
- pyshp
- pytest # test
- pytest-cov # test
- pyyaml
- recommonmark # docs
- scipy
- sphinx # docs
- sphinx_rtd_theme # docs
- pip:
- isoweek
- ruamel.yaml==0.15.50
224 changes: 64 additions & 160 deletions energy_demand/model.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@
and into class methods, and then calling each method in turn
- suggest explictly creating class properties to access results, rather than
the ``setattr`` magic
- potential for parallelisation of the for-loop around regions for generation
- potential for parallelisation of the for-loop around regions for generation
by dwelling stocks
- suggest splitting ``aggregate_across_all_regions`` into separate methods as each
of these requires creating a very large array which is then held in memory,
Expand All @@ -26,7 +26,14 @@
import logging
from collections import defaultdict
import numpy as np
import matplotlib.pyplot as plt

try:
from tqdm import tqdm
except ImportError:
def tqdm(iterator, *_, **__):
"""Alternative to tqdm, with no progress bar - ignore any arguments after the first
"""
return iterator

from energy_demand.basic import lookup_tables
import energy_demand.enduse_func as endusefunctions
Expand Down Expand Up @@ -97,12 +104,7 @@ def __init__(
# Simulate regions
# -------------------------------------------
print("... generating by dwelling stocks", flush=True)
for reg_array_nr, region in enumerate(regions):

logging.info("... Simulate: region %s, simulation year: %s, percent: (%s)",
region, assumptions.curr_yr, round((100/assumptions.reg_nrs)*reg_array_nr, 2))
print("... Simulate: region %s, simulation year: %s, percent: (%s)",
region, assumptions.curr_yr, round((100/assumptions.reg_nrs)*reg_array_nr, 2), flush=True)
for reg_array_nr, region in enumerate(tqdm(regions)):

all_submodels = simulate_region(
region,
Expand Down Expand Up @@ -298,30 +300,6 @@ def aggregate_load_factors(

return reg_load_factor_y, reg_load_factor_yd

def get_all_submodels(submodels, submodel_name):
"""Collect all submodel objects for a
specific submodel name
Arguments
---------
submodels : list
All calculated model objects
submodel_name : str
Name of submodels to collect
Returns
-------
specific_submodels : list
Contains all submodels
"""
specific_submodels = []

for submodel in submodels:
if submodel.submodel_name == submodel_name:
specific_submodels.append(submodel)

return specific_submodels

def get_disaggregated_fuel_of_reg(submodel_names, fuel_disagg, region):
"""Get disaggregated region for every submodel
for a specific region
Expand Down Expand Up @@ -370,8 +348,6 @@ def simulate_region(
all_submodels : list
List with submodel objects depending on submodels
"""
submodel_objs = []

submodel_names = assumptions.submodels_names

# Get closest weather region object
Expand Down Expand Up @@ -434,7 +410,7 @@ def simulate_region(
# ---------------
# Create submodel for region and enduse
# ---------------
submodel_obj = endusefunctions.Enduse(
yield endusefunctions.Enduse(
submodel_name=submodel_name,
region=region_obj.name,
scenario_data=data['scenario_data'],
Expand All @@ -457,18 +433,6 @@ def simulate_region(
reg_scen_drivers=assumptions.scenario_drivers,
flat_profile_crit=flat_profile_crit)

submodel_objs.append(submodel_obj)

# Collect all submodels with respect to submodel names and store in list
# e.g. [[all_residential-submodels], [all_service_submodels]...]
all_submodels = []
submodels = lookup_tables.basic_lookups()['submodels_names']
for submodel in submodels:
submodels = get_all_submodels(submodel_objs, submodel)
all_submodels.append(submodels)

return all_submodels

def fuel_aggr(
sector_models,
sum_crit,
Expand All @@ -482,7 +446,7 @@ def fuel_aggr(
Arguments
----------
sector_models : list
sector_models : list[Enduse]
Sector models
sum_crit : str
Criteria
Expand All @@ -507,23 +471,22 @@ def fuel_aggr(
"""
input_array = np.zeros(shape_aggregation_array, dtype="float")

for sector_model in sector_models:
for enduse_object in sector_model:
for enduse_object in sector_models:

fuels = get_fuels_yh(
enduse_object,
attribute_technologies)

if isinstance(fuels, dict):
for tech, fuel_tech in fuels.items():
tech_fueltype = technologies[tech].fueltype_int
input_array[tech_fueltype] += fuel_tech
else:
# Fuel per technology
fuels = get_fuels_yh(
enduse_object,
attribute_technologies)

if isinstance(fuels, dict):
for tech, fuel_tech in fuels.items():
tech_fueltype = technologies[tech].fueltype_int
input_array[tech_fueltype] += fuel_tech
else:
# Fuel per technology
fuels = get_fuels_yh(
enduse_object,
attribute_non_technology)
input_array += fuels
attribute_non_technology)
input_array += fuels

if sum_crit == 'no_sum':
return input_array
Expand Down Expand Up @@ -565,7 +528,7 @@ def get_fuels_yh(enduse_object, attribute_to_get):
elif attribute_to_get == 'shape_non_peak_yd':
fuels = fuels_reg_y * flat_shape_yd
elif attribute_to_get == 'fuel_yh' or attribute_to_get == 'techs_fuel_yh':
f_hour = 0.00011415525114155251 #1 / 8760
f_hour = 1 / 8760
flat_shape = np.full((enduse_object.fuel_y.shape[0], 365, 24), f_hour, dtype="float")
fuels = fuels_reg_y[:, np.newaxis, np.newaxis] * flat_shape
else: #If not flat shape, use yh load profile of enduse
Expand Down Expand Up @@ -598,24 +561,22 @@ def aggr_complete_result(
full_result_aggr : dict
Full result aggregation dict
"""
for _, sector_model in enumerate(sector_models):
for model_object in sector_model:

fuels = get_fuels_yh(model_object, 'techs_fuel_yh')
enduse_array_nr = lookup_enduses[model_object.enduse]

if isinstance(fuels, dict):
for tech, fuel_tech in fuels.items():
tech_fueltype = technologies[tech].fueltype_int
full_result_aggr[enduse_array_nr][tech_fueltype][reg_array_nr] += fuel_tech.reshape(8760)
else:
fueltype_yh_365_24 = get_fuels_yh(model_object, 'fuel_yh')
for model_object in sector_models:
fuels = get_fuels_yh(model_object, 'techs_fuel_yh')
enduse_array_nr = lookup_enduses[model_object.enduse]

if isinstance(fuels, dict):
for tech, fuel_tech in fuels.items():
tech_fueltype = technologies[tech].fueltype_int
full_result_aggr[enduse_array_nr][tech_fueltype][reg_array_nr] += fuel_tech.reshape(8760)
else:
fueltype_yh_365_24 = get_fuels_yh(model_object, 'fuel_yh')

# Iterate over fueltype and add to region
fueltype_yh_8760 = fueltype_yh_365_24.reshape(fueltype_yh_365_24.shape[0], 8760)
# Iterate over fueltype and add to region
fueltype_yh_8760 = fueltype_yh_365_24.reshape(fueltype_yh_365_24.shape[0], 8760)

for fueltype_nr, fuels_8760 in enumerate(fueltype_yh_8760):
full_result_aggr[enduse_array_nr][fueltype_nr][reg_array_nr] += fuels_8760
for fueltype_nr, fuels_8760 in enumerate(fueltype_yh_8760):
full_result_aggr[enduse_array_nr][fueltype_nr][reg_array_nr] += fuels_8760

return full_result_aggr

Expand Down Expand Up @@ -827,7 +788,7 @@ def aggregate_results_constrained(
(key of this will be made self.attributes)
reg_array_nr : int
Region array number
all_submodels : list
all_submodels : list[Enduse]
Submodel objects
mode_constrained : bool
Mode of how to run the model
Expand Down Expand Up @@ -857,97 +818,40 @@ def aggregate_results_constrained(
all_submodels,
technologies)

'''
if assumptions.curr_yr > 2015:
fueltype_nr = 2 #electricity
sum_all_enduses = np.sum(aggr_results['ed_submodel_enduse_fueltype_regs_yh'], axis=0) #full_result_aggr[enduse_array_nr][fueltype_nr][reg_array_nr] += fuels_8760
# print peak electrictiy demand of region
_mak_demand = np.max(sum_all_enduses[fueltype_nr][reg_array_nr])
_sum_dayily = np.sum(sum_all_enduses[fueltype_nr][reg_array_nr].reshape(365, 24), axis=1)
print(_sum_dayily.shape)
_peak_day = np.argmax(_sum_dayily)
print("info.... reg_array_nr: {} {}".format(reg_array_nr, assumptions.curr_yr))
print("MAX ELEC DEMAND " + str(_mak_demand))
print("_-- rs_space_heating " + str(aggr_results['ed_submodel_enduse_fueltype_regs_yh'].shape))
sum_rs_space_heating = aggr_results['ed_submodel_enduse_fueltype_regs_yh'][6]
print(sum_rs_space_heating.shape)
print(np.sum(sum_rs_space_heating[fueltype_nr][reg_array_nr]))
_mak_demand = np.max(sum_rs_space_heating[fueltype_nr][reg_array_nr])
print("MAX ELEC DEMAND RS SPACE: " + str(_mak_demand))
print("MAX ELEC DEMAND consumer electronics: " + str(np.max(aggr_results['ed_submodel_enduse_fueltype_regs_yh'][3][fueltype_nr][reg_array_nr])))
enduses_lu = {
'rs_lighting': 0, 'rs_cold': 1, 'rs_wet': 2, 'rs_consumer_electronics': 3,
'rs_home_computing': 4, 'rs_cooking': 5, 'rs_space_heating': 6, 'rs_water_heating': 7,
'ss_catering': 8, 'ss_small_power': 9, 'ss_cooled_storage': 10, 'ss_other_electricity': 11,
'ss_other_gas': 12, 'ss_ICT_equipment': 13, 'ss_space_heating': 14, 'ss_water_heating': 15,
'ss_cooling_humidification': 16, 'ss_lighting': 17, 'ss_fans': 18, 'is_other': 19, 'is_motors': 20,
'is_lighting': 21, 'is_drying_separation': 22, 'is_high_temp_process': 23, 'is_refrigeration': 24,
'is_low_temp_process': 25, 'is_compressed_air': 26, 'is_space_heating': 27}
#for enduse, enduse_nr in enduses_lu.items():
# plt.plot(aggr_results['ed_submodel_enduse_fueltype_regs_yh'][enduse_nr][fueltype_nr][reg_array_nr], label="{} ({})".format(
# enduse, enduse_nr))
#plt.plot(sum_all_enduses[fueltype_nr][reg_array_nr], linewidth=1, label="all")
all_end = aggr_results['ed_submodel_enduse_fueltype_regs_yh'][:, fueltype_nr, reg_array_nr]
print("shape all enduse" + str(all_end.shape))
print(sum_rs_space_heating[fueltype_nr][reg_array_nr].shape)
#for enduse_nr in range(28):
# plt.plot(all_end[enduse_nr], label="enduse_nr {}".format(enduse_nr))
#plt.plot(sum_all_enduses[fueltype_nr][reg_array_nr], label="Combined all end uses")
#plt.plot(aggr_results['ed_submodel_enduse_fueltype_regs_yh'][13][fueltype_nr][reg_array_nr], linewidth=4, label="tt")
#plt.plot(aggr_results['ed_submodel_enduse_fueltype_regs_yh'][3][fueltype_nr][reg_array_nr], linewidth=4, label="CONSUM")
#plt.plot(aggr_results['ed_submodel_enduse_fueltype_regs_yh'][23][fueltype_nr][reg_array_nr], linewidth=4, label="hoch temp")
#plt.plot(sum_rs_space_heating[fueltype_nr][reg_array_nr], linewidth=0.1, label="HEATING")
plt.plot(sum_all_enduses[fueltype_nr][reg_array_nr].reshape(365, 24)[_peak_day], linewidth=1, label="all_peak_day")
_z = np.zeros((24))
for enduse, enduse_nr in enduses_lu.items():
enduse_fuel = all_end[enduse_nr].reshape(365,24)
#_z += enduse_fuel[_peak_day]
#print("Z: {} {}".format(enduse_nr, _z[10:20]))
plt.plot(enduse_fuel[_peak_day], label="{} - ({})".format(enduse, enduse_nr))
plt.legend()
plt.show()
raise Exception("BLBLBAL")
'''

if mode_constrained:
submodel_to_idx = {
name: idx
for idx, name in enumerate(lookup_tables.basic_lookups()['submodels_names'])
}

# -----------------------------------------------------------------
# Aggregate fuel of constrained technologies for heating
# -----------------------------------------------------------------
for submodel_nr, submodel in enumerate(all_submodels):
for enduse_object in submodel:
for enduse_object in all_submodels:
submodel_nr = submodel_to_idx[enduse_object.submodel_name]

# Aggregate only over heating technologies
if enduse_object.enduse in enduse_space_heating:
# Aggregate only over heating technologies
if enduse_object.enduse in enduse_space_heating:

techs_fueltypes_yh = get_fuels_yh(
enduse_object, 'techs_fuel_yh')
techs_fueltypes_yh = get_fuels_yh(enduse_object, 'techs_fuel_yh')

# All used heating technologies
heating_techs = enduse_object.enduse_techs
# All used heating technologies
heating_techs = enduse_object.enduse_techs

# Iterate technologies and get fuel per technology
for heating_tech in heating_techs:
# Iterate technologies and get fuel per technology
for heating_tech in heating_techs:

tech_fuel = techs_fueltypes_yh[heating_tech] # Fuel of technology
fueltype_tech_int = technologies[heating_tech].fueltype_int # Fueltype of technology
tech_fuel = techs_fueltypes_yh[heating_tech] # Fuel of technology
fueltype_tech_int = technologies[heating_tech].fueltype_int # Fueltype of technology

# Aggregate Submodel (sector) specific enduse for fueltype
if heating_tech in aggr_results['results_constrained'].keys():
aggr_results['results_constrained'][heating_tech][submodel_nr][reg_array_nr][fueltype_tech_int] += tech_fuel.reshape(8760)
else:
aggr_results['results_constrained'][heating_tech] = np.zeros((len(all_submodels), reg_nrs, fueltypes_nr, 8760), dtype="float")
aggr_results['results_constrained'][heating_tech][submodel_nr][reg_array_nr][fueltype_tech_int] += tech_fuel.reshape(8760)
# Aggregate Submodel (sector) specific enduse for fueltype
if heating_tech in aggr_results['results_constrained'].keys():
aggr_results['results_constrained'][heating_tech][submodel_nr][reg_array_nr][fueltype_tech_int] += tech_fuel.reshape(8760)
else:
# TODO query this - are we overwriting otherwise
if heating_tech not in aggr_results['results_constrained']:
aggr_results['results_constrained'][heating_tech] = np.zeros((len(submodel_to_idx), reg_nrs, fueltypes_nr, 8760), dtype="float")
aggr_results['results_constrained'][heating_tech][submodel_nr][reg_array_nr][fueltype_tech_int] += tech_fuel.reshape(8760)

return aggr_results

Expand Down
Empty file removed energy_demand/plotting/__pycache__/basic_plot_functions.cpython-36.pyc.425738032
View file
Empty file.
18 changes: 7 additions & 11 deletions energy_demand/read_write/data_loader.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -224,7 +224,7 @@ def load_local_user_defined_vars(
var_name = file_name[:-4] #remove ".csv"

raw_file_content = pd.read_csv(os.path.join(path_csv, file_name))

strategy_vars_as_narratives[var_name] = narrative_related.read_user_defined_param(
raw_file_content,
simulation_base_yr=simulation_base_yr,
Expand Down Expand Up @@ -257,12 +257,12 @@ def load_ini_param(path):
config.read(os.path.join(path, 'model_run_sim_param.ini'))

regions = ast.literal_eval(config['REGIONS']['regions'])

assumptions = {}
assumptions['reg_nrs'] = int(config['SIM_PARAM']['reg_nrs'])
assumptions['base_yr'] = int(config['SIM_PARAM']['base_yr'])
assumptions['sim_yrs'] = ast.literal_eval(config['SIM_PARAM']['sim_yrs'])

# -----------------
# Other information
# -----------------
Expand Down Expand Up @@ -478,14 +478,10 @@ def convert_config_to_correct_type(config):
"""
out_dict = defaultdict(dict)

for path in config['PATHS']:
out_dict['PATHS'][path] = config.get('PATHS', path)

for path in config['DATA_PATHS']:
out_dict['DATA_PATHS'][path] = config.get('DATA_PATHS', path)

for path in config['CONFIG_DATA']:
out_dict['CONFIG_DATA'][path] = config.get('CONFIG_DATA', path)
string_sections = ['PATHS', 'DATA_PATHS', 'CONFIG_DATA', 'RESULT_DATA']
for section in string_sections:
for path in config[section]:
out_dict[section][path] = config.get(section, path)

for config_section in config['CONFIG']:
out_dict['CONFIG'][config_section] = config.getint('CONFIG', config_section)
Expand Down

0 comments on commit c4e4bae

Please sign in to comment.