Skip to content

Commit

Permalink
Merge pull request #151 from nismod/weather_model_runs
Browse files Browse the repository at this point in the history 
Weather model runs
  • Loading branch information
eggimasv committed Jan 23, 2019
2 parents 0c9a0f9 + 0012bc1 commit 0a0c938
Show file tree
Hide file tree
Showing 13 changed files with 841 additions and 217 deletions.
230 changes: 175 additions & 55 deletions energy_demand/enduse_func.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@
import math
import numpy as np
import matplotlib.pyplot as plt
import copy

from energy_demand.basic import date_prop
from energy_demand.profiles import load_factors as lf
Expand Down Expand Up @@ -127,8 +128,6 @@ def __init__(
#print("------INFO {} {} {} {}".format(self.enduse, sector, region, curr_yr))
#print("FUEL TRAIN A0: " + str(np.sum(self.fuel_y)))

#if curr_yr > 2015 and enduse == 'rs_lighting':
# print("tt")
# Get technologies of enduse
self.enduse_techs = get_enduse_techs(fuel_tech_p_by)

Expand Down Expand Up @@ -230,21 +229,12 @@ def __init__(
# Demand management for non-technology enduse
self.fuel_yh = load_shifting(
enduse,
curr_yr,
strategy_vars['dm_improvement'],
fuel_yh,
mode_constrained=False,
param_lf_improved_cy=strategy_vars['dm_improvement'][enduse][curr_yr],
make_all_flat=make_all_flat)
#print("FUEL TRAIN Y" + str(np.sum(fuel_yh)))

'''print("-----otside-------------")
if curr_yr > 2015 and enduse == 'rs_space_heating':
plt.plot(self.fuel_yh.reshape(8760), label="after")
plt.plot(fuel_yh.reshape(8760), label="before")
plt.legend()
plt.show()
print("----")'''

#if curr_yr > 2015:
# _a = np.max(fuel_yh)
# _b = np.max(self.fuel_yh)
Expand Down Expand Up @@ -345,39 +335,183 @@ def __init__(
mode_constrained)

# --------------------------------------
# Demand Management
# Peak shifting
# --------------------------------------
if mode_constrained:
self.techs_fuel_yh = {}
self.fuel_yh = None

for tech in fuel_yh:
self.techs_fuel_yh[tech] = load_shifting(
# If no demand management improvenent, no peak shifting
if strategy_vars['dm_improvement'][enduse][curr_yr] == 0:
self.techs_fuel_yh = fuel_yh
else:
self.techs_fuel_yh = load_shifting_multiple_tech(
enduse,
curr_yr,
strategy_vars['dm_improvement'],
fuel_yh[tech],
mode_constrained=True,
make_all_flat=make_all_flat)

self.fuel_yh = None
fueltypes,
self.enduse_techs,
assumptions.technologies,
fuel_yh,
param_lf_improved_cy=strategy_vars['dm_improvement'][enduse][curr_yr])
else:
#TODO TEST SHIFTING??
print("SHAPE " + str(self.fuel_yh))
self.fuel_yh = load_shifting(
enduse,
curr_yr,
strategy_vars['dm_improvement'],
fuel_yh,
mode_constrained=False,
param_lf_improved_cy=strategy_vars['dm_improvement'][enduse][curr_yr],
make_all_flat=make_all_flat)
print("SHAPE " + str(self.fuel_yh))

def load_shifting_multiple_tech(
enduse,
fueltypes,
enduse_techs,
technologies,
fuel_yh,
param_lf_improved_cy
):
"""Shift demand in case of multiple technologiesself.
Check how much of each technology is shifted in peak hours.
Calculate the absolute and relative demand of shifited demand
for every technology. Then substract and assign the new demands
per technology.
"""
fuel_tech_shifted = {}

for fueltype in fueltypes:

# Get technologies of fueltype
technologies_fueltype = tech_related.get_technologies_per_fueltype(
enduse_techs, technologies, fueltype)

if technologies_fueltype != []:

# Sum across al technologies
sum_all_techs = np.zeros((365, 24))
for tech in technologies_fueltype:
sum_all_techs += fuel_yh[tech]

# Shift demand for enduse (summed demand of technologies)
sum_all_techs_shifted = load_shifting(
enduse,
sum_all_techs,
mode_constrained=True,
param_lf_improved_cy=param_lf_improved_cy,
make_all_flat=False)
#plt.plot(sum_all_techs.reshape(8760), label="test taase")
#plt.plot(sum_all_techs_shifted.reshape(8760), label="test after")
#plt.legend()
#plt.show()

# Shift demand of peak (removed demand)
removed_demand = sum_all_techs - sum_all_techs_shifted
removed_demand = removed_demand.reshape(8760)
removed_demand[removed_demand < 0] = 0
tot_removed_demand = sum(removed_demand)
shape_removed = (1 / tot_removed_demand) * removed_demand
shape_removed[np.isnan(shape_removed)] = 0
peak_hour_positions = list(np.where(removed_demand > 0)[0])

# Demand in off-peak (demand added)
added_demand = sum_all_techs - sum_all_techs_shifted
added_demand = added_demand.reshape(8760)
added_demand[added_demand > 0] = 0
added_demand = abs(added_demand)
tot_added_demand = sum(added_demand)
shape_added = (1 / tot_added_demand) * added_demand
shape_added[np.isnan(shape_added)] = 0

assert round(np.sum(tot_removed_demand), 6) == round(tot_added_demand, 6)

for tech in technologies_fueltype:

#fuel_tech_8760 = fuel_yh[tech].reshape(8760)
#TODO Probably does not need to be copied but just assigned
fuel_tech_8760 = copy.copy(fuel_yh[tech].reshape(8760))
sum_all_techs_8760 = sum_all_techs.reshape(8760)
#print("-------------FUEL {} {}".format(np.sum(fuel_tech_8760), tech))
#print("shape_added: " + str(shape_added[:24]))
#print("ff " + str(np.sum(shape_added)))
#print("param_lf_improved_cy: " + str(param_lf_improved_cy))
#print("sum_all_techs_8760 {} {}".format(np.sum(sum_all_techs_8760), tech))
# Calculate share of technology specific demand to overall enduse demand per hour before shift
h_contribution_before = (1 / sum_all_techs_8760) * fuel_tech_8760
h_contribution_before[np.isnan(h_contribution_before)] = 0
#print("0 A " + str(np.sum(fuel_yh[tech])))
#print(np.sum(fuel_yh[tech]))
#print("A " + str(h_contribution_before))
# Get absolute demand in peak hours
tech_peak_shifted_demand = 0
for peak_hour_pos in peak_hour_positions:
tech_peak_shifted_demand += h_contribution_before[peak_hour_pos] * removed_demand[peak_hour_pos]

#print("peak_hour_positions: " + str(len(peak_hour_positions)))
#print("tech_peak_shifted_demand: " + str(tech_peak_shifted_demand))
#print("tot_removed_demand: " + str(tot_removed_demand))
# Percentage of total shifted demand of technology
p_abs_tech_shifted = (1.0 / tot_removed_demand) * tech_peak_shifted_demand

if np.isnan(p_abs_tech_shifted):
p_abs_tech_shifted = 0
#print("p_abs_tech_shifted: " + str(p_abs_tech_shifted))
#print(np.sum(fuel_yh[tech]))
#print(np.sum(fuel_tech_8760))
#print("--------")
#print(tot_removed_demand)
#print(p_abs_tech_shifted)
#print(p_abs_tech_shifted * tot_removed_demand)
#print(p_abs_tech_shifted * tot_removed_demand * shape_added)
#_a = p_abs_tech_shifted * tot_removed_demand * shape_added
#print("TEST davor" + str(np.sum(_a)))
#print(np.sum(fuel_tech_8760))
# Add shifted demand to total demand
fuel_tech_8760 += (p_abs_tech_shifted * tot_removed_demand) * shape_added
#plt.plot(fuel_tech_8760, label="intermediar")
#print("ZZZ")
#_a = shape_removed * tot_removed_demand
#print(shape_removed[:24])
#print(_a[:24])
##print(np.sum(fuel_tech_8760))
# Substract peak from total demand
fuel_tech_8760 -= (p_abs_tech_shifted * tot_removed_demand) * shape_removed

#print("KROKODIL")
#_max_before = np.max(fuel_yh[tech].reshape(8760))
#_max_after = np.max(fuel_tech_8760)
#print(_max_before)
#print(_max_after)
#print("PERCENT NEW " + str(((1/_max_before) * _max_after)))
#print(np.sum(fuel_tech_8760))
#print(np.sum(fuel_yh[tech]))
#plt.plot(fuel_yh[tech].reshape(8760), linewidth=4, label="before taase {}".format(tech))
#plt.plot(fuel_tech_8760, label="after")
#plt.legend()
#plt.show()
##print(np.sum(fuel_tech_8760))
#print(np.sum(fuel_yh[tech]))
assert round(np.sum(fuel_tech_8760), 6) == round(np.sum(fuel_yh[tech]), 6)

fuel_tech_shifted[tech] = fuel_tech_8760.reshape(365, 24)

'''if enduse == 'rs_space_heating':
plt.plot(sum_all_techs.reshape(8760), label="COMBINED BEFORE SHFITING")
# Sum across al technologies
sum_all_techs = np.zeros((365, 24))
for tech in technologies_fueltype:
sum_all_techs += fuel_tech_shifted[tech]
plt.plot(sum_all_techs.reshape(8760), label="COMBINED AFTER SHFITING")
#plt.plot(sum_all_techs_shifted.reshape(8760), label="test after")
plt.legend()
plt.show()'''
return fuel_tech_shifted

def load_shifting(
enduse,
curr_yr,
dm_improvement,
fuel_yh,
mode_constrained,
param_lf_improved_cy,
make_all_flat=False
):
"""This function shifts peak per of this enduse
Expand All @@ -386,25 +520,14 @@ def load_shifting(
Arguments
----------
enduse : str
Enduse
curr_yr : int
Current year
dm_improvement : dict
Assumptions of strategy variables
fuel_yh : array
Fuel per hours
enduse_techs : list
Enduse specfic technologies
sector : str
Sector
tech_stock : obj
Technology stock
load_profiles : obj
Load profiles
mode_constrained : bool
Running mode
If mode_constrained, always only one technology imported
Running mode, If mode_constrained, always only one technology imported
param_lf_improved_cy : dict
Assumptions of strategy variables (dm_improvement)
Load factor improvement parameter in current year
make_all_flat : bool
If true, all shapes are flat
Expand All @@ -413,11 +536,10 @@ def load_shifting(
fuel_yh : array
Fuel of yh
"""
# Get assumed load shift
if dm_improvement[enduse][curr_yr] == 0:
if param_lf_improved_cy == 0:
pass # no load management
else:

# Calculate average for every day
if mode_constrained:
average_fuel_yd = np.average(fuel_yh, axis=1)
Expand All @@ -428,9 +550,6 @@ def load_shifting(
loadfactor_yd_cy = lf.calc_lf_d(
fuel_yh, average_fuel_yd, mode_constrained)

# Load factor improvement parameter in current year
param_lf_improved_cy = dm_improvement[enduse][curr_yr]
#param_lf_improved_cy = 0.1
#print("Loading shifting .. {} {} {}".format(enduse, curr_yr, param_lf_improved_cy))

# Calculate current year load factors
Expand All @@ -447,10 +566,9 @@ def load_shifting(
fuel_yh,
mode_constrained)

'''
'''
if enduse == "rs_space_heating":
print("ENDUSE " + str(enduse))
print(curr_yr)
print(fuel_yh.shape)
_sum_dayily = np.sum(fuel_yh, axis=1)
print(_sum_dayily.shape)
Expand All @@ -469,14 +587,16 @@ def load_shifting(
print(param_lf_improved_cy)
print(loadfactor_yd_cy[_peak_day])
print(lf_improved_cy[_peak_day])
# Get maximum hour for electricity demand
__max_after = np.max(fuel_yh[_peak_day])
__max_before = np.max(_a[_peak_day])
print("before; {} after: {}".format(__max_before, __max_after))
print("peak reduction: " + str((100/__max_before) *__max_after))
_new_lf = lf.calc_lf_y_8760(fuel_yh[_peak_day])
_old_lf = lf.calc_lf_y_8760(_a[_peak_day])
print("before; {} after: {}".format(_old_lf, _new_lf))
print("lf before; {} after: {}".format(_old_lf, _new_lf))
print(make_all_flat)
'''

# -------------------------------------------------
Expand Down
31 changes: 16 additions & 15 deletions energy_demand/main.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -97,14 +97,20 @@ def energy_demand_model(
data = {}
sim_yrs = [2015, 2020, 2025, 2030, 2035, 2040, 2045, 2050]
#sim_yrs = [2015, 2020, 2050] #, 2050]
sim_yrs = [2015, 2030, 2050]
#sim_yrs = [2015, 2030, 2050]
sim_yrs = [2015, 2050]

if len(sys.argv) > 3: #user defined arguments are provide
scenario_name = str(sys.argv[2])
weather_realisation = str(sys.argv[3]) # Weather realisation
weather_realisation = str(sys.argv[3]) # Weather realisation
name_config_path = str(sys.argv[4])
else:
scenario_name = "_dm_10P_"
scenario_name = "_dm_5_"
weather_realisation = 'NF1'
#name_config_path = 'h_max'
#name_config_path = 'h_min'
#name_config_path = 'l_max'
#name_config_path = 'l_min'

print("-------------------------------------")
print("Information")
Expand All @@ -113,26 +119,18 @@ def energy_demand_model(
print("Configuration path: " + str(path_config))
print("Simulated yrs " + str(sim_yrs))

#name_config_path = 'h_c'
#name_config_path = 'l_c'
#name_config_path = 'h_max'
#name_config_path = 'h_min'
name_config_path = 'h_min_zero'
#name_config_path = 'h_min_5dm'
name_config_path = 'h_min_10dm'
#name_config_path = 'l_max'
#name_config_path = 'l_min'


# Local path configurations
if config['CRITERIA']['cluster_calc']:
path_weather_data = "/soge-home/staff/cenv0553/_weather_realisation"
else:
path_weather_data = "C:/Users/cenv0553/ED/data/scenarios"

if name_config_path == 'h_max' or name_config_path == 'l_max':
if name_config_path == 'h_max' or name_config_path == 'l_max' or name_config_path in ['h_max_0', 'h_max_10', 'h_max_0_only', 'h_max_10_only']:
local_scenario = 'pop-b_econ-c_fuel-c' #high
elif name_config_path == 'h_c' or name_config_path == 'l_c':
local_scenario = 'pop-baseline16_econ-c16_fuel-c16' #middle
#elif name_config_path == 'h_c' or name_config_path == 'l_c':
# local_scenario = 'pop-baseline16_econ-c16_fuel-c16' #middle
elif name_config_path == 'h_min' or name_config_path == 'l_min' or name_config_path in ['h_min_zero', 'h_min_5dm', 'h_min_10dm']:
local_scenario = 'pop-f_econ-c_fuel-c' #low

Expand Down Expand Up @@ -170,6 +168,9 @@ def energy_demand_model(
# ----------------------------------------------------------------------
data['scenario_data'] = defaultdict(dict)
data['enduses'], data['sectors'], data['fuels'], lookup_enduses, lookup_sector_enduses = data_loader.load_fuels(data['paths'])
#print("a")
#print(lookup_enduses)
#raise Exception
data['regions'] = read_data.get_region_names(name_region_set)
data['reg_coord'] = basic_functions.get_long_lat_decimal_degrees(read_data.get_region_centroids(name_region_set))

Expand Down

0 comments on commit 0a0c938

Please sign in to comment.