Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Thermal model updates: feasibility & unit tests #374

Draft
wants to merge 3 commits into
base: master
Choose a base branch
from
Draft

Thermal model updates: feasibility & unit tests #374

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
fccd9ed
Updates for thermal management to address feasibility issues:
werdnum Jun 28, 2024
bcc462b
Fixes for cases where the def_start_timestep, def_end_timestep and si…
werdnum Jun 28, 2024
1ff13e1
Unit tests for thermal management
werdnum Jun 28, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
82 changes: 65 additions & 17 deletions src/emhass/optimization.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -157,6 +157,12 @@ def perform_optimization(self, data_opt: pd.DataFrame, P_PV: np.array, P_load: n
def_end_timestep = self.optim_conf['end_timesteps_of_each_deferrable_load']
type_self_conso = 'bigm' # maxmin

num_deferrable_loads = self.optim_conf['number_of_deferrable_loads']

def_total_hours = def_total_hours + [0] * (num_deferrable_loads - len(def_total_hours))
def_start_timestep = def_start_timestep + [0] * (num_deferrable_loads - len(def_start_timestep))
def_end_timestep = def_end_timestep + [0] * (num_deferrable_loads - len(def_end_timestep))

#### The LP problem using Pulp ####
opt_model = plp.LpProblem("LP_Model", plp.LpMaximize)

Expand All @@ -173,7 +179,7 @@ def perform_optimization(self, data_opt: pd.DataFrame, P_PV: np.array, P_load: n
name="P_grid_pos{}".format(i)) for i in set_I}
P_deferrable = []
P_def_bin1 = []
for k in range(self.optim_conf['number_of_deferrable_loads']):
for k in range(num_deferrable_loads):
if type(self.optim_conf['nominal_power_of_deferrable_loads'][k]) == list:
upBound = np.max(self.optim_conf['nominal_power_of_deferrable_loads'][k])
else:
Expand Down Expand Up @@ -402,7 +408,7 @@ def create_matrix(input_list, n):
rhs = 0)
for i in set_I})

elif "def_load_config" in self.optim_conf.keys():
elif "def_load_config" in self.optim_conf.keys() and len(self.optim_conf["def_load_config"]) > k:
if "thermal_config" in self.optim_conf["def_load_config"][k]:
# Special case of a thermal deferrable load
def_load_config = self.optim_conf['def_load_config'][k]
Expand All @@ -415,6 +421,9 @@ def create_matrix(input_list, n):
outdoor_temperature_forecast = data_opt['outdoor_temperature_forecast']
desired_temperatures = hc["desired_temperatures"]
sense = hc.get('sense', 'heat')

sense_coeff = (1 if sense == 'heat' else -1)

predicted_temp = [start_temperature]
for I in set_I:
if I == 0:
Expand All @@ -423,26 +432,63 @@ def create_matrix(input_list, n):
predicted_temp[I-1]
+ (P_deferrable[k][I-1] * (heating_rate * self.timeStep / self.optim_conf['nominal_power_of_deferrable_loads'][k]))
- (cooling_constant * (predicted_temp[I-1] - outdoor_temperature_forecast[I-1])))
if len(desired_temperatures) > I and desired_temperatures[I]:
constraints.update({"constraint_defload{}_temperature_{}".format(k, I):

# Set up a variable representing whether we are past the overshoot temperature.
is_overshoot = plp.LpVariable("defload_{}_overshoot_{}".format(k, I), cat='Binary')
constraints.update({
# If we are past the overshoot temperature, make sure is_overshoot is true.
f"constraint_defload{k}_overshoot_{I}_1":
plp.LpConstraint(
predicted_temp[I] - overshoot_temperature - (100 * sense_coeff * is_overshoot),
sense = plp.LpConstraintLE if sense == 'heat' else plp.LpConstraintGE,
rhs=0
),
# If we are not past the overshoot temperature, make sure is_overshoot is false.
f"constraint_defload{k}_overshoot_{I}_2":
plp.LpConstraint(
e = predicted_temp[I],
predicted_temp[I] - overshoot_temperature + (100 * sense_coeff * (1 - is_overshoot)),
sense = plp.LpConstraintGE if sense == 'heat' else plp.LpConstraintLE,
rhs = desired_temperatures[I],
)
})
constraints.update({"constraint_defload{}_overshoot_temp_{}".format(k, I):
plp.LpConstraint(
e = predicted_temp[I],
sense = plp.LpConstraintLE if sense == 'heat' else plp.LpConstraintGE,
rhs = overshoot_temperature,
)
for I in set_I})
rhs=0
),
# Either load was off last cycle or we haven't overshot
"constraint_defload{}_overshoot_temp_{}".format(k, I):
plp.LpConstraint(
e = is_overshoot + P_def_bin2[k][I-1],
sense = plp.LpConstraintLE,
rhs = 1,
)})

if len(desired_temperatures) > I and desired_temperatures[I]:
if hc.get('mode', 'constrain') == 'constrain':
constraints.update({"constraint_defload{}_temperature_{}".format(k, I):
plp.LpConstraint(
e = predicted_temp[I],
sense = plp.LpConstraintGE if sense == 'heat' else plp.LpConstraintLE,
rhs = desired_temperatures[I],
)
})
elif hc['mode'] == 'penalize':
penalty_factor = hc.get('penalty_factor', 10)
if penalty_factor < 0:
raise ValueError("penalty_factor must be positive, otherwise the problem will become unsolvable")
# If heating, sense_coeff = 1 and therefore predicted_temp < desired_temperature makes the below expression negative as desired.
penalty_value = (predicted_temp[I] - desired_temperatures[I]) * penalty_factor * sense_coeff
# ... but we don't 'reward' for going over the desired temperature, so we need a variable that is <= 0 and also <= the calculated penalty value.
penalty_var = plp.LpVariable("defload_{}_thermal_penalty_{}".format(k, I), cat='Continuous', upBound=0)
constraints.update({
"constraint_defload{}_penalty_{}".format(k, I):
plp.LpConstraint(
# Max value is penalty_var if penalty_var is negative, otherwise 0.
e = penalty_var - penalty_value,
sense = plp.LpConstraintLE,
rhs = 0
)
})
opt_model.setObjective(opt_model.objective + penalty_var)
predicted_temps[k] = predicted_temp

else:

if def_total_hours[k] > 0:
if len(def_total_hours) > k and def_total_hours[k] > 0:
# Total time of deferrable load
constraints.update({"constraint_defload{}_energy".format(k) :
plp.LpConstraint(
Expand Down Expand Up @@ -642,6 +688,8 @@ def create_matrix(input_list, n):
})
opt_model.constraints = constraints

print(repr(opt_model))

## Finally, we call the solver to solve our optimization model:
# solving with default solver CBC
if self.lp_solver == 'PULP_CBC_CMD':
Expand Down
Loading
Loading