"RecursionError: maximum recursion depth exceeded" while adding a new constraints to a convex problem in cvxpy - constraints

Question

I'm new in cvxpy and working on a certain energy optimization model I'm trying to add new constraints. When I try to append the new constraints i get the following error "RecursionError: maximum recursion depth exceeded". If I exclude the last lines related (rules.appen(...)) everything works. I have tried to increase the system recursion limit but this doesn't work. Any suggestion?
class ProductionRamping(Constraint):
# it runs on both operation and planning, single- and multi-region
def __get_energy_prod_differences(self):
"""Creates variables to constrain the increase and decrease
in energy production when passing from one timestep to the next"""
# Create a dataframe containing the differences between the energy production of each tech
# in each timestep and the production in the previous timestep
energy_prod_differences = {}
# for each region
for reg in self.model_data.settings.regions:
energy_prod_differences_regional = {}
#for each tech, except demand, transmission, and storage
for tech_type in self.model_data.settings.technologies[reg].keys():
if tech_type == "Demand" or tech_type == "Transmission" or tech_type == "Storage":
continue
# select the first row
first_row = cp.reshape(
self.variables.technology_prod[reg][tech_type][0,:],
(1, self.variables.technology_prod[reg][tech_type][0,:].shape[0])
)
# in the case the timesteps are just one, the shifted dataframe containing the energy production
# will only be the first row; if timesteps are > 1 instead we have a shifted dataframe
if self.variables.technology_prod[reg][tech_type].shape[0] > 1:
shifted_prod = cp.vstack([first_row, (self.variables.technology_prod[reg][tech_type][:-1,:])])
else:
shifted_prod = first_row
# compute the difference between the energy production in each timestep and the the production in the previouos timestep
energy_prod_differences_regional[tech_type] = self.variables.technology_prod[reg][tech_type] - shifted_prod
energy_prod_differences[reg] = energy_prod_differences_regional
return energy_prod_differences ## 8760*nyears x ntechs
def transform_total_capacity(capacity, regions, timeslice_fraction, years):
totcapacity ={}
for reg in regions:
totcapacity_regional = {}
for tech_type in capacity[reg].keys():
if tech_type == "Demand" or tech_type == "Transmission" or tech_type == "Storage":
continue
totcapacity_hourly = cp.reshape(
capacity[reg][tech_type][0,:],
(1, capacity[reg][tech_type][0,:].shape[0]))
totcapacity_yearly = totcapacity_hourly
for hours in np.arange(len(timeslice_fraction)-1):
totcapacity_yearly = cp.vstack([totcapacity_yearly,totcapacity_hourly])
totcapacity_multiyear_regional = totcapacity_yearly
for indx in np.arange(len(years)-1):
totcapacity_hourly1 = cp.reshape(
capacity[reg][tech_type][indx,:],
(1, capacity[reg][tech_type][indx,:].shape[0]))
totcapacity_yearly1 = totcapacity_hourly1
for hours in np.arange(len(timeslice_fraction)-1):
totcapacity_yearly1 = cp.vstack([totcapacity_yearly1,totcapacity_hourly1])
totcapacity_multiyear_regional = cp.vstack ([totcapacity_multiyear_regional,totcapacity_yearly1])
totcapacity_regional[tech_type] = totcapacity_multiyear_regional
totcapacity[reg] = totcapacity_regional
return totcapacity
def _check(self):
assert hasattr(self.variables, 'totalcapacity'), "totalcapacity must be defined"
def rules(self):
rules = []
totcapacity = transform_total_capacity(
self.variables.totalcapacity,
self.model_data.settings.regions,
self.model_data.settings.timeslice_fraction,
self.model_data.settings.years)
for reg in self.model_data.settings.regions:
for tech_type, value in totcapacity[reg].items():
max_percentage_ramps = self.model_data.regional_parameters[reg]["prod_max_ramp"].loc[:, tech_type]
max_percentage_ramps = max_percentage_ramps.reindex(max_percentage_ramps.index.repeat(
len(self.model_data.settings.timeslice_fraction))
).values
min_percentage_ramps = self.model_data.regional_parameters[reg]["prod_min_ramp"].loc[:, tech_type]
min_percentage_ramps = min_percentage_ramps.reindex(min_percentage_ramps.index.repeat(
len(self.model_data.settings.timeslice_fraction))
).values
energy_prod_differences = self.__get_energy_prod_differences()
for indx, year in enumerate(self.model_data.settings.years):
max_ramp_in_timestep = cp.multiply ( value [indx * len(self.model_data.settings.timeslice_fraction) : (indx + 1)
* len(self.model_data.settings.time_steps), :] , max_percentage_ramps)
min_ramp_in_timestep = cp.multiply ( value [indx * len(self.model_data.settings.timeslice_fraction) : (indx + 1)
* len(self.model_data.settings.time_steps), :] , min_percentage_ramps)
energy_prod = energy_prod_differences[reg][tech_type][indx * len(self.model_data.settings.timeslice_fraction) : (indx + 1)
* len(self.model_data.settings.timeslice_fraction), :]
diff = energy_prod - max_ramp_in_timestep
diff1 = energy_prod * -1 - min_ramp_in_timestep
rules.append(
diff <= 0
)
rules.append(
diff1 <= 0
)
return rules