TypeError: '_SumExpression' object is not iterable - pyomo

Question

Please I am new to Pyomo. I have tried to run the following codes:
from pyutilib.misc import import_file
from pyomo.environ import *
model = ConcreteModel()
model.name = "Transmission Investment planning problem_"
#Sets
#Epoch
model.E = RangeSet(0,3) #No. of Epochs
#System nodes
model.N = ['N1', 'N2', 'N3'] #Names of Bus nodes
#model.n_name= Param(model.N)
model.G = ['G1', 'G2', 'G3'] #Names of generators
model.LI = ['L1', 'L2', 'L3'] #Nnames of Transmission lines
#Scalar Parameters
model.int_rate = 0.05 #interest rate
model.vll = 3000 #value of loss load(£/MWh)
model.tau_period = 8760 #Time duration of demand period (hours)
model.base = 100 #MVA base
model.ref = {'N3'} #reference node
model.vadegree = 0 #phase angle of reference node
#Discount factors
L = 5
Y= len(model.E)*L
irate = range(0,(Y-1))
def disc_factor(i):
disc = [1/((1 + model.int_rate)**i) for i in irate]
return disc
model.cum_disc_inv_cost = [sum(disc_factor(i)[(i*L):]) for i in model.E] #investment discount factor
model.cum_disc_op_cost = [sum(disc_factor(i)[(i*L):((i+1)*L)]) for i in model.E] #operation discount factor
#Demand Periods
model.t_demand = {'N1': 105, 'N2': 210, 'N3': 735} #demand at nodes (MW)
model.demand_curtailed = Var(model.E, model.N, initialize = 0) #curtailed demand (MW)
#Generation Units
model.ge_max = {'G1': 200, 'G2': 200, 'G3': 1000} #maximum stable power generation(MW)
model.ge_marginal_cost = {'G1': 30, 'G2': 35, 'G3': 40} #marginal cost of generation units (£/MWh)
#Bus to generation incidence matrix
model.B = {('N1','G1'): 1, ('N1','G2'): 0, ('N1','G3'): 0, ('N2','G1'): 0, ('N2','G2'): 1, ('N2','G3'): 0, ('N3','G1'): 0, ('N3','G2'): 0, ('N3','G3'): 1,}
#Transmission lines
model.li_x = {'L1': 0.2, 'L2': 0.2, 'L3': 0.2} #reactance of transmission line(p.u)
model.li_max_f = 150 #maximum capaciy provided for line expansion (MW)
model.li_f = {'L1': 100, 'L2': 100, 'L3': 100} #initial capacity for line l (MW)
model.li_sending_bus = {'L1': 'N1', 'L2': 'N1', 'L3': 'N2'} #sending bus for line l
model.li_receiving_bus = {'L1': 'N2', 'L2': 'N3', 'L3': 'N3'} #receiving bus for line l
model.li_length = {'L1': 100, 'L2': 100, 'L3': 100} #length of line l (km)
#Expansion Options
model.inv_cost_var = 4000000 #Annuitized variable investment cost for line l (£/MW.km.yr)
#Bus to line incidence matrix
model.I = {('N1','L1'): 1, ('N1','L2'): 1, ('N1','L3'): 0, ('N2','L1'): -1, ('N2','L2'): 0, ('N2','L3'): 1, ('N3','L1'): 0, ('N3','L2'): -1, ('N3','L3'): -1,}
#Variables
#Transmission line power flow limits
def fl_inv(model, i, l):
return (0, model.li_max_f)
model.li_f_inv = Var(model.E, model.LI, bounds = fl_inv) #transmission capacity to be built for line l (MW)
#Transmission line investment and operation contraints
model.f = Var(model.LI, model.E, initialize=0)
def fl_rule(model, l, j, i):
if i:
return model.f[l,j] >= -(model.li_f_inv[j,l] + model.li_f[l])
else:
return model.f[l,j] <= (model.li_f_inv[j,l] + model.li_f[l])
model.bound_f = Constraint(model.LI, model.E, [0,1], rule=fl_rule)
##generation limit
def fg(model, i, g):
return (0, model.ge_max[g])
model.ge_output = Var(model.E, model.G, initialize = 0, bounds = fg)
#phase angles for the nodes
def theta(model, e, n):
for n in model.N:
if n == model.ref:
model.theta[e, n].fixed = True
return model.vadegree
else: return 0
model.theta = Var(model.E, model.N, initialize = theta)
def line_equation(model, l, e):
return model.f[l, e] == model.base/model.li_x[l] *(sum(model.theta[e, n] for n in model.N if n == model.li_sending_bus[l]) - sum(model.theta[e, n] for n in model.N if n == model.li_receiving_bus[l]))
model.line_equation = Constraint(model.LI, model.E, rule = line_equation)
def system_balance(model, e, n):
return sum(model.B[n, g] * model.ge_output[e, g] for g in model.G) \
+ sum(model.I[n, l] * model.f[l, e] for l in model.LI) \
- model.t_demand[n] + model.demand_curtailed[e, n] == 0
model.SystemBalance = Constraint(model.E, model.N, rule=system_balance)
#OBJECTIVE FUNCTION
def objective_mincost(model):
for i in model.E:
return sum(model.cum_disc_inv_cost[i] * sum(model.li_f_inv[i, l] * model.inv_cost_var * model.li_length[l] for l in model.LI) + model.cum_disc_op_cost[i] * (model.tau_period * (sum(model.ge_max[g] * (model.ge_marginal_cost[g]) for g in model.G) + sum(model.demand_curtailed[i, n] for n in model.N * model.vll))))
model.objective = Objective(rule = objective_mincost, sense = minimize)
opt = SolverFactory('gurobi')
results = opt.solve(model) # solves and updates instance
model.display()
I got the following errorcodes:
ERROR: Rule failed when generating expression for objective objective:
TypeError: '_SumExpression' object is not iterable ERROR: Constructing component 'objective' from data=None failed:
TypeError: '_SumExpression' object is not iterable
Please any suggestions on where the problem might be and possible solution?
Thank you.

Answer 1

You have two places where model.E is multiplied by something that is not a set. You probably meant to parenthesize your sums differently (i.e., this is mostly a problem with how your sums are organized)