def construct_nvm():
from msppy.msp import MSLP
import numpy as np
nvm = MSLP(T=3, sense=-1, bound=500)
def sample_path_generator(random_state, size):
a = np.zeros([size, 3, 1])
for t in range(1, 3):
a[:, t, :] = (0.5 * a[:, t - 1, :] +
random_state.lognormal(2.5, 1, size=[size, 1]))
return a
nvm.add_Markovian_uncertainty(sample_path_generator)
for t in range(3):
m = nvm[t]
buy_now, buy_past = m.addStateVar(name='bought', obj=-1.0)
if t != 0:
sold = m.addVar(name='sold', obj=2)
unsatisfied = m.addVar(name='unsatisfied')
recycled = m.addVar(name='recycled', obj=0.5)
m.addConstr(sold + unsatisfied == 0,
uncertainty_dependent={'rhs': 0})
m.addConstr(sold + recycled == buy_past)
return nvm
def g(random_state, size):
a = numpy.empty([size, T, 2])
a[:, 0, :] = [[1.2, 1.06]]
for t in range(1, T):
noise = random_state.multivariate_normal(
mean=[1.0, 1.0],
cov=[[0.1, 0], [0, 0.01]],
size=size,
)
a[:, t, :] = 0.1 * a[:, t - 1, :] + noise
return a
AssetMgt = MSLP(T=T, sense=-1, bound=1e5, outputFlag=0)
AssetMgt.add_Markovian_uncertainty(g)
for t in range(T):
m = AssetMgt[t]
now, past = m.addStateVars(2, lb=0, obj=0)
if t == 0:
m.addConstr(now[0] + now[1] == 55)
if t in [1, 2]:
m.addConstr(past[0] + past[1] == now[0] + now[1],
uncertainty_dependent={
past[0]: 0,
past[1]: 1
})
if t == 3:
y = m.addVar(obj=1)
w = m.addVar(obj=-4)
m.addConstr(past[0] + past[1] - y + w == 80,
}
pandas.DataFrame(
[pandas.Series(result_simulated),pandas.Series(result)],
index=['Simulated','True']
).to_csv("./result/statistics_MCA.csv")
# evaluation table
evaluation_tab = []
evaluationTrue_tab = []
for lambda_ in [0.0,0.25,0.5,0.75]:
AssetMgt = MSLP(
T=T, sense=-1, bound=200
)
N = 100
K = 5
AssetMgt.add_Markovian_uncertainty(generator_augmented)
for t in range(T):
m = AssetMgt[t]
now, past = m.addStateVars(N+1, lb=0, obj=0, name='asset')
if t == 0:
buy = m.addVars(N, name='buy')
sell = m.addVars(N, name='sell')
m.addConstrs(now[j] == buy[j] - sell[j] for j in range(N))
m.addConstr(
now[N] == 100
- (1+fee) * gurobipy.quicksum(buy[j] for j in range(N))
+ (1-fee) * gurobipy.quicksum(sell[j] for j in range(N))
)
elif t != T-1:
sell = m.addVars(N, name='sell')
buy = m.addVars(N, name='buy')
def generator(random_state, size):
inflow = numpy.empty([size, T, 4])
inflow[:, 0, :] = inflow_initial[numpy.newaxis:, ]
for t in range(T - 1):
noise = numpy.exp(
random_state.multivariate_normal(mean=[0] * 4,
cov=sigma[t % 12],
size=size))
inflow[:, t + 1, :] = noise * ((1 - gamma[t % 12]) * exp_mu[t % 12] +
gamma[t % 12] * exp_mu[t % 12] / exp_mu[
(t - 1) % 12] * inflow[:, t, :])
return inflow
HydroThermal = MSLP(T=T, bound=0, discount=0.9906)
HydroThermal.add_Markovian_uncertainty(generator)
for t in range(T):
m = HydroThermal[t]
stored_now, stored_past = m.addStateVars(4,
ub=hydro_['UB'][:4],
name="stored")
spill = m.addVars(4, obj=0.001, name="spill")
hydro = m.addVars(4, ub=hydro_['UB'][-4:], name="hydro")
deficit = m.addVars(
[(i, j) for i in range(4) for j in range(4)],
ub=[
demand.iloc[t % 12][i] * deficit_['DEPTH'][j] for i in range(4)
for j in range(4)
],
obj=[deficit_['OBJ'][j] for i in range(4) for j in range(4)],
name="deficit")