def test_stagewise_inflows_and_markov_prices(self):
markov_stagewise_model = hydrovalleymodel(hasstagewiseinflows=True,
hasmarkovprice=True)
solve_default(markov_stagewise_model,
iteration_limit=10,
print_level=0)
self.assertTrue(
isapprox(markov_stagewise_model.getbound(), 855.0, atol=1e-3))
def test_riskaverse(self):
"""
风险厌恶者
"""
riskaverse_model = hydrovalleymodel(
riskmeasure=EAVaR(lamb=0.5, beta=0.66))
solve_default(riskaverse_model, iteration_limit=10, print_level=0)
self.assertTrue(
isapprox(riskaverse_model.getbound(), 828.157, atol=1e-3))
def test_deterministic(self):
deterministic_model = hydrovalleymodel(hasmarkovprice=False,
hasstagewiseinflows=False)
status = solve_default(deterministic_model,
iteration_limit=10,
cut_selection_frequency=1,
print_level=0)
self.assertTrue(
isapprox(deterministic_model.getbound(), 835.0, atol=1e-3))
def test_worst_case(self):
worst_case_model = hydrovalleymodel(riskmeasure=EAVaR(lamb=0.5,
beta=0.0),
sense=Sense.Min)
solve_default(worst_case_model,
iteration_limit=10,
simulation=MonteCarloSimulation(frequency=2,
steps=list(
range(20, 51, 10))))
self.assertTrue(
isapprox(worst_case_model.getbound(), -780.867, atol=1e-3))
# Copyright 2017, Oscar Dowson, Zhao Zhipeng
# This Source Code Form is subject to the terms of the Mozilla Public
# License, v. 2.0. If a copy of the MPL was not distributed with this
# file, You can obtain one at http://mozilla.org/MPL/2.0/.
#############################################################################
from sddp.SDDP import solve_default
from sddp.example.HydroValley.hydro_valley import hydrovalleymodel
if __name__ == '__main__':
deterministic_model = hydrovalleymodel(hasmarkovprice=False,
hasstagewiseinflows=False)
status = solve_default(deterministic_model,
iteration_limit=100,
cut_selection_frequency=1,
print_level=0)
def test_DRO2(self):
dro_model = hydrovalleymodel(hasmarkovprice=False,
riskmeasure=DRO(1 / 6))
solve_default(dro_model, iteration_limit=20, print_level=0)
self.assertTrue(isapprox(dro_model.getbound(), 836.695, atol=1e-3))
def test_DRO(self):
dro_model = hydrovalleymodel(hasmarkovprice=False,
riskmeasure=DRO(math.sqrt(2 / 3) - 1e-6))
solve_default(dro_model, iteration_limit=10, print_level=0)
self.assertTrue(isapprox(dro_model.getbound(), 835.0, atol=1e-3))
def test_markov_prices(self):
markov_model = hydrovalleymodel(hasstagewiseinflows=False)
status = solve_default(markov_model, iteration_limit=10, print_level=0)
self.assertTrue(isapprox(markov_model.getbound(), 851.8, atol=1e-3))
def test_stagewise(self):
stagewise_model = hydrovalleymodel(hasmarkovprice=False)
solve_default(stagewise_model, iteration_limit=20, print_level=0)
self.assertTrue(isapprox(stagewise_model.getbound(), 838.33,
atol=1e-2))