#################################################################################
# The Institute for the Design of Advanced Energy Systems Integrated Platform
# Framework (IDAES IP) was produced under the DOE Institute for the
# Design of Advanced Energy Systems (IDAES), and is copyright (c) 2018-2021
# by the software owners: The Regents of the University of California, through
# Lawrence Berkeley National Laboratory, National Technology & Engineering
# Solutions of Sandia, LLC, Carnegie Mellon University, West Virginia University
# Research Corporation, et al. All rights reserved.
#
# Please see the files COPYRIGHT.md and LICENSE.md for full copyright and
# license information.
#################################################################################
from itertools import zip_longest
from pyomo.common.dependencies import attempt_import
from pyomo.common.config import ConfigDict, ConfigValue
import pyomo.environ as pyo
from idaes.apps.grid_integration.utils import convert_marginal_costs_to_actual_costs
prescient, prescient_avail = attempt_import("prescient")
[docs]class DoubleLoopCoordinator:
"""
Coordinate Prescient, tracker and bidder.
"""
def __init__(self, bidder, tracker, projection_tracker):
"""
Initializes the DoubleLoopCoordinator object and registers functionalities
in Prescient's plugin system.
Arguments:
bidder: an initialized bidder object
tracker: an initialized bidder object
projection_tracker: an initialized bidder object, this object is
mimicking the behaviro of the projection SCED in
Prescient and to projecting the system states
and updating bidder model.
Returns:
None
"""
self.bidder = bidder
self.tracker = tracker
self.projection_tracker = projection_tracker
[docs] def register_plugins(self, context, options, plugin_config):
"""
Register functionalities in Prescient's plugin system.
Arguments:
context: Prescient plugin PluginRegistrationContext from prescient.plugins.plugin_registration
options: Prescient options from prescient.simulator.config
plugin_config: Prescient plugin config
Returns:
None
"""
self.plugin_config = plugin_config
context.register_initialization_callback(self.initialize_customized_results)
context.register_before_ruc_solve_callback(self.pass_static_params_to_DA)
context.register_before_ruc_solve_callback(self.bid_into_DAM)
context.register_after_ruc_generation_callback(self.fetch_DA_prices)
context.register_after_ruc_generation_callback(self.fetch_DA_dispatches)
context.register_before_operations_solve_callback(self.pass_static_params_to_RT)
context.register_before_operations_solve_callback(self.bid_into_RTM)
context.register_after_operations_callback(self.track_sced_signal)
context.register_update_operations_stats_callback(self.update_observed_dispatch)
context.register_after_ruc_activation_callback(self.activate_pending_DA_data)
context.register_finalization_callback(self.write_plugin_results)
return
[docs] def get_configuration(self, key):
"""
Register customized commandline options.
Arguments:
key: plugin name
Returns:
config: Prescient config dict
"""
config = ConfigDict()
# Add command line options
config.declare(
"bidding_generator",
ConfigValue(
domain=str,
description="Specifies the generator we derive bidding strategis for.",
default=None,
),
).declare_as_argument("--bidding-generator")
return config
[docs] def initialize_customized_results(self, options, simulator):
"""
This method is outdated.
"""
simulator.data_manager.extensions["customized_results"] = {}
customized_results = simulator.data_manager.extensions["customized_results"]
customized_results["Generator"] = []
customized_results["Date"] = []
customized_results["Hour"] = []
customized_results["State"] = []
customized_results["RUC Schedule"] = []
customized_results["SCED Schedule"] = []
customized_results["Power Output"] = []
return
def _update_bids(self, gen_dict, bids, start_hour, horizon):
"""
This method takes bids and pass the information in the bids to generator
dictionary from Prescient.
Arguments:
gen_dict: a dictionary from Prescient's RUC/SCED instance that stores
the generator parameters.
bids: the bids we want to pass into the market.
start_hour: the effective start hour of the bid
horizon: the length of the bids
Returns:
None
"""
gen_name = self.bidder.bidding_model_object.model_data.gen_name
def _update_p_cost(gen_dict, bids, param_name, start_hour, horizon):
# update the "p_cost" element in the generator's dict
gen_dict["p_cost"] = {
"data_type": "time_series",
"values": [
{
"data_type": "cost_curve",
"cost_curve_type": "piecewise",
"values": bids[t][gen_name]["p_cost"],
}
for t in range(start_hour, horizon + start_hour)
],
}
# because the p_cost is updated, so delete p_fuel
if "p_fuel" in gen_dict:
gen_dict.pop("p_fuel")
return
def _update_time_series_params(gen_dict, bids, param_name, start_hour, horizon):
value_list = [
bids[t][gen_name].get(param_name, None)
for t in range(start_hour, start_hour + horizon)
]
if param_name in gen_dict:
gen_dict[param_name] = {
"data_type": "time_series",
"values": value_list,
}
return
def _update_non_time_series_params(
gen_dict, bids, param_name, start_hour, horizon
):
if param_name in gen_dict:
gen_dict[param_name] = bids[start_hour][gen_name].get(param_name, None)
return
param_update_func_map = {
"p_cost": _update_p_cost,
"p_max": _update_time_series_params,
"p_min": _update_time_series_params,
"fixed_commitment": _update_time_series_params,
"min_up_time": _update_non_time_series_params,
"min_down_time": _update_non_time_series_params,
"startup_capacity": _update_time_series_params,
"shutdown_capacity": _update_time_series_params,
"startup_fuel": _update_non_time_series_params,
"startup_cost": _update_non_time_series_params,
}
for param_name in bids[start_hour][gen_name].keys():
update_func = param_update_func_map[param_name]
update_func(gen_dict, bids, param_name, start_hour, horizon)
return
def _pass_DA_bid_to_prescient(self, options, ruc_instance, bids):
"""
This method passes the bids into the RUC model for day-ahead market clearing.
Arguments:
options: Prescient options from prescient.simulator.config.
ruc_instance: Prescient RUC object
bids: the bids we want to pass into the day-ahead market.
Returns:
None
"""
gen_name = self.bidder.bidding_model_object.model_data.gen_name
# fetch the generator's parameter dictionary from Prescient UC instance
gen_dict = ruc_instance.data["elements"]["generator"][gen_name]
self._update_bids(gen_dict, bids, start_hour=0, horizon=options.ruc_horizon)
return
[docs] def assemble_project_tracking_signal(self, options, simulator, hour):
"""
This function assembles the signals for the tracking model to estimate the
state of the bidding model at the begining of next RUC.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
hour: the simulation hour.
Returns:
market_signals: the market signals to be tracked.
"""
tracking_horizon = len(self.projection_tracker.time_set)
market_signals = self._assemble_sced_tracking_market_signals(
hour=hour,
sced_dispatch=None,
tracking_horizon=tracking_horizon,
)
return market_signals
[docs] def project_tracking_trajectory(self, options, simulator, ruc_hour):
"""
This function projects the full power dispatch trajectory from the
tracking model so we can use it to update the bidding model, i.e. advance
the time for the bidding model.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
ruc_hour: the hour RUC is being solved
Returns:
full_projected_trajectory: the full projected power dispatch trajectory.
"""
current_date = simulator.time_manager.current_time.date
current_hour = simulator.time_manager.current_time.hour
self._clone_tracking_model()
for hour in range(ruc_hour, 24):
# assemble market_signals
market_signals = self.assemble_project_tracking_signal(
options=options, simulator=simulator, hour=hour
)
# solve tracking
self.projection_tracker.track_market_dispatch(
market_dispatch=market_signals, date=current_date, hour=current_hour
)
# merge the trajectory
full_projected_trajectory = {}
for stat in self.tracker.daily_stats:
full_projected_trajectory[stat] = self.tracker.daily_stats.get(
stat
) + self.projection_tracker.daily_stats.get(stat)
# clear the projection stats
self.projection_tracker.daily_stats = None
return full_projected_trajectory
def _clone_tracking_model(self):
"""
Clone the model in tracker and replace that of projection tracker. In this
way, tracker and projection tracker have the same states before projection.
Arguments:
None
Returns:
None
"""
# iterate all the variables and params and clone the values
objects_list = [pyo.Var, pyo.Param]
for obj in objects_list:
for tracker_obj, proj_tracker_obj in zip_longest(
self.tracker.model.component_objects(
obj, sort=pyo.SortComponents.alphabetizeComponentAndIndex
),
self.projection_tracker.model.component_objects(
obj, sort=pyo.SortComponents.alphabetizeComponentAndIndex
),
):
if tracker_obj.name != proj_tracker_obj.name:
raise ValueError(
f"Trying to copy the value of {tracker_obj} to {proj_tracker_obj}, but they do not have the same name and possibly not the corresponding objects. Please make sure tracker and projection tracker do not diverge. "
)
for idx in tracker_obj.index_set():
if pyo.value(proj_tracker_obj[idx]) != pyo.value(tracker_obj[idx]):
proj_tracker_obj[idx] = round(pyo.value(tracker_obj[idx]), 4)
return
def _update_static_params(self, gen_dict):
"""
Update static parameters in the Prescient generator parameter data dictionary.
Args:
gen_dict: Prescient generator parameter data dictionary.
Returns:
None
"""
for param, value in self.bidder.bidding_model_object.model_data:
if param == "gen_name" or value is None:
continue
elif param == "p_cost":
curve_value = convert_marginal_costs_to_actual_costs(value)
gen_dict[param] = {
"data_type": "cost_curve",
"cost_curve_type": "piecewise",
"values": curve_value,
}
if "p_fuel" in gen_dict:
gen_dict.pop("p_fuel")
else:
gen_dict[param] = value
if param == "startup_cost" and "startup_fuel" in gen_dict:
gen_dict.pop("startup_fuel")
[docs] def pass_static_params_to_DA(
self, options, simulator, ruc_instance, ruc_date, ruc_hour
):
"""
This method pass static generator parameters to RUC model in Prescient
before it is solved.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
ruc_instance: Prescient RUC object.
ruc_date: the date of the day-ahead market we bid into.
ruc_hour: the hour the RUC is being solved in the day before.
Returns:
None
"""
gen_name = self.bidder.bidding_model_object.model_data.gen_name
gen_dict = ruc_instance.data["elements"]["generator"][gen_name]
self._update_static_params(gen_dict)
return
[docs] def bid_into_DAM(self, options, simulator, ruc_instance, ruc_date, ruc_hour):
"""
This function uses the bidding objects to bid into the day-ahead market
(DAM).
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
ruc_instance: Prescient RUC object.
ruc_date: the date of the day-ahead market we bid into.
ruc_hour: the hour the RUC is being solved in the day before.
Returns:
None
"""
# check if it is first day
is_first_day = simulator.time_manager.current_time is None
if not is_first_day:
# solve rolling horizon to get the trajectory
full_projected_trajectory = self.project_tracking_trajectory(
options, simulator, options.ruc_execution_hour
)
# update the bidding model
self.bidder.update_day_ahead_model(**full_projected_trajectory)
# generate bids
bids = self.bidder.compute_day_ahead_bids(date=ruc_date)
if is_first_day:
self.current_bids = bids
self.next_bids = bids
# pass to prescient
self._pass_DA_bid_to_prescient(options, ruc_instance, bids)
return
[docs] def fetch_DA_prices(self, options, simulator, result, uc_date, uc_hour):
"""
This method fetches the day-ahead market prices from unit commitment results,
and save it as a coordinator attribute.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
result: a Prescient RucPlan object.
ruc_date: the date of the day-ahead market we bid into.
ruc_hour: the hour the RUC is being solved in the day before.
Returns:
None
"""
current_bus = self.bidder.bidding_model_object.model_data.bus
is_first_day = simulator.time_manager.current_time is None
DA_prices = [
result.ruc_market.day_ahead_prices.get((current_bus, t)) for t in range(24)
]
if is_first_day:
self.current_avail_DA_prices = DA_prices
else:
self.current_avail_DA_prices = self.current_DA_prices + DA_prices
self.next_DA_prices = DA_prices
[docs] def fetch_DA_dispatches(self, options, simulator, result, uc_date, uc_hour):
"""
This method fetches the day-ahead dispatches from unit commitment results,
and save it as a coordinator attribute.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
result: a Prescient RucPlan object.
ruc_date: the date of the day-ahead market we bid into.
ruc_hour: the hour the RUC is being solved in the day before.
Returns:
None
"""
is_first_day = simulator.time_manager.current_time is None
gen_name = self.bidder.bidding_model_object.model_data.gen_name
gen_type = self.bidder.bidding_model_object.model_data.generator_type
gen_type_dispatch_mapping = {
"thermal": result.ruc_market.thermal_gen_cleared_DA,
"renewable": result.ruc_market.renewable_gen_cleared_DA,
"virtual": result.ruc_market.virtual_gen_cleared_DA,
}
dispatch_dict = gen_type_dispatch_mapping[gen_type]
DA_dispatches = [dispatch_dict.get((gen_name, t)) for t in range(24)]
if is_first_day:
# self.current_DA_dispatches = DA_dispatches
self.current_avail_DA_dispatches = DA_dispatches
else:
self.current_avail_DA_dispatches = (
self.current_DA_dispatches + DA_dispatches
)
self.next_DA_dispatches = DA_dispatches
def _pass_RT_bid_to_prescient(self, options, simulator, sced_instance, bids, hour):
"""
This method passes the bids into the SCED model for real-time market
clearing.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
sced_instance: Prescient SCED object
bids: the bids we want to pass into the real-time market.
hour: the hour of the real-time market.
Returns:
None
"""
gen_name = self.bidder.bidding_model_object.model_data.gen_name
# fetch generator's parameter dictionary from SCED instance
gen_dict = sced_instance.data["elements"]["generator"][gen_name]
self._update_bids(gen_dict, bids, start_hour=hour, horizon=options.sced_horizon)
return
[docs] def pass_static_params_to_RT(self, options, simulator, sced_instance):
"""
This method pass static generator parameters to SCED model in Prescient
before it is solved.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
sced_instance: Prescient SCED object.
Returns:
None
"""
gen_name = self.bidder.bidding_model_object.model_data.gen_name
gen_dict = sced_instance.data["elements"]["generator"][gen_name]
self._update_static_params(gen_dict)
return
[docs] def bid_into_RTM(self, options, simulator, sced_instance):
"""
This function bids into the real-time market. At this moment I just copy the
corresponding day-ahead bid here.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
sced_instance: Prescient SCED object.
Returns:
None
"""
# fetch the bids
date = simulator.time_manager.current_time.date
hour = simulator.time_manager.current_time.hour
bids = self.bidder.compute_real_time_bids(
date=date,
hour=hour,
realized_day_ahead_prices=self.current_avail_DA_prices,
realized_day_ahead_dispatches=self.current_avail_DA_dispatches,
)
# pass bids into sced model
self._pass_RT_bid_to_prescient(options, simulator, sced_instance, bids, hour)
return
[docs] def assemble_sced_tracking_market_signals(
self, options, simulator, sced_instance, hour
):
"""
This function assembles the signals for the tracking model.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
sced_instance: Prescient SCED object
hour: the simulation hour.
Returns:
market_signals: the market signals to be tracked.
"""
gen_name = self.bidder.bidding_model_object.model_data.gen_name
# fecth the sced signals for the generation from sced instance
sced_dispatch = sced_instance.data["elements"]["generator"][gen_name]["pg"][
"values"
]
tracking_horizon = len(self.tracker.time_set)
market_signals = self._assemble_sced_tracking_market_signals(
hour=hour,
sced_dispatch=sced_dispatch,
tracking_horizon=tracking_horizon,
)
return market_signals
def _assemble_sced_tracking_market_signals(
self, hour, sced_dispatch, tracking_horizon
):
"""
This function assembles the signals for the tracking model.
Arguments:
hour: the simulation hour
sced_dispatch: current sced dispatch (a list)
tracking_horizon: length of the tracking horizon
Returns:
market_signals: the market signals to be tracked.
"""
market_signals = []
# append the sced dispatch
if sced_dispatch is None:
dispatch = self.current_DA_dispatches[hour]
else:
dispatch = sced_dispatch[0]
market_signals.append(dispatch)
# append corresponding RUC dispatch
for t in range(hour + 1, hour + tracking_horizon):
if t < len(self.current_avail_DA_dispatches):
market_signals.append(self.current_avail_DA_dispatches[t])
return market_signals
[docs] def track_sced_signal(self, options, simulator, sced_instance, lmp_sced):
"""
This methods uses the tracking object to track the current real-time market
signals.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
sced_instance: Prescient SCED object
lmp_sced: Prescient SCED LMP object
Returns:
None
"""
current_date = simulator.time_manager.current_time.date
current_hour = simulator.time_manager.current_time.hour
# get market signals
market_signals = self.assemble_sced_tracking_market_signals(
options=options,
simulator=simulator,
sced_instance=sced_instance,
hour=current_hour,
)
# actual tracking
implemented_profiles = self.tracker.track_market_dispatch(
market_dispatch=market_signals, date=current_date, hour=current_hour
)
# update models
self.tracker.update_model(**implemented_profiles)
self.bidder.update_real_time_model(**implemented_profiles)
return
[docs] def update_observed_dispatch(self, options, simulator, ops_stats):
"""
This methods extract the actual power delivered by the tracking model and
inform Prescient, so Prescient can use this data to calculate the settlement
and etc.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
ops_stats: Prescient operation statitstic object
Returns:
None
"""
g = self.bidder.bidding_model_object.model_data.gen_name
# store the dictionaries that the observed/delivered power levels
observed_dispatch_level_dicts = [
ops_stats.observed_thermal_dispatch_levels,
ops_stats.observed_renewables_levels,
ops_stats.observed_virtual_dispatch_levels,
]
for observed_dispatch_level in observed_dispatch_level_dicts:
if g in observed_dispatch_level:
observed_dispatch_level[g] = self.tracker.get_last_delivered_power()
return
[docs] def activate_pending_DA_data(self, options, simulator):
"""
This function puts the day-ahead data computed in the day before into effect,
i.e. the data for the next day become the data for the current day.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
Returns:
None
"""
self.current_bids, self.next_bids = self.next_bids, None
self.current_DA_prices, self.next_DA_prices = self.next_DA_prices, None
self.current_DA_dispatches, self.next_DA_dispatches = (
self.next_DA_dispatches,
None,
)
# update the available DA signals
self.current_avail_DA_prices = self.current_DA_prices
self.current_avail_DA_dispatches = self.current_DA_dispatches
[docs] def write_plugin_results(self, options, simulator):
"""
After the simulation is completed, the plugins can write their own customized
results. Each plugin will have to have a method named 'write_results'.
Arguments:
options: Prescient options from prescient.simulator.config.
simulator: Prescient simulator.
Returns:
None
"""
self.bidder.write_results(path=options.output_directory)
self.tracker.write_results(path=options.output_directory)
return