Methods, systems, and computer readable media for model-free privacy preserving thermal load management
Methods, systems, and computer readable media for thermal load management of a collection of power consumers managed by an electric power aggregator. In some examples, a system includes a scheduling subsystem, implemented on one or more processors, configured for determining a plurality of cooling or heating control schedules for the collection of power consumers by forecasting one or more wholesale electricity price peaks. The system includes a control system implemented on one or more processors. The control system is configured for carrying out the cooling or heating control schedules at the individual consumer-level. The control system is configured for guaranteeing, for each power consumer of the collection of power consumers, a comfort constraint specified by a bilevel thermostat for the power consumer, wherein the comfort constraint comprises a lower level bound temperature and an upper bound temperature that the power consumer is willing to tolerate.
1 . A system for thermal load management of a collection of power consumers managed by an electric power aggregator, the system comprising:
a scheduling subsystem, implemented on one or more processors, configured for determining a plurality of cooling or heating control schedules for the collection of power consumers by forecasting one or more wholesale electricity price peaks, wherein forecasting the one or more wholesale electricity price peaks incudes using a long-short term memory (LSTM) based recurrent neural network architecture to forecast hourly electric prices from historical price data and weather forecasts and wherein determining the heating or cooling control schedules includes solving, using a heuristic relaxation, an optimization problem that, given the forecast hourly electric prices, minimizes a total cost of procuring electricity by an electric power load serving entity (LSE); and
a control subsystem, implemented on one or more processors, configured for:
carrying out the cooling or heating control schedules at the individual consumer-level, wherein carrying out the heating or cooling control schedules at the individual consumer level includes broadcasting, from the electric power LSE to smart bilevel thermostats, control commands for switching heating or cooling equipment on or off in accordance with distributed open loop control laws determined from results of the optimization problem; and
guaranteeing, for each power consumer of the collection of power consumers, a comfort constraint specified by a bilevel thermostat for the power consumer, wherein the comfort constraint comprises a lower bound temperature and an upper bound temperature that the power consumer is willing to tolerate.
2 . The system of claim 1 , wherein the scheduling subsystem is configured for determining the cooling or heating control schedules without sensing or modeling of a plurality of buildings of the collection of power consumers.
3 . The system of claim 1 , wherein the scheduling subsystem is configured for solving the optimization problem using the relaxation heuristic by approximating that a function that characterizes the forecast hourly electric prices is unimodal.
4 . The system of claim 1 , wherein forecasting the one or more wholesale electricity price peaks comprises forecasting day-ahead or real-time wholesale electricity price peaks.
5 . The system of claim 1 , wherein forecasting the one or more wholesale electricity price peaks comprises forecasting based on one or more outdoor temperature forecasts.
6 . The system of claim 1 , wherein the control subsystem is configured for carrying out the cooling or heating control schedules without sharing temperature or power consumption measurements from the collection of power consumers with the electric power aggregator.
7 . The system of claim 1 , wherein the control subsystem is configured for carrying out the cooling or heating control schedules without using thermal models of buildings of the power consumers.
8 . A method for thermal load management of a collection of power consumers managed by an electric power aggregator, the method comprising:
determining, at a scheduling subsystem implemented on one or more processors, a plurality of cooling or heating control schedules for the collection of power consumers by forecasting one or more wholesale electricity price peaks, wherein forecasting the one or more wholesale electricity price peaks incudes using a long-short term memory (LSTM) based recurrent neural network architecture to forecast hourly electric prices from historical price data and weather forecasts and wherein determining the heating or cooling control schedules includes solving, using a heuristic relaxation, an optimization problem that, given the forecast hourly electric prices, minimizes a total cost of procuring electricity by an electric power load serving entity (LSE);
carrying out the cooling or heating control schedules at the individual consumer-level using a control subsystem implemented on one or more processors, wherein carrying out the heating or cooling control schedules at the individual consumer level includes broadcasting, from the electric power LSE to smart bilevel thermostats, control commands for switching heating or cooling equipment on or off in accordance with distributed open loop control laws determined from results of the optimization problem; and
guaranteeing, for each power consumer of the collection of power consumers, a comfort constraint specified by a bilevel thermostat for the power consumer, wherein the comfort constraint comprises a lower bound temperature and an upper bound temperature that the power consumer is willing to tolerate.