Reward generating method for reducing peak load of power consumption and computing device for performing the same
AssigneeElectronics and Telecommunications Research Institute
InventorCheol-ho Shin
Life of this patent
Estimating the term…
Abstract
A reward generating method for reducing peak load of power consumption and a computing device for performing the same are provided. The reward generating method for reducing the peak load of the power consumption includes calculating an energy index according to a predetermined time interval using power consumption data of a specific building, to which an energy storage system (ESS) is applied, during a predetermined period, determining a reward index according to an action of the ESS using a reward function generated based on the energy index, and training a training model to which the reward function is applied through a reward generated based on the reward index.
Claims20
A reward generating method for reducing peak load of power consumption, the method comprising:
- calculating an energy index according to a predetermined time interval using power consumption data of a specific building, to which an energy storage system (ESS) is applied, during a predetermined period;
- determining a reward index according to an action of the ESS using a reward function generated based on the energy index;
- and
- training a training model to which the reward function is applied through a reward generated based on the reward index.
A computing device comprising:
- a processor;
- a memory configured to load a program executed by the processor;
- and
- a storage configured to store the program,
- wherein the program comprises instructions to cause the processor to perform calculating an energy index according to a predetermined time interval using power consumption data of a specific building, to which an energy storage system (ESS) is applied, during a predetermined period, determining a reward index according to an action of the ESS using a reward function generated based on the energy index, and training a training model to which the reward function is applied through a reward generated based on the reward index.
Dependencies are inferred from the claim text.