Autonomous community energy management system and method
An embodiment provides an energy management system of a community in which a plurality of buildings, each divided into a plurality of zones, are located, the energy management system including: a community agent that is a device managing energy of the community and is configured to receive a first incentive about a Demand Response (DR) or a Frequency Regulation (FR) from a demand management device, calculate a first response capacity responding to the first incentive among a community management capacity, and transmit a second incentive to a building agent; a plurality of building agents, each of which is a device managing energy of a corresponding building and is configured to calculate a second response capacity responding to the second incentive among a building management capacity, and transmit a third incentive to a zone agent; and a plurality of zone agents, each of which is a device managing energy of a corresponding zone and is configured to calculate a third response capacity responding to the third incentive among a zone management capacity and transmit the third response capacity to a corresponding building agent, wherein each of the building agents is configured to receive third response capacities from the plurality of zone agents, and transmit a fourth response capacity obtained by adding the plurality of third response capacities and the second response capacity to the community agent, and the community agent is configured to receive fourth response capacities from the plurality of building agents, and transmit a fifth response capacity obtained by adding the plurality of fourth response capacities and the first response capacity to the demand management device.
1. An energy management system of a building divided into a plurality of zones,
the energy management system comprising:
a building agent that is a device managing energy of the building and is configured to receive a first incentive about a Demand Response (DR) or a Frequency Regulation (FR) from a community agent configured to manage energy of a community comprising the building, calculate a first response capacity responding to the first incentive among a building management capacity, and transmit a second incentive to a zone agent; and
a plurality of zone agents, each of which is a device managing energy of a corresponding zone and is configured to, when the each of zone agents does not participate in the demand response or the frequency regulation (case 1), estimate an energy demand amount (a first energy demand amount) of the corresponding zone through machine learning (first machine learning) or calculation, when the each of zone agents participates in the demand response or the frequency regulation (case 2), estimate an energy demand amount (a second energy demand amount) of the corresponding zone through machine learning (second machine learning), determine, as a second response capacity responding to the second incentive, a difference between the first energy demand amount and the second energy demand amount, and transmit the determined second response capacity to the building agent,
wherein the building agent is configured to receive second response capacities from the plurality of zone agents and transmit a third response capacity obtained by adding the plurality of second response capacities and the first response capacity to the community agent, and
the second machine learning comprises, as input data, the first energy demand amount and the second incentive and calculates the second energy demand amount to maximize an objective function comprising an incentive profit obtained by the second incentive and the second response capacity and a quantified user satisfaction degree.
2. The energy management system of claim 1 , wherein each of the zone agents is configured to set whether to allow each occupant of the corresponding zone to participate in the demand response or in the frequency regulation.
3. The energy management system of claim 1 , wherein each of the zone agents is configured to store user configurations for case 1 and case 2, respectively.
4. The energy management system of claim 3 , wherein each of the zone agents is configured to store the user configuration for case 1 to be a particular value and the user configuration for case 2 to be a range.
5. The energy management system of claim 4 , wherein each of the zone agents is configured to quantify a user satisfaction degree for each detail section in the range and store a quantified value of the degree.
6. The energy management system of claim 1 , wherein each of the zone agents is configured to further receive incentive information from the building agent,
the incentive information comprises a load capacity ranking and an incentive ranking of a zone corresponding to a relevant zone agent among all the zones of the building, and
the second energy demand capacity is calculated through the second machine learning comprising, as input data, the load capacity ranking and the incentive ranking.
7. The energy management system of claim 1 , wherein the second machine learning is independently executed for each zone and fails to recognize the second incentive of a different zone.
8. The energy management system of claim 1 , wherein the building agent and the plurality of zone agents are configured to determine an internal variable through machine learning, and machine learnings corresponding to the building agent and the plurality of zone agents are operated independently and in parallel to each other.