Collaborative load balancing within a community of energy nodes

1. A system for collaborative load balancing within a community of a plurality of energy nodes, the system comprising:

a central allocation server having a processor and a transmitter; and

a plurality of local agent servers, each of the local agent servers being connected to the central allocation server and to a respective one of the energy nodes, each of the local agent servers having a processor configured to:

receive input variables or parameters from the respective one of the energy nodes,

predict, using the received input variables or parameters, a non-zero energy generation amount that power generation equipment of the respective one of the energy nodes can generate over a planning horizon and an energy consumption amount that the respective one of the energy nodes will consume over the planning horizon,

solve, using the predicted energy generation amount and the predicted energy consumption amount, an optimization problem over the planning horizon, and

communicate a solution to the respective optimization problem to the central allocation server,

wherein the processor of the central allocation server is configured to:

receive, from each respective local agent server of the plurality of local agent servers, the solution to the optimization problem solved by the respective local agent server, and

run an allocation algorithm using the solutions received from each of the local agent servers to determine an amount of energy to be received from or supplied to individual ones of the energy nodes so as to provide the collaborative load balancing within the community, and

wherein the transmitter of the central allocation server is configured to transmit, to each of the local agent servers, an indication of the amount of energy which will be received from or supplied to the energy node associated therewith,

wherein each of the energy nodes includes power generation equipment, power transmission equipment, and power storage equipment,

wherein at least one of the central allocation server and the local agent servers include a memory device configured to store a credit balance for each of the energy nodes,

wherein the credit balance is respectively increased or decreased based on whether a respective one of the energy nodes has supplied energy to the community or received energy from the community, and

wherein the central allocation server is configured to favor requests to supply energy to the community for the energy nodes which have a negative credit balance and to favor requests to receive energy from the community for the energy nodes which have a positive balance.

2. The system according to claim 1 , wherein the central allocation server is configured to estimate a possible energy exchange for each of the energy nodes based on the credit balances.

3. The system according to claim 1 , wherein the local agent servers are configured to predict the energy generation amount using a weather forecast and are configured to predict the energy consumption amount using past usage data stored in a memory device located at one or more of the central allocation server and the local agent servers.

4. A central allocation server associated with a community of a plurality of energy nodes, the central allocation server comprising:

a receiver configured to receive, from a plurality of local agent servers, requests to supply energy to the community or to receive energy from the community, wherein each of the plurality of local agent servers is connected to the central allocation server and to a respective one of the energy nodes, wherein each of the requests received from the plurality of local agent servers is a solution provided by a respective one of the plurality of local agent servers, the solution obtained by solving an optimization problem by using a predicted energy generation amount and a predicted energy consumption amount, wherein the predicted energy generation amount is a non-zero amount of energy that power generation equipment of the respective one of the energy nodes is predicted to be able to generate over a planning horizon, and wherein the predicted energy consumption amount is an amount of energy that the respective one of the energy nodes is predicted to consume over the planning horizon;

a computational processor configured to run an allocation algorithm configured to determine an amount of energy to be received from or supplied to individual ones of the energy nodes so as to provide collaborative load balancing within the community;

a transmitter configured to communicate to each of the local agent servers the amount of energy which will be received from or supplied to the energy node associated therewith; and

memory devices containing a credit balance for each of the energy nodes,

wherein each of the energy nodes includes power generation equipment, power transmission equipment, and power storage equipment,

wherein the central allocation server is configured to respectively increase or decrease the credit balances based on whether a respective one of the energy nodes has supplied energy to the community or received energy from the community, and

wherein the central allocation server is configured to favor the requests to supply energy to the community for the energy nodes which have a negative credit balance and to favor the requests to receive energy from the community for the energy nodes which have a positive balance.

5. The central allocation server according to claim 4 , wherein the central allocation server is configured to estimate a possible energy exchange for each of the energy nodes based on the credit balances.

6. A method of providing a collaborative load balancing in a community of a plurality of energy nodes, the method comprising:

receiving, by a receiver of a central allocation server from a plurality of local agent servers, each of which is connected to the central allocation server and to a respective one of the energy nodes, requests to supply energy to the community or to receive energy from the community, wherein each of the requests received from the plurality of local agent servers is a solution provided by a respective one of the plurality of local agent servers, the solution obtained by solving an optimization problem by using a predicted energy generation amount and a predicted energy consumption amount, wherein the predicted energy generation amount is a non-zero amount of energy that power generation equipment of the respective one of the energy nodes is predicted to be able to generate over a planning horizon, and wherein the energy consumption amount is a predicted amount of energy that the respective one of the energy nodes is predicted to consume over the planning horizon;

determining, by a computational processor of a central allocation server, an amount of energy to be received from or supplied to individual ones of the energy nodes based on the requests received from the plurality of local agent servers and credit balances of the energy nodes so as to provide the collaborative load balancing within the community;

transmitting, to each of the local agent servers, an indication of the amount of energy which will be received from or supplied to the energy node associated therewith; and

increasing or decreasing the credit balances, stored in a memory of at least one of the central allocation server and the local agent servers, for each of the energy nodes respectively based on whether a respective one of the energy nodes has supplied energy to the community or received energy from the community;

wherein each of the energy nodes includes power generation equipment, power transmission equipment, and power storage equipment, and

wherein the central allocation server, during the determining, is configured to favor requests to supply energy to the community for the energy nodes which have a negative credit balance and to favor requests to receive energy from the community for the energy nodes which have a positive balance.

7. The method according to claim 6 , further comprising estimating, by the central allocation server, a possible energy exchange for each of the energy nodes based on the credit balances and communicating the estimation to respective ones of the local agent servers.

8. The method according to claim 6 , wherein the energy generation amount is predicted by the respective one of the plurality of local agent servers using a weather forecast and wherein the energy consumption amount is predicted by using past usage data stored in a memory device of at least one of the central allocation server and the local agent servers.