Method and system for the hierarchical scheduling of energy flows in microgrids deploying battery energy storage systems

1 . A method for managing energy flows between an electrical grid and a plurality of microgrids each microgrid deploying at least one battery energy storage system and controlled by a microgrid controller, the method comprising:

receiving by a central controller electrical grid information from external sources, wherein the central controller comprises a virtual power plant (VPP) controller;

generating by the VPP controller using the electrical grid information an aggregate long-term schedule and an aggregate short-term schedule for energy flows and energy prices for the plurality of microgrids connected to the VPP controller, wherein the aggregate long-term schedule comprises at least a day-ahead plan, over a long-term scheduling horizon, defining at least grid power exchange, and power and energy pricing for each of the plurality of microgrids connected to the VPP controller, wherein the aggregate short-term schedule comprises a near real-time plan, over a short-term scheduling horizon shorter than the long-term scheduling horizon, defining updated power import and updated power export for the plurality of microgrids;

generating and training, by the VPP controller, a microgrid controller scheduler model for power import and export for the plurality of microgrids connected to the VPP controller;

receiving by a microgrid scheduler associated with each microgrid, the aggregate long-term schedule and the aggregate short-term schedule;

generating by each microgrid scheduler using the aggregate long-term schedule, a microgrid long-term schedule for controlling energy flows between the electrical grid and each microgrid and its connected battery energy storage system; and

correcting, by the microgrid scheduler, microgrid long-term schedule using a microgrid short-term near real-time schedule comprising an updated near real-time schedule generated at a higher update frequency than the microgrid long-term schedule to compensate for forecast uncertainties.

2 . The method of claim 1 , wherein the method further comprises:

training a microgrid controller scheduler model at the central controller; and

downloading the microgrid controller scheduler model to each microgrid controller.

3 . The method of claim 1 , wherein the microgrid long-term schedule of each microgrid scheduler for controlling energy flows comprises:

generating a long-term battery energy storage system charge/discharge schedule;

generating a long-term load dispatch schedule; and

generating a long-term controllable generation schedule.

4 . The method of claim 1 , wherein the microgrid short-term near real-time schedule corrects the microgrid long-term schedule for the effects of uncertainties in microgrid long-term forecasts.

5 . The method of claim 4 , wherein the method further comprises:

correcting the microgrid long-term schedule using a microgrid short-term near real-time schedule for the effects of uncertainties in the microgrid long-term forecasts using short-term forecasts including at least:

power generated from renewable generators;

the battery energy storage system state of charge;

response times of the power generated by the renewable generators and the battery energy storage system; and

manually entered long-term schedules.

6 . The method of claim 5 , wherein the microgrid long-term schedule is continuously updated by the microgrid short-term near real-time scheduler based on the short-term forecasts.

7 . The method of claim 6 , wherein the microgrid short-term near-real time schedule is compared to a schedule generated by a mixed integer non-linear programming (MINLP) solver using data from a historian.

8 . The method of claim 7 , wherein the method further comprises:

improving the microgrid scheduler by analyzing the performance difference between the microgrid short-term near real time schedule and the schedule to generate a value used for improving the microgrid scheduler.

9 . The method of claim 8 , wherein the method further comprises:

downloading the improved microgrid scheduler to the microgrid controller, wherein the microgrid controller is used to control the energy flows between the electrical grid and the microgrids for storing energy in the microgrid battery energy storage system and that dispatches energy from the microgrid battery energy storage system to the electrical grid and microgrid.

10 . A system for managing energy flows between an electrical grid and a plurality of microgrids each microgrid deploying at least one battery energy storage system and controlled by a microgrid controller, the system comprising:

a central controller comprising a virtual power plant (VPP) controller, the VPP controller arranged to receive electrical grid information from external sources and arranged to generate an aggregate long-term schedule and an aggregate short-term schedule for energy flows and energy prices for the plurality microgrids connected to the VPP controller;

wherein the aggregate long-term schedule comprises at least a day-ahead plan, over a long-term scheduling horizon, defining at least grid power exchange, and power and energy pricing for each of the plurality of microgrids connected to the VPP controller,

wherein the aggregate short-term schedule comprises a near real-time plan, over a short-term scheduling horizon shorter than the long-term scheduling horizon, defining updated power import and updated power export for the plurality of microgrids;

wherein the VPP controller comprises:

a VPP long-term scheduler configured to generate the aggregate long-term schedule for the energy flows and energy price for the plurality of microgrids connected to the VPP;

a VPP short-term scheduler configured to generate the aggregate short-term schedule for power import and export for the plurality of microgrids connected to the VPP controller; and

a microgrid scheduler associated with each microgrid connected to the VPP controller, each microgrid scheduler comprising:

a long-term day(s)-ahead scheduler configured to receive the aggregate long-term schedule and to generate a microgrid long-term schedule for controlling energy flows between the electrical grid and each microgrid and its connected battery energy storage system; and

a short-term near real-time scheduler configured to receive the aggregate short-term near real-time schedule and to correct microgrid long-term schedule using a microgrid short-term near real-time schedule, wherein the microgrid short-term near real-time schedule comprising an updated near real-time schedule generated at a higher update frequency than the microgrid long-term schedule to compensate for forecast uncertainties.

11 . The system of claim 10 , wherein the VPP controller generates a microgrid controller scheduler model for each microgrid connected to the VPP controller and each microgrid controller scheduler model is downloaded from the VPP controller to a respective microgrid controller.

12 . The system of claim 10 , wherein the microgrid long-term schedule generated by each microgrid scheduler for controlling energy flows comprises:

generating a long-term battery energy storage system charge/discharge schedule;

generating a long-term load dispatch schedule; and

generating a long-term controllable generation schedule.

13 . The system of claim 10 , wherein the system further includes a forecast uncertainty estimator running in the microgrid long-term day(s) ahead scheduler that estimates a measure of uncertainty from microgrid long-term forecasts.

14 . The system of claim 13 , wherein the forecast uncertainty estimator estimates the uncertainties in the microgrid long-term forecasts to correct the microgrid long-term schedule for the effects of uncertainties in the microgrid long-term forecasts including at least:

power generated from renewable generators;

the battery energy storage system state of charge;

response times of the power generated by the renewable generators and the battery energy storage system; and

manually entered long-term schedules.

15 . The system of claim 14 , wherein the system further comprises:

a first mixed integer non-linear programing MINLP solver connected to a historian, wherein the first MINLP solver is arranged to develop a schedule; and a second MINLP solver associated with the long-term day(s)-ahead scheduler and connected to the forecast uncertainty estimator, wherein the microgrid long-term schedule is generated using the uncertainty estimates for the microgrid long-term forecasts.

16 . The system of claim 15 , wherein a scheduler performance analyzer receives the schedule from the first MINLP solver and uses the microgrid short-term near real-time schedule with actual and ideal cost revenue inputs to generate a value that is a function of a difference between the total cost associated with an actual short-term near real-time schedule generated by the microgrid short-term near real-time scheduler and the schedule generated by the first MINLP solver.

17 . The system of claim 16 , wherein the microgrid short-term near real-time scheduler includes a recurrent neural network (RNN) that receives as an input the value generated by the scheduler performance analyzer to continuously improve the microgrid scheduler.

18 . The system of claim 17 , wherein each microgrid scheduler downloads an improved microgrid short-term near real-time scheduler to each microgrid controller associated with the microgrid scheduler, wherein the microgrid controller uses the microgrid controller scheduler model to generate a microgrid short-term near real-time schedule to control the energy flows between the electrical grid and the microgrids for storing energy in the microgrid battery energy storage system and for dispatching energy from the microgrid battery energy storage system to the electrical grid and microgrid.