US-12627168-B2 - Energy storage control methods for optimal VPP energy management

Abstract

A method of operating a virtual power plant controller, that manages a power consumer connected to a power grid, a battery storage system, and an independent power plant, includes: obtaining a first data set including time-series information for each of power usage of the power consumer, power output of the independent power plant, power output capacity of the power grid, and state of charge of the battery storage system; training a machine learning (ML) model based on the time-series information using a ML algorithm, the ML model determines one or more parameters of an energy management system (EMS) policy to satisfy a power demand of the power consumer over a predetermined horizon; obtaining a second data set including power availability information; determining the one or more parameters of the EMS policy by inputting the second data set into the ML model; transmitting a command based on the EMS policy.

Inventors

• Othman Elkhomri

Assignees

• Banpu Innovation & Ventures LLC

Dates

Publication Date

20260512

Application Date

20230831

Claims (18)

1. 1 . A method of operating a virtual power plant (VPP) controller that manages a power consumer connected to a power grid, a battery storage system, and an independent power plant, the method comprising: obtaining a first data set including time-series information for each of: power usage of the power consumer; power output of the independent power plant; power output capacity of the power grid; and state of charge (SOC) of the battery storage system; training a machine learning (ML) model based on the time-series information using a ML algorithm, wherein the ML model determines one or more parameters of an energy management system (EMS) policy to satisfy a power demand of the power consumer over a predetermined horizon; obtaining a second data set including power availability information from each of the battery storage system, the power grid, and the independent power plant; determining, by inputting the second data set into the ML model, the one or more parameters of the EMS policy to satisfy the power demand of the power consumer; transmitting a command, based on the EMS policy, to at least one of the power grid, the battery storage system, and the independent power plant to control an amount of power supplied to the power consumer, wherein the EMS policy includes a comparison of the SOC of the battery storage system to the following ranges: SOC≥SOCref+ΔSOC, SOC-A: SOC≤SOCref−ΔSOC, SOC-B: SOCref−ΔSOC<SOC<SOCref+ΔSOC, SOC-C: wherein SOCref is a charge reference level of the battery storage system, and wherein ΔSOC is a range parameter that is one of the one or more parameters determined by the ML model, and wherein the EMS policy includes different commands based on the SOC satisfying one of the ranges SOC-A, SOC-B, and SOC-C.
2. 2 . The method of claim 1 , wherein the charge reference level, SOCref, is one of the one or more parameters determined by the ML model.
3. 3 . The method of claim 1 , wherein the command includes instructions for the independent power plant to provide power to the power consumer, and when the SOC satisfies the range SOC-A, the command includes additional instructions for the battery storage system to provide additional power to satisfy the power demand of the power consumer, when the SOC satisfies the range SOC-B, the command includes additional instructions for: the power grid to provide additional power to satisfy the power demand of the power consumer; and the power grid to provide power to charge the battery storage system with an entirety of a reference power allocation, when the SOC satisfies the range SOC-C, the command includes additional instructions for the power grid to split the reference power allocation between the power consumer and the battery storage system, wherein a first portion of the reference power allocation provides additional power to satisfy the power demand of the power consumer, and a second portion of the reference power allocation provides power to charge the battery storage system.
4. 4 . The method of claim 3 , wherein when the SOC satisfies the range SOC-A, the command includes additional instructions for the power grid to provide additional power to satisfy the power demand of the power consumer.
5. 5 . The method of claim 3 , wherein when the SOC satisfies the range SOC-A, the command includes additional instructions for the battery storage system to export energy to the power grid after satisfying the power demand of the power consumer.
6. 6 . The method of claim 3 , wherein the power consumer is an electric vehicle (EV) charging station, the time-series information used to train the ML model is based on a maximum charge rate of the EV charging station, and the reference power allocation is based on the maximum charge rate of the EV charging station.
7. 7 . A non-transitory computer readable medium storing instructions executable by a computer processor of a virtual power plant (VPP) controller that manages a power consumer connected to a power grid, a battery storage system, and an independent power plant, the instructions comprising functionality for: obtaining a first data set including time-series information for each of: power usage of the power consumer; power output of the independent power plant; power output capacity of the power grid; and state of charge (SOC) of the battery storage system; training a machine learning (ML) model based on the time-series information using a ML algorithm, wherein the ML model determines one or more parameters of an energy management system (EMS) policy to satisfy a power demand of the power consumer over a predetermined horizon; obtaining a second data set including power availability information from each of the battery storage system, the power grid, and the independent power plant; determining, by inputting the second data set into the ML model, the one or more parameters of the EMS policy to satisfy the power demand of the power consumer; transmitting a command, based on the EMS policy, to at least one of the power grid, the battery storage system, and the independent power plant to control an amount of power supplied to the power consumer, wherein the EMS policy includes a comparison of the SOC of the battery storage system to the following ranges: SOC≥SOCref+ΔSOC, SOC-A: SOC≤SOCref−ΔSOC, SOC-B: SOCref−ΔSOC<SOC<SOCref+ΔSOC, SOC-C: wherein SOCref is a charge reference level of the battery storage system, and wherein ΔSOC is a range parameter that is one of the one or more parameters determined by the ML model, and wherein the EMS policy includes different commands based on the SOC satisfying one of the ranges SOC-A, SOC-B, and SOC-C.
8. 8 . The non-transitory computer readable medium of claim 7 , wherein the charge reference level, SOCref, is one of the one or more parameters determined by the ML model.
9. 9 . The non-transitory computer readable medium of claim 7 , wherein the command includes instructions for the independent power plant to provide power to the power consumer, and when the SOC satisfies the range SOC-A, the command includes additional instructions for the battery storage system to provide additional power to satisfy the power demand of the power consumer, when the SOC satisfies the range SOC-B, the command includes additional instructions for: the power grid to provide additional power to satisfy the power demand of the power consumer; and the power grid to provide power to charge the battery storage system with an entirety of a reference power allocation, when the SOC satisfies the range SOC-C, the command includes additional instructions for the power grid to split the reference power allocation between the power consumer and the battery storage system, wherein a first portion of the reference power allocation provides additional power to satisfy the power demand of the power consumer, and a second portion of the reference power allocation provides power to charge the battery storage system.
10. 10 . The non-transitory computer readable medium of claim 9 , wherein when the SOC satisfies the range SOC-A, the command includes additional instructions for the power grid to provide additional power to satisfy the power demand of the power consumer.
11. 11 . The non-transitory computer readable medium of claim 9 , wherein when the SOC satisfies the range SOC-A, the command includes additional instructions for the battery storage system to export energy to the power grid after satisfying the power demand of the power consumer.
12. 12 . The non-transitory computer readable medium of claim 9 , wherein the power consumer is an electric vehicle (EV) charging station, the time-series information used to train the ML model is based on a maximum charge rate of the EV charging station, and the reference power allocation is based on the maximum charge rate of the EV charging station.
13. 13 . A virtual power plant (VPP) controller that manages a power consumer connected to a power grid, a battery storage system, and an independent power plant, the VPP controller comprising: a processor configured as: a power grid interface that communicates with the power grid; a battery storage interface that communicates with the battery storage system; a power consumer interface that communicates with the power consumer; and a power plant interface that communicates with the independent power plant; and a memory storing an energy management system (EMS) policy and instructions that, when executed, cause the processor to: obtain a first data set including time-series information for each of: power usage of the power consumer; power output of the independent power plant; power output capacity of the power grid; and state of charge (SOC) of the battery storage system; train a machine learning (ML) model based on the time-series information using a ML algorithm, wherein the ML model determines one or more parameters of the EMS policy to satisfy a power demand of the power consumer over a predetermined horizon; obtain a second data set including power availability information from each of the battery storage system, the power grid, and the independent power plant; determine, by inputting the second data set into the ML model, the one or more parameters of the EMS policy to satisfy the power demand of the power consumer; transmit a command, based on the EMS policy, to at least one of the power grid, the battery storage system, and the independent power plant to control an amount of power supplied to the power consumer, wherein the EMS policy includes a comparison of the SOC of the battery storage system to the following ranges: SOC≥SOCref+ΔSOC, SOC-A: SOC≤SOCref−ΔSOC, SOC-B: SOCref−ΔSOC<SOC<SOCref+ΔSOC, SOC-C: wherein SOCref is a charge reference level of the battery storage system, and wherein ΔSOC is a range parameter that is one of the one or more parameters determined by the ML model, and wherein the EMS policy includes different commands based on the SOC satisfying one of the ranges SOC-A, SOC-B, and SOC-C.
14. 14 . The VPP controller of claim 13 , wherein the charge reference level, SOCref, is one of the one or more parameters determined by the ML model.
15. 15 . The VPP controller of claim 13 , wherein the command includes instructions for the independent power plant to provide power to the power consumer, and when the SOC satisfies the range SOC-A, the command includes additional instructions for the battery storage system to provide additional power to satisfy the power demand of the power consumer, when the SOC satisfies the range SOC-B, the command includes additional instructions for: the power grid to provide additional power to satisfy the power demand of the power consumer; and the power grid to provide power to charge the battery storage system with an entirety of a reference power allocation, when the SOC satisfies the range SOC-C, the command includes additional instructions for the power grid to split the reference power allocation between the power consumer and the battery storage system, wherein a first portion of the reference power allocation provides additional power to satisfy the power demand of the power consumer, and a second portion of the reference power allocation provides power to charge the battery storage system.
16. 16 . The VPP controller of claim 15 , wherein when the SOC satisfies the range SOC-A, the command includes additional instructions for the power grid to provide additional power to satisfy the power demand of the power consumer.
17. 17 . The VPP controller of claim 15 , wherein when the SOC satisfies the range SOC-A, the command includes additional instructions for the battery storage system to export energy to the power grid after satisfying the power demand of the power consumer.
18. 18 . The VPP controller of claim 15 , wherein the power consumer is an electric vehicle (EV) charging station, the time-series information used to train the ML model is based on a maximum charge rate of the EV charging station, and the reference power allocation is based on the maximum charge rate of the EV charging station.

Description

BACKGROUND An intelligent energy system is defined as an approach in which intelligent electricity, thermal, and gas grids are combined with power storage technologies and coordinated to identify synergies between them in order to achieve an optimal solution for each individual sector as well as for the overall energy system. In aggregated systems, different power sources and power consumption units could potentially offer flexibility to the power system. However, in aggregated systems, power production from each power source may not be directly controlled. For example, the power generated by solar power plants (i.e., one or more arrays of solar panels) and wind power plants is dependent on weather conditions, time of day, and/or time of year. To maximize utilization of these variable independent power sources and minimize the risk of failing to satisfy the power demand of a power consumer at any given time, techniques for efficiently managing the power storage technologies are required. A virtual power plant is a system that connects different components of an intelligent grid ecosystem (power sources, power storage technologies (e.g., battery systems, electric vehicle batteries), and power consumers (e.g., buildings, electric vehicles, intelligent systems)) and coordinates optimal control solutions for the overall power ecosystem. In other words, the aim of the virtual power plant control system is to manage the power flow through the grid components in order to increase the economic performance and sustainability of the grid. SUMMARY In general, embodiments of the invention relate to a method of operating a virtual power plant (VPP) controller that manages a power consumer connected to a power grid, a battery storage system, and an independent power plant. The method includes: obtaining a first data set including time-series information for each of power usage of the power consumer, power output of the independent power plant, power output capacity of the power grid, and state of charge (SOC) of the battery storage system; training a machine learning (ML) model based on the time-series information using a ML algorithm, the ML model determines one or more parameters of an energy management system (EMS) policy to satisfy a power demand of the power consumer over a predetermined horizon; obtaining a second data set including power availability information from each of the battery storage system, the power grid, and the independent power plant; determining, by inputting the second data set into the ML model, the one or more parameters of the EMS policy to satisfy the power demand of the power consumer; transmitting a command, based on the EMS policy, to at least one of the power grid, the battery storage system, and the independent power plant to control an amount of power supplied to the power consumer. The EMS policy includes a comparison of the SOC of the battery storage system to the following ranges: SOC-A: SOC≥SOCref+ΔSOC; SOC-B: SOC≤SOCref−ΔSOC; SOC-C: SOCref−ΔSOC<SOC<SOCref+ΔSOC, where SOCref is a charge reference level of the battery storage system, and ΔSOC is a range parameter that is one of the one or more parameters determined by the ML model. The EMS policy includes different commands based on the SOC satisfying one of the ranges SOC-A, SOC-B, and SOC-C. In addition, embodiments of the invention relate to a non-transitory computer readable medium storing instructions executable by a computer processor of a virtual power plant (VPP) controller that manages a power consumer connected to a power grid, a battery storage system, and an independent power plant. The instructions comprising functionality for: obtaining a first data set including time-series information for each of power usage of the power consumer, power output of the independent power plant, power output capacity of the power grid, and state of charge (SOC) of the battery storage system; training a machine learning (ML) model based on the time-series information using a ML algorithm, the ML model determines one or more parameters of an energy management system (EMS) policy to satisfy a power demand of the power consumer over a predetermined horizon; obtaining a second data set including power availability information from each of the battery storage system, the power grid, and the independent power plant; determining, by inputting the second data set into the ML model, the one or more parameters of the EMS policy to satisfy the power demand of the power consumer; transmitting a command, based on the EMS policy, to at least one of the power grid, the battery storage system, and the independent power plant to control an amount of power supplied to the power consumer. The EMS policy includes a comparison of the SOC of the battery storage system to the following ranges: SOC-A: SOC≥SOCref+ΔSOC; SOC-B: SOC≤SOCref−ΔSOC; SOC-C: SOCref−ΔSOC<SOC<SOCref+ΔSOC, where SOCref is a charge reference level of the battery storage system, and ΔSOC is a r