FLEXIBLE, SECURE ENERGY MANAGEMENT SYSTEM

• US 20160274608A1
• Filed: 03/16/2015
• Published: 09/22/2016
• Est. Priority Date: 03/16/2015
• Status: Active Grant

• Current Assignee: The Florida International University Board of Trustees (State University System of Florida)
• Original Assignee: The Florida International University Board of Trustees (State University System of Florida)
• Inventors: MOHAMMED, Osama, YOUSSEF, Tarek A., ELSAYED, Ahmed
• Granted Patent: US 9,915,965 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G05F 1/66 Regulating electric power
• CPC Class Codes: G06Q 30/0206 Price or cost determination based on market factors

Abstract

Systems and techniques are disclosed for flexible, secure energy management that enable control of individual devices in a customer site'"'"'s energy infrastructure, which can allow optimization (e.g., by minimizing energy cost or usage) across the infrastructure. A system management component may centralize optimization and control capabilities across the energy infrastructure by determining an energy model and directing consumption, storage, and energy draw activities for the totality of the energy infrastructure devices. Systems may include modular, secure terminal devices that can be attached to or paired with the devices in an energy infrastructure in order to provide input data such as power load metrics, as well as to control the operation of the devices. A dynamic, adaptive prediction of system parameters, such as future demand, cost, generation capacity, and other metrics, may be provided. Dynamic, adaptive prediction may be performed without detailed historical usage data. A system gateway device may centralize communications between the system devices so that energy customer privacy is maintained.

Claims

1. A system for energy management comprising:

one or more control unit, wherein each control unit is paired to a particular energy consuming device, energy storage device, or energy source, wherein each control unit receives an input data, including at least a power flow data, from, and directs a power flow through, the particular energy consuming device, energy storage device, or energy source; and

a system management unit comprising a processing system, a network interface, one or more computer readable storage media, and program instructions stored on the one or more computer readable storage media that direct the processing system to;

determine an energy model for a time period based on;

a total power demand calculated from the input data received from the one or more control unit,

a current utility energy price received from a smart meter or a web service,

a current local energy generation capacity, and

a cost model for energy storage; and

modify, in accordance with the energy model for the time period, at least one of

an energy consumption activity for one or more energy consuming device,

an energy storage activity for one or more energy storage device, and

an energy draw activity for one or more energy source,

wherein the modifying is performed by sending an instruction set to a particular one of the one or more control unit that is paired to the particular energy consuming device, energy storage device, or energy source.

2. The system of claim 1, wherein determining the energy model for the time period is further based on:

one or more predicted value of an energy management system parameter, received from a forecasting component, for a prospective time period.

3. The system of claim 2, wherein the energy management system parameter includes a predicted total power demand, a predicted local energy generation capacity, and a predicted utility energy price.

4. The system of claim 2, wherein the forecasting component comprises an artificial neural network with dynamic adaptation of the weights and biases.

5. The system of claim 1, wherein the cost model for energy storage comprises a power loss rate and a depreciation cost per discharge cycle for one or more energy storage devices.

6. The system of claim 1, wherein a type of the control unit is a smart terminal unit comprising:

a sensing unit;

a device demand control component;

a network interface;

a processing system, one or more computer readable storage media, and program instructions stored on the one or more computer readable storage media that direct the processing system to;

in response to receiving a request from the system management unit, via the network interface, for the input data from the particular energy consuming device, energy storage device, or energy source to which the control unit is paired, retrieve the input data via the sensing unit and send the input data to the system management unit;

in response to receiving the instruction set from the system management unit, via the network interface, modify the energy consumption, energy storage, or energy draw activity by directing the power flow through the particular energy consuming device, energy storage device, or energy source via the device demand control component.

7. The system of claim 6, further comprising program instructions stored on the one or more computer readable storage media of the smart terminal unit that direct the processing system of the smart terminal unit to:

encrypt the input data before sending the input data; and

decrypt the instruction set from the system management unit before modifying the energy consumption, energy storage, or energy draw activity.

8. The system of claim 6, wherein the sensing unit detects one or more of:

the power flow through the particular energy consuming device, energy storage device, or energy source;

a temperature;

a humidity reading;

a current wind speed; and

a current storage level in the particular energy storage device.

9. The system of claim 6, further comprising program instructions stored on the one or more computer readable storage media of the system management unit that direct the processing system of the system management unit to:

encrypt the instruction set before sending the instruction set; and

decrypt the input data before determining the energy model.

10. The system of claim 1, further comprising a system gateway unit for facilitating communication between the system management unit, one or more control units, the smart meter, and the web service, the system gateway unit comprising:

a processing system, a network interface, one or more computer readable storage media, and program instructions stored on the one or more computer readable storage media that direct the processing system to;

transmit a message, via the network interface, between the system management unit and the one or more control unit, the smart meter, or the web service, wherein the message contains a request, the input data, or the instruction set.

11. The system of claim 1, further comprising program instructions stored on the one or more computer readable storage media that direct the processing system to display a user interface to select an optimization objective for the energy model;

and

wherein determining the energy model for the time period is further based on the optimization objective for the energy model.

12. A method for energy management comprising:

receiving an input data from one or more control unit;

receiving a current utility energy price from a smart meter or a web service;

determining an energy model for a time period based on a total power demand calculated from the input data, the current utility energy price, a current local energy generation capacity, and a cost model for energy storage; and

modifying, in accordance with the energy model for the time period, at least one of an energy consumption activity for one or more energy consuming device, an energy storage activity for one or more energy storage device, and an energy draw activity for one or more energy source.

13. The method of claim 12, wherein determining the energy model for the time period is further based on:

one or more predicted value of an energy management system parameter for a prospective time period.

14. The method of claim 13, wherein the energy management system parameter includes a predicted total power demand, a predicted local energy generation capacity, and a predicted utility energy price.

15. The method of claim 12, wherein the energy model is tuned in accordance with an optimization objective for the energy model.

16. The method of claim 15, wherein the optimization objective for the energy model is one or more of:

to minimize energy cost, to minimize power usage from a set of energy sources, to minimize total cost, and to maximize the use of local renewable resources.

17. A method for dynamic adaptation in training an energy management system forecasting component having an artificial neural network, the method comprising:

calculating a predicted value of an energy management system parameter for a prospective time period;

storing the predicted value in a buffer of a memory until the prospective time period has elapsed;

receiving an actual value for the energy management system parameter for the elapsed prospective time period;

calculating a resultant error between the predicted value and the actual value; and

adjusting a plurality of weights and biases of the artificial neural network using the resultant error.

18. The method of claim 17, wherein the energy management system parameter includes a total power demand, a local energy generation capacity, and a utility energy price.

19. The method of claim 17, wherein the predicted value is overwritten in the buffer at the time of the calculating of a next predicted value of the energy management system parameter for a next prospective time period.

20. The method of claim 17, wherein the adjusting of the plurality of weights and biases of the artificial neural network is performed using the resultant error as an input to a Levenberg-Marquardt algorithm.