PREDICTIVE GENERATOR EVENTS

• US 20180316220A1
• Filed: 04/26/2017
• Published: 11/01/2018
• Est. Priority Date: 04/26/2017
• Status: Active Grant

• Current Assignee: Discovery Energy LLC
• Original Assignee: Kohler Company
• Inventors: Frampton, Isaac S.
• Granted Patent: US 10,447,077 B2
• Time in Patent Office: Days
• CPC Class Codes: G05B 13/026 using a predictor
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: H02J 13/00016 using a wired telecommunication network or a data transmission bus
• CPC Class Codes: H02J 3/003 Load forecast, e.g. methods or systems for forecasting future load demand
• CPC Class Codes: H02J 3/466 Scheduling the operation of the generators, e.g. connecting or disconnecting generators to meet a given demand
• CPC Class Codes: H02K 99/10 Generators
• CPC Class Codes: Y02E 60/00 Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
• CPC Class Codes: Y04S 10/50 Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
• CPC Class Codes: Y04S 40/124 using wired telecommunication networks or data transmission busses

Abstract

An apparatus for predicting a load event in a generator system or a method for predicting the load event in the generator system may include monitoring output data associated with an electrical output of a generator, monitoring external sensor data associated with the generator, detecting a load event for the generator from the output data, identifying a subset of the external sensor data that preceded the load event from the output data, performing an analysis of the subset of the external sensor data, and determining a load event characteristic from the analysis, the load event characteristic indicative of a subsequent load event for the generator.

Claims

1. A method comprising:

monitoring data associated with a load supplied by an output of a generator;

detecting a load event from the data associated with the load;

performing an analysis of the load event; and

determining a load event characteristic from the analysis, the load event characteristic indicative of a subsequent load event for the generator; and

providing a signal when the subsequent load event is predicted.

2. The method of claim 1, further comprising:

monitoring external sensor data associated with the generator;

identifying a subset of the external sensor data that preceded the load event from the output data, wherein the analysis is performed on the subset of the external sensor data.

3. The method of claim 2, wherein the external sensor data includes an environmental condition of the load.

4. The method of claim 2, wherein the external sensor data includes at least one electrical parameter.

5. The method of claim 2, wherein the external sensor data includes scheduling information.

6. The method of claim 2, wherein the external sensor data includes occupancy information.

7. The method of claim 1, wherein determining the load event characteristic is performed according to a learning algorithm based on the data associated with a load supplied by the electrical output of the generator.

8. The method of claim 1, wherein the data includes electrical quantities associated with the electrical output of the generator.

9. The method of claim 8, wherein the electrical quantities associated with the electrical output of the generator include a voltage, a frequency, or a current of a power signal of the electrical output of the generator.

10. The method of claim 8, further comprising:

measuring a response characteristic of the generator; and

defining a time period based on the response characteristic of the generator.

11. The method of claim 1, wherein providing the signal when the subsequent load event is detected includes generating an instruction to start an additional generator and connect the additional generator to the load.

12. The method of claim 1, wherein providing the signal when the subsequent load event is detected includes generating an instruction to add or remove one or more loads from the bus.

13. An apparatus comprising:

a communication interface configured to receive output data associated with a load of a generator and receive external sensor data associated with the generator; and

a controller configured to detect a load event for the generator from the output data and identify a subset of the external sensor data that preceded the load event from the output data,

wherein an analysis of the subset of the external sensor data determines a load event characteristic, the load event characteristic indicative of a subsequent load event for the generator.

14. The apparatus of claim 13, wherein the output data associated with the load includes real power provided to the load.

15. The apparatus of claim 13, wherein the sensor data associated with the load includes an environmental condition associated with the load.

16. The apparatus of claim 13, wherein the sensor data associated with the load includes a scheduling condition associated with the load.

17. A generator system comprising:

a power bus;

a first generator coupled to the power bus;

a generator sensor configured to collect data indicative of an output of the first generator;

a sensor configured to collect additional data associated with the generator system;

wherein a load event detected from the data indicative of the output of the first generator is associated with a time window from the additional data associated with the generator system.

18. The generator system of claim 17, further comprising:

a controller configured to analyze the time window of additional data to identify a load trigger for the load event.

19. The generator system of claim 18, wherein the controller is configured to connect a second generator to the power bus in response to a subsequent occurrence of the load trigger.

20. The generator system of claim 18, wherein the controller is configured to calculate a certainty value for the load trigger and provide a generator command based on the load trigger and the certainty value.