SYSTEM AND METHOD FOR CONTROLLING POWER PRODUCTION FROM A WIND FARM

• US 20170337495A1
• Filed: 05/22/2017
• Published: 11/23/2017
• Est. Priority Date: 05/23/2016
• Status: Active Grant

• Current Assignee: GE Infrastructure Technology, LLC (GE Vernova, Inc.)
• Original Assignee: General Electric Company
• Inventors: GHOSH, Subhankar, MONDAL, Abhirup, DOGANAKSOY, Necip, LIU, Hongyan, ZHANG, Zhanpan, KAUCIC, Robert August, CAO, Jay Zhiqiang
• Granted Patent: US 11,334,834 B2
• Time in Patent Office: Days
• CPC Class Codes: F03D 7/048 controlling wind farms
• CPC Class Codes: F05B 2260/821 Parameter estimation or prediction
• CPC Class Codes: F05B 2270/335 Output power or torque
• CPC Class Codes: G06F 16/489 using time information
• CPC Class Codes: G06Q 10/06375 Prediction of business process outcome or impact based on a proposed change
• CPC Class Codes: G06Q 10/0639 Performance analysis of employees; Performance analysis of enterprise or organisation operations
• CPC Class Codes: H02J 2203/20 Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
• CPC Class Codes: H02J 2300/28 The renewable source being wind energy
• CPC Class Codes: H02J 3/004 Generation forecast, e.g. methods or systems for forecasting future energy generation
• CPC Class Codes: H02J 3/381 Dispersed generators
• CPC Class Codes: H02J 3/388 Islanding, i.e. disconnection of local power supply from the network
• CPC Class Codes: H02J 3/46 Controlling of the sharing of output between the generators, converters, or transformers
• CPC Class Codes: Y02E 10/76 Power conversion electric or electronic aspects
• 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 40/20 Information technology specific aspects, e.g. CAD, simulation, modelling, system security

Abstract

The system and method described herein relate to production of power from the wind farm that incorporate tunable power production forecasts for optimal wind farm performance, where the wind farm power production is controlled at least in part by the power production forecasts. The system and method use a tunable power forecasting model to generate tunable coefficients based on asymmetric loss function applied on actual power production data, along with tuning factor(s) that tune forecast towards under forecasting or over forecasting. The power production forecasts are generated using the tunable coefficients 34 and power characteristic features that are derived from actual power production data. The power production forecasts are monitored for any degradation, and a control action to regenerate the coefficients or retune the model is undertaken if degradation is observed.

Claims

1. A method for controlling power production for a wind farm wherein the wind farm is electrically coupled to a power grid, the method comprising:

receiving time series data of actual power produced by the wind farm at a plurality of time instances in a data receiving time interval;

deriving a plurality of power characteristic features from the time series data at each time instance from the plurality of time instances;

generating power production forecast data for the wind farm for a forecast time interval using the plurality of power characteristic features and a set of tunable coefficients derived using a tunable power forecasting model, wherein the tunable power forecasting model is based on an asymmetric loss function applied on a history of time series data of actual power produced by the wind farm, and wherein the tunable power forecasting model characterizes optimal wind farm performance; and

controlling wind farm power production based at least in part on the power production forecast data.

2. The method of claim 1 further comprising generating a forecasting performance output by monitoring the power production forecast data by using forecast deviation parameters for a plurality of forecast time intervals, wherein each forecast deviation parameter is derived from a difference between the power production forecast data and actual power produced for each forecast time interval.

3. The method of claim 2 wherein the forecasting performance output is generated by applying statistical process control on the forecast deviation parameters.

4. The method of claim 2 further comprising reconfiguring at least one of the tunable power forecasting model or the set of tunable coefficients based on the forecasting performance output.

5. The method of claim 2 wherein the forecast deviation parameters are compared with at least one of an over forecasting error and an under forecasting error, wherein the over forecasting error and the under forecasting error are user defined.

6. The method of claim 1 wherein the optimal wind farm performance is based on a combination of different weights assigned for an over forecasting error and an under forecasting error.

7. The method of claim 1 wherein the forecast time interval is in a range of about 5 minutes to about 30 minutes.

8. The method of claim 1 wherein controlling wind farm power production based at least in part on the power production forecast data comprises:

transmitting the power production forecast data to a power grid management system of the power grid;

receiving production set-points based on the power production forecast data from the power grid; and

controlling wind farm power production based on the production set-points.

9. A system for controlling power production for a wind farm based at least in part on power production forecast data, wherein the wind farm is electrically coupled to a power grid, the system comprising:

a power grid management system for generating production set-points for wind farm power production based on the power production forecast data; and

a wind farm management system for generating the power production forecast data and for controlling the wind farm power production, wherein the wind farm management system is communicatively coupled to the power grid management system, the wind farm management system comprising;

a controller for controlling wind farm power production based on production set points received from the power grid management module, and for generating actual power production data of the wind farm,

a feature extraction processor for generating a plurality of power characteristic features from the actual power production data for a data receiving time interval,

a tunable power forecasting model for generating a set of tunable coefficients for optimal wind farm performance, wherein the tunable power forecasting model is derived from applying an asymmetric loss function on a history of actual power production data of the wind farm, and

a forecast generation module for generating the power production forecast data for a plurality of forecast time intervals, wherein the power production forecast data for each forecast time interval is based on the plurality of power characteristic features and the set of tunable coefficients.

10. The system of claim 9 wherein the wind farm management system further comprises:

a forecast monitoring module for monitoring the power production forecast data over a defined time period by comparing actual power production data and corresponding power production forecast data, and generating a forecasting performance output, and

a forecast control module for generating a control action for initiating at least one of a re-configuration of the tunable power forecasting model, or regeneration of the set of tunable coefficients, wherein the control action is based on the forecasting performance output from the forecast monitoring module.

11. The system of claim 10 wherein the forecast monitoring module is used for obtaining forecast deviation parameters based on a difference between actual production data and corresponding power production forecast data, and for performing statistical process control on the forecast deviation parameters for generating the forecasting performance output.

12. The system of claim 10 wherein the control action comprises at least a trigger for reconfiguration of tunable power forecasting model or regeneration of tunable forecast coefficients.

13. The system of claim 12 wherein the trigger comprises at least one of a text message, an audio alarm or a visual alarm.

14. The system of claim 9 further comprising a data acquisition module for receiving and storing the actual power production data received from the controller, and wherein the data acquisition module is communicatively coupled to the feature extraction processor and the tunable power forecasting model.

15. The system of claim 9 wherein the wind farm management system further comprises a wind farm server to store the power production forecast data, actual power production data, and production set-points.

16. The system of claim 9 wherein the tunable power forecasting model is derived by using one or more tuning factors to characterize optimal wind farm performance.

17. A wind farm management system for controlling power production for a wind farm, the wind farm management system comprising:

a controller for generating actual power production data of the wind farm;

a feature extraction processor for generating a plurality of power characteristic features from the actual power production data for a data receiving time interval;

a tunable power forecasting model for generating a set of tunable coefficients for optimal wind farm performance, wherein the tunable power forecasting model is derived from applying an asymmetric loss function on a history of actual power production data of the wind farm; and

a forecast generation module for generating the power production forecast data for a plurality of forecast time intervals, wherein the power production forecast data for each forecast time interval is based on the plurality of power characteristic features and the set of tunable coefficients,

wherein the controller is used for controlling the wind farm power production based at least in part on the power production forecast data.

18. The wind farm management system of claim 17 further comprising:

a forecast monitoring module for monitoring the power production forecast data over a defined time period by comparing actual power production data and corresponding power production forecast data, and generating a forecasting performance output; and

a forecast control module for generating a control action for initiating at least one of a re-configuration of the tunable power forecasting model, or regeneration of the set of tunable coefficients, wherein the control action is based on the forecasting performance output from the forecast monitoring module.

19. The wind farm management system of claim 17 further comprising a data acquisition module for receiving and storing the actual power production data received from the controller, and wherein the data acquisition module is communicatively coupled to the feature extraction processor and the tunable power forecasting model.

20. The wind farm management system of claim 17 further comprising a wind farm server to store at least the power production forecast data, and actual power production data.