Systems and methods for improved wind power generation

Patent metadata

Field	Value
Publication number	US-10443577-B2
Application number	US-201615198749-A
Country	US
Kind code	B2
Filing date	Jun 30, 2016
Priority date	Jul 17, 2015
Publication date	Oct 15, 2019
Grant date	Oct 15, 2019

Abstract

A wind power generation system includes one or both of a memory or storage device storing one or more processor-executable executable routines, and one or more processors configured to execute the one or more executable routines which, when executed, cause acts to be performed. The acts include receiving weather data, wind turbine system data, or a combination thereof; transforming the weather data, the wind turbine system data, or the combination thereof, into a data subset, wherein the data subset comprises a first time period data; selecting one or more wind power system models from a plurality of models; transforming the one or more wind power system models into one or more trained models at least partially based on the data subset; and executing the one or more trained models to derive a forecast, wherein the forecast comprises a predicted electrical power production for the wind power system.

First claim

What is claimed is:

1. A wind power forecasting system communicatively coupled to one or more wind turbine systems, wherein each of the one or more wind turbine systems comprises sensors for sensing environmental conditions or parameters related to the respective wind turbine system, the system comprising:
one or both of a memory or storage device storing one or more processor-executable executable routines; and
one or more processors configured to execute the one or more processor-executable routines which, when executed, cause acts to be performed comprising:
receiving weather data, wind turbine system data, or a combination thereof, from the sensors;
transforming the weather data, the wind turbine system data, or the combination thereof, into a data subset, wherein the data subset comprises a first time period data corresponding to a wind ramp event, wherein the transforming step comprises applying a slope-based analysis, a first order derivative, or a combination thereof, and wherein the data subset is shortened to enhance accuracy in case of higher slop values or higher first order derivatives of wind values;
selecting one or more wind power system models from a plurality of models;
transforming the one or more wind power system models into one or more trained models at least partially based on the data subset; and
executing the one or more trained models to derive a forecast, wherein the forecast comprises a predicted electrical power production for the wind turbine systems.

2. The system of claim 1 , wherein transforming the weather data, the wind turbine system data, or the combination thereof, into the data subset comprises applying an accuracy analysis to the weather data, to the wind turbine system data, or to the combination thereof, to select the first time period data.

3. The system of claim 1 , wherein the acts to be performed comprise comparing a first slope of the first time period data to a second slope of a second time period data, comparing a first mean value of the first time period data to a second mean value of the second time period data, comparing a first first order derivative of the first time period data to a second first order derivative of the second time period data, or a combination thereof, to select the first time period data, and wherein the second time period data comprises a different time period from the first time period.

4. The system of claim 1 , wherein selecting the one or more wind power system models comprises selecting the one or more wind power system models based on a predictive accuracy of the one or more wind power system models over the first time period data, based on a number of data points in the data subset, based on an amount of noise in the data subset, or a combination thereof.

5. The system of claim 1 , wherein the acts to be performed further comprise updating at least one of the one or more wind power system models based on the forecast.

6. The system of claim 1 , wherein the one or more wind power system models comprise a wind speed prediction model, a wind power prediction model, a wind direction prediction model, an availability forecasting model, a condition based monitoring (CBM) model, an actual power curve model, a farm level model, or a combination thereof.

7. The system of claim 6 , wherein the wind speed prediction model, the wind power prediction model, the wind direction prediction model, the availability forecasting model, the condition based monitoring (CBM) model, the actual power curve model, the farm level model, or the combination thereof, comprise an autoregressive (AR) model, an autoregressive integrated moving average (ARIMA) model, an autoregressive models for circular time series data (ARCS) model, an autoregressive conditional heteroskedasticity (ARCH) model, an autoregressive moving average (ARMA) model, a generalized autoregressive conditional heteroskedasticity (GARCH) model, a moving-average (MA) model, a neural network, a search-based model, or a combination thereof.

8. The system of claim 1 , wherein the forecast is configured to meet a Renewable Regulatory Fund (RRF) metric, a regulatory metric, or a combination thereof.

9. The system of claim 1 , further comprising a controller comprising the one or more processors, the one or both of the memory or storage device, or a combination thereof, wherein the controller is configured to control the wind turbine systems.

10. A method of foresting an electrical power production for wind turbine systems, comprising:
receiving weather data, wind turbine system data, or a combination thereof, from sensors;
transforming the weather data, the wind turbine system data, or the combination thereof, into a data subset, wherein the data subset comprises a first time period data corresponding to a wind ramp event, wherein the transforming step comprises applying a slope-based analysis, a first order derivative, or a combination thereof and wherein the data subset is shortened to enhance accuracy in case of higher slope values or higher first order derivatives of wind values;
selecting one or more wind power system models from a plurality of models;
transforming the one or more wind power system models into one or more trained models at least partially based on the data subset; and
executing the one or more trained models to derive a forecast, wherein the forecast comprises a predicted electrical power production for the wind turbine systems.

11. The method of claim 10 , wherein transforming the weather data, the wind turbine system data, or the combination, into the data subset comprises applying an accuracy analysis, a slope-based analysis, a mean value based analysis, a first order derivative, or a combination thereof.

12. The method of claim 11 , wherein applying the slope-based analysis, the mean value based analysis, the first order derivative, or the combination thereof, comprises comparing a first slope of the first time period data to a second slope of a second time period data, comparing a first mean value of the first time period data to a second mean value of the second time period data, comparing a first first order derivative of the first time period data to a second first order derivative of the second time period data, or a combination thereof, wherein the second time period data comprises a different time period from the first time period.

13. The method of claim 10 , wherein selecting the one or more wind power system models comprises analyzing a predictive accuracy of the one or more wind power system models over the first time period data, analyzing a number of data points in the data subset, analyzing an amount of noise in the data subset, or a combination thereof.

14. The method of claim 10 , further comprising updating at least one of the one or more wind power system models based on the forecast.

15. The method of claim 10 , wherein the one or more wind power system models comprise a wind speed prediction model, a wind power prediction model, a wind direction prediction model, an availability forecasting model, a condition based monitoring (CBM) model, an actual power curve model, a farm level model, or a combination thereof.

16. A tangible, non-transitory, computer-readable medium comprising instructions that when executed by a processor cause the processor to:
receive weather data, wind turbine system data, or a combination thereof, from sensors;
transform the weather data, the wind turbine system data, or the combination thereof, into a data subset, wherein the data subset comprises a first time period data corresponding to a wind ramp event, wherein t

Assignees

• Gen Electric

Inventors

• Swaminathan Krishna Kumar
• Sagi Deepak Raj
• Jain Pritesh
• Dasaratha Sridhar
• Bhaskar Nitika
• Srivastava Rahul Kumar
• Gogad Milesh Shrichandra

Classifications

• F03D7/045

  with model-based controls · CPC title

• F03D7/028 Primary

  controlling wind motor output power · CPC title

• F05B2260/84

  Modelling or simulation · CPC title

• F03D7/048

  controlling wind farms · CPC title

• F05B2260/821

  Parameter estimation or prediction · CPC title