Temporal bias correction in wind forecasting

• US 10,578,771 B2
• Filed: 05/30/2019
• Issued: 03/03/2020
• Est. Priority Date: 08/01/2016
• Status: Active Grant

• Current Assignee: The Weather Co. LLC
• Original Assignee: International Business Machines Corporation
• Inventors: Cipriani, James P., Khabibrakhmanov, Ildar, Kim, Younghun, Praino, Anthony P., Treinish, Lloyd A.
• Primary Examiner(s): Islam, Mohammad K
• Application Number:
  US16/426,782
• Publication Number:
  US 20190317244A1
• Time in Patent Office: 278 Days
• CPC Class Codes: G01W 1/10 Devices for predicting weather conditions computers per se G06; display devices G09

Abstract

A set of characteristics of a period is extracted from a forecast that is applicable to the period and a geographical area. From a repository, a set of historical data is selected corresponding to a historical period, the set of historical data including a historical forecast related to the geographical area, and an actual measurement of wind at a location in the geographical area, the set of historical data having a subset of the set of characteristics. From the set of historical data, a function is computed to represent a discrepancy between the historical forecast and the actual measurement over the historical period. A bias value is computed from the function. By applying the bias to a value in the forecast, an adjusted forecast is produced of winds at a wind turbine located at a location in the geographical area.

Claims

1. A method of bias correction in wind forecasting, the method comprising:

extracting, from a wind forecast that is applicable to a future period and a geographical area, a set of weather characteristics of the future period;

selecting, from a repository, a set of historical data corresponding to a historical period, the set of historical data comprising a historical forecast related to the geographical area, and an actual measurement of wind at a location in the geographical area, the set of historical data having a subset of the set of weather characteristics;

computing, using a processor and a memory, from the set of historical data, a time function to represent a time discrepancy between the historical forecast reaching a value of wind speed and the actual measurement reaching the value of wind speed;

computing a bias value from the time function;

adjusting the wind forecast, using the processor and the memory, by applying the bias to a value in the wind forecast, to produce an adjusted wind forecast of winds at a wind turbine located at a location in the geographical area.

2. The method of claim 1, further comprising:

adjusting, using the adjusted forecast of winds, a power generation forecast.

3. The method of claim 1, further comprising:

computing a time difference between the historical forecast reaching a value and the actual measurement reaching the value, wherein the function represents a curve formed by a plurality of time differences, the time difference being a member of the plurality of time differences.

4. The method of claim 1, wherein the set of historical data comprises a plurality of pairs of historical forecast related to the geographical area and actual measurement of wind at a location in the geographical area, wherein each pair in the plurality of historical forecast applies to a different historical time.

5. The method of claim 1, further comprising:

receiving, from a weather forecasting service, the forecast including a wind forecast of winds during the period at the geographical area.

6. The method of claim 1, wherein the historical forecast related to the geographical area comprises:

a historical wind forecast, wherein the historical wind forecast is an extrapolated value that applies to the location in the geographical area, wherein a value at a second location is extrapolated to form the extrapolated value at the location.

7. A computer usable program product comprising one or more computer-readable storage medium, and program instructions stored on at least one of the one or more storage medium, the stored program instructions comprising:

program instructions to extract, from a wind forecast that is applicable to a future period and a geographical area, a set of weather characteristics of the future period;

program instructions to select, from a repository, a set of historical data corresponding to a historical period, the set of historical data comprising a historical forecast related to the geographical area, and an actual measurement of wind at a location in the geographical area, the set of historical data having a subset of the set of weather characteristics;

program instructions to compute from the set of historical data, a time function to represent a time discrepancy between the historical forecast reaching a value of wind speed and the actual measurement reaching the value of wind speed;

program instructions to compute a bias value from the time function;

program instructions to adjust the wind forecast, using the processor and the memory, by applying the bias to a value in the wind forecast, to produce an adjusted wind forecast of winds at a wind turbine located at a location in the geographical area.

8. The computer usable program product of claim 7, further comprising:

program instructions to adjust, using the adjusted forecast of winds, a power generation forecast.

9. The computer usable program product of claim 7, further comprising:

program instructions to compute a time difference between the historical forecast reaching a value and the actual measurement reaching the value, wherein the function represents a curve formed by a plurality of time differences, the time difference being a member of the plurality of time differences.

10. The computer usable program product of claim 7, wherein the set of historical data comprises a plurality of pairs of historical forecast related to the geographical area and actual measurement of wind at a location in the geographical area, wherein each pair in the plurality of historical forecast applies to a different historical time.

11. The computer usable program product of claim 7, further comprising:

program instructions to receive, from a weather forecasting service, the forecast including a wind forecast of winds during the period at the geographical area.

12. The computer usable program product of claim 7, wherein the historical forecast related to the geographical area comprises:

a historical wind forecast, wherein the historical wind forecast is an extrapolated value that applies to the location in the geographical area, wherein a value at a second location is extrapolated to form the extrapolated value at the location.

13. The computer usable program product of claim 7, wherein the computer usable code is stored in a computer readable storage medium in a data processing system, and wherein the computer usable code is transferred over a network from a remote data processing system.

14. The computer usable program product of claim 7, wherein the computer usable code is stored in a computer readable storage medium in a server data processing system, and wherein the computer usable code is downloaded over a network to a remote data processing system for use in a computer readable storage medium associated with the remote data processing system.

15. A computer system comprising one or more processors, one or more computer-readable memories, and one or more computer-readable storage devices, and program instructions stored on at least one of the one or more storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, the stored program instructions comprising:

program instructions to extract, from a wind forecast that is applicable to a future period and a geographical area, a set of weather characteristics of the future period;

program instructions to select, from a repository, a set of historical data corresponding to a historical period, the set of historical data comprising a historical forecast related to the geographical area, and an actual measurement of wind at a location in the geographical area, the set of historical data having a subset of the set of weather characteristics;

program instructions to compute from the set of historical data, a time function to represent a time discrepancy between the historical forecast reaching a value of wind speed and the actual measurement reaching the value of wind speed;

program instructions to compute a bias value from the time function;

program instructions to adjust the wind forecast, using the processor and the memory, by applying the bias to a value in the wind forecast, to produce an adjusted wind forecast of winds at a wind turbine located at a location in the geographical area.

16. The computer system of claim 15, further comprising:

program instructions to adjust, using the adjusted forecast of winds, a power generation forecast.

17. The computer system of claim 15, further comprising:

program instructions to compute a time difference between the historical forecast reaching a value and the actual measurement reaching the value, wherein the function represents a curve formed by a plurality of time differences, the time difference being a member of the plurality of time differences.

18. The computer system of claim 15, wherein the set of historical data comprises a plurality of pairs of historical forecast related to the geographical area and actual measurement of wind at a location in the geographical area, wherein each pair in the plurality of historical forecast applies to a different historical time.

19. The computer system of claim 15, further comprising:

program instructions to receive, from a weather forecasting service, the forecast including a wind forecast of winds during the period at the geographical area.

20. The computer system of claim 15, wherein the historical forecast related to the geographical area comprises:

a historical wind forecast, wherein the historical wind forecast is an extrapolated value that applies to the location in the geographical area, wherein a value at a second location is extrapolated to form the extrapolated value at the location.