Method of evaluating a software upgrade of a wind turbine

Patent metadata

Field	Value
Publication number	US-11157261-B2
Application number	US-201816651414-A
Country	US
Kind code	B2
Filing date	Sep 3, 2018
Priority date	Sep 27, 2017
Publication date	Oct 26, 2021
Grant date	Oct 26, 2021

Abstract

A method of estimating a parameter relating to an upgrade of a wind turbine software includes toggling the wind turbine alternatingly between two modes, where the upgrade is implemented in one but not the other. Data is collected during toggling and divided into pairs of data parts, one from each mode. The data from the two modes is ordered separately and a quantile-to-quantile comparison is made.

First claim

The invention claimed is:

1. A method of estimating a parameter relating to an upgrade of a wind turbine software, the method comprising:
toggling a wind turbine alternatingly between a first mode where the upgrade is not implemented and a second mode where the upgrade is implemented, wherein toggling the wind turbine alternating between the first mode and the second mode occurs at a frequency such that a wind speed experienced by the wind turbine during the first mode is the same as a wind speed experienced by the wind turbine during the second mode;
during the first and second modes, collecting data relating to a performance of the wind turbine, wherein the data collected during the first, mode comprises a first number of data parts and the data collected during the second mode comprises a second number of data parts;
discarding one or more data parts from the data collected during the first mode or the second mode such that the first number equals the second number:
deriving pairs of data parts from remaining data parts from the data collected during the first mode and the data parts from the data collected during the second mode, where each pair of data parts comprises a first data part corresponding to the data collected during the first mode and during a first period of time and a second data part corresponding to the data collected during the second mode and during a second period of time;
estimating the parameter from the pairs of data parts, wherein estimating the parameter comprises:
ordering the first data parts in the pairs of data parts in increasing or decreasing order to produce first ordered data parts;
ordering the second data parts in the pairs of data parts in increasing or decreasing order to produce second ordered data parts; and
comparing a first data part at a first position in the first ordered data parts and a second data part at the first position in the second ordered data parts; and
operating the wind turbine in the second mode if the estimated parameter indicates that the upgrade improved a power production of the wind turbine.

2. The method according to claim 1 , wherein estimating the parameter further comprises comparing a third data part at a second position in the first ordered data parts and a fourth data part at the second position in the second ordered data parts.

3. The method according to claim 1 , further comprising:
operating the wind turbine in the first mode while collecting the data,
deriving second pairs of data parts, where a first data part of a second pair corresponds to data collected during a third period of time and a second data part of the second pair corresponds to data collected during a fourth period of time,
ordering the first data parts of the second pairs in increasing or decreasing order, and
ordering the second data parts of the second pairs in increasing or decreasing order, wherein estimating the parameter further comprises comparing the data part of a first part of the second pairs at the first position in the ordered first parts of the second pairs and the data part of a second part of the second pairs at the first position in the ordered second parts of the second pairs.

4. A method of estimating a parameter relating to an upgrade of a wind turbine software, the method comprising:
toggling a wind turbine alternatingly between a first mode where the upgrade is not implemented and a second mode Where the upgrade is implemented, wherein toggling the wind turbine alternatingly between the first mode and the second mode occurs at a frequency such that a wind speed experienced by the wind turbine during the first mode is the same as a wind speed experienced by the wind turbine during the second mode;
during the first and second modes, collecting data relating to a performance of the wind turbine, wherein the data collected during the first mode comprises a first number of data parts and the data collected during the second mode comprises a second number of data parts;
discarding one or more data parts from the data collected during the first mode or the second mode such that the first number equals the second number;
deriving pairs of data parts from remaining data parts from the data collected during the first mode and the data, parts from the data collected during the second mode, where each pair of data parts comprises a first data part corresponding to the data collected during the first mode and during a first period of time and a second data part corresponding to the data collected during the second mode and during a second period of time;
estimating the parameter from the pairs of data, parts, wherein estimating the parameter comprises:
determining, for each pair of data parts, a difference value of a first data part and a second data part of that pair of data parts;
ordering the first data parts in the pairs of data parts in increasing or decreasing order to produce first ordered data parts;
ordering the second data parts in the pairs of data parts in increasing or decreasing order to produce second ordered data parts; and
fitting a curve to the difference values ordered according to the ordering of the first data parts and the second data parts; and

operating the wind turbine in the second mode if the estimated parameter indicates that the upgrade improved a power production of the wind turbine.

5. The method according to claim 4 , further comprising:
operating the wind turbine in the first mode while collecting data relating to a performance of the wind turbine,
deriving second pairs of data parts, where a first part of a second pair corresponds to the data collected during the first mode and derived during one period of time and a second part of the second pair corresponds to the data collected during the first mode and collected through a second period of time,
determining, for each second pair, a difference value of the first and second data parts of the second pair,
ordering the second pairs in accordance with a value determined from the first or second parts of the second pairs and
fitting a second curve to difference values ordered according to the of the second pairs, wherein estimating the parameter is based also on the second curve.

6. The method according to claim 4 , wherein first and second periods of time of the data parts of each pair are adjacent to each other.

7. The method according to claim 4 , wherein toggling between the first mode and the second mode occurs at least one time every hour.

8. The method according to claim 4 , wherein the first and second periods of time are no more than three hours apart.

9. The method according to claim 4 , wherein the collected data comprises a power production of the wind turbine.

10. A wind turbine comprising a controller configured to:
toggle the wind turbine alternatingly between a first mode where an upgrade is not implemented and a second mode Where the upgrade is implemented, wherein toggling the wind turbine alternatingly between the first mode and the second mode occurs at a frequency such that a wind speed experienced by the wind turbine during the first mode is the same as a wind speed experienced by the wind turbine during the second mode;
during the first and second modes, collect data relating to a performance of the wind turbine, wherein the data collected during the first mode comprises a first number of data parts and the data collected during the second mode, comprises a second number of data parts;
discard one or more data parts from the data collected during the first mode or the second mode such that the first number equals the second number;
derive pairs of data parts from remaining data parts from the data collected during the first mode

Assignees

• Vestas Wind Sys As

Inventors

• Berra Asier
• Bertelsen Kresten
• Nielsen Carsten Krough

Classifications

• H02J2101/28

  Wind energy · CPC title

• H02J3/003

  Load forecast, e.g. methods or systems for forecasting future load demand · CPC title

• F03D17/00

  Monitoring or testing of wind motors, e.g. diagnostics (testing during commissioning of wind motors F03D13/30) · CPC title

• Y02E10/72

  Wind turbines with rotation axis in wind direction · CPC title

• G06F8/65 Primary

  Updates (security arrangements therefor G06F21/57) · CPC title