WO2009016020A1

WO2009016020A1 - Wind turbine monitoring system

Info

Publication number: WO2009016020A1
Application number: PCT/EP2008/059086
Authority: WO
Inventors: Paul Happy
Original assignee: Vestas Wind Systems AS
Current assignee: Vestas Wind Systems AS
Priority date: 2007-07-31
Filing date: 2008-07-11
Publication date: 2009-02-05
Anticipated expiration: 2010-01-31

Abstract

A computer-implemented method of monitoring at least one operation parameter of one or more devices comprises the steps of obtaining measurement data of at least one operation parameter of the devices, and comparing a first value of said operation parameter to a second value of the operation parameter, and emitting a signal in case the first value deviates by more than a predetermined value from the second value. The one more devices may be comprised in groups with similar location, and they may be divided into groups according to characteristics, such as type of gearbox included in the turbines. Operation parameters e.g. include generator temperature or gearbox oil pressure.

Description

WIND TURBINE MONITORING SYSTEM

Technical field

The present invention relates to the field of monitoring of one or more wind turbines, such as wind turbines in a wind farm. More specifically, the invention provides a computer- implemented method for monitoring any data produced by a measurement device related to the operation of the wind turbine, such as transformer temperature, gearbox oil temperature, event logs or other parameters of wind turbines, and a wind power production facility comprising a computer system for carrying such a method.

Background of the invention

Wind turbine monitoring and maintenance is a field of increasing importance to wind turbine manufacturers and operators. Much effort is put into predicting life time of mechanical and electrical components of wind turbines, and predefined service sessions are conducted in an attempt to prevent system breakdown due to component failure, or at least to reduce the risk of breakdowns.

However, components may fail, even if their actual running time is far from the expected life time. Such failures usually occur in consequence of abnormal operation conditions or as a result of a chain of occurrences.

Therefore, operators of wind turbines often monitor certain operation parameters, such as a temperature known to be critical, e.g. the gearbox oil temperature of each wind turbine. In case such an operation parameter is found to exceed a value, which, according to the operator's experience, indicates potential failure of a component of the wind turbine, the operator may decide to take appropriate action to alleviate the possible malfunction, which is believed to constitute the cause of the unusually high or unusually low value of the parameter in question. Current processes for determining values indicating possible failure are deterred by the variable nature of the parameters being monitored.

Summary of the invention

The present inventor has found that current processes of determining values which can trigger further analysis is unreliable. It is hence an object of preferred embodiments of the present invention to provide a system and a method, which is more reliable in the sense that potential failures may be more precisely predicted. It is a further objection of preferred embodiments of the present invention to provide a partially automated system and a method.

In a first aspect, the invention provides a system for monitoring at least one wind turbine, the at least one wind turbine being characterised by a plurality of wind turbine characteristics, the system comprising at least one sensor element for obtaining measurement data of at least one operation parameter of the at least one wind turbine, the system further comprising:

- a computer system comprising at least a processor, a data input device, and an output device;

- data transmission elements for transmitting said measurement data of the at least one wind turbine to the data input device of the computer system; the processor of the computer system being configured to:

- compare a first value of said operation parameter to a second value of said operation parameter;

- cause the output device to emit an output signal in case the first value deviates by more than a predetermined value from said second value.

In a second aspect, the invention provides a computer-implemented method of monitoring at least one operation parameter of at least one wind turbine, the wind turbine being characterised by a plurality of wind turbine characteristics, the method comprising the steps of:

- obtaining measurement data of at least one operation parameter of the wind turbine;

- comparing a first value of said operation parameter to a second value of said operation parameter;

- emitting an output signal in case the first value deviates by more than a predetermined value from said second value.

It will be understood that preferred embodiments of the method of the second aspect of the invention are carried out by the system of the first aspect of the invention.

The first and second values may be actual, average, median, distribution or deviation values of a given operation parameter. The comparison between the first and second values may optionally be performed in a continuous manner. Thus, the processor of the computer system may be configured to continuously compare a first value of an operation parameter to a second value of said operation parameter.

The computer system allows collection of large amounts of data in a continuous manner as well as fast processing thereof. As the computer system collects measurement data, optionally in a continuous manner, and compares a first value to a second value, monitoring no longer relies solely on experience or pre-defined threshold values, which may be subject to significant inaccuracies, but rather on a comparison between measured data. Hence, if for example a generator temperature of a wind turbine rises beyond a certain value, such temperature rise may be acceptable, if the average generator temperature also rises. Indeed, such temperature rise may occur as a consequence of a number of operating and/or ambient conditions, which do not occur regularly or which only last for a relatively short period of time. Whereas manual monitoring of the wind turbine would be likely to result in emission of an error signal or even shutting down of the wind turbine, because the parameter in question would be compared to a static threshold value, the present invention allows the parameter in question to be compared to a dynamic value deriving from measured data.

It is within the scope of the present invention to measure values of operating parameters, as well as derivatives thereof, such as first or second order derivatives. Hence, if a measured value may is not in itself of interest, but if a derivate thereof may indicate the onset of a fault or breakdown, such onset may be captured and appropriate action may be taken. For example, even a sudden rise in mechanical load of e.g. a shaft of the wind turbine may be acceptable, if it occurs in average on all wind turbines. If, however, the rise occurs at a higher rate at one wind turbine than at other turbines, a critical situation may exist in that one wind turbine.

Further, it is within the scope of the present invention that durations of abnormal parameter values may be determined. For example, the method may be configured not to emit a warning signal, if an excessive value only occurs for a predefined period of time.

The first value may be an average of past values, such as past actual values, of a single first wind turbine or of values of a plurality of wind turbines in a first group of turbines. The second average value may be an average, a median, a distribution or a deviation of values of a single second wind turbine or of values of plurality of wind turbines in a second group of turbines. If desirable, the second group of wind turbines may include the first wind turbine or the plurality of wind turbines of the first group of turbines.

It will be appreciated that, in addition to comparing first wind turbine measurement data to second data, the present method may also include the step of emitting an output signal and/or shutting a wind turbine down, if current measurement data exceed a pre-defined threshold level. The output signal may e.g. comprise an acoustic signal, a visual signal on a computer monitor, or a text message, such as an SMS or e-mail message. In addition, the output signal emitted in case the first value deviates by more than the predetermined value from said second value may involve booking of a technician for example for scheduling service.

The threshold value may be set close to the point, at which the wind turbine or a component thereof is known to fail, or at which the lifetime of wind turbine or the component is severely affected.

Generally, the at least one sensor element of each wind turbine may be arranged to determine at least one of: a temperature of a component of the wind turbine, a temperature of a fluid in the wind turbine, a pressure of a fluid in a component of the wind turbine, a level of a fluid in a component of the wind turbine, a rotational speed of a component of the wind turbine, mechanical stress or load in a component of the wind turbine, a position or a displacement of a component of the wind turbine, a mass or a mass distribution of a component of the wind turbine, a frequency and/or an amplitude of vibrations in a component of the wind turbine, a frequency and/or amplitude of displacement in a component of the wind turbine, a power, voltage or current in a component of the wind turbine, a number of faults of a component of the wind turbine, and a number of predefined events determined by a control system of the wind turbine. In the present context, event includes a number of times a signal exceeds a certain set point, or a number of times a certain combination of events occurs. The determined parameters may e.g. include:

- Blades: temperature, pressure(s), stress, load, vibration, displacement, rotation, tolerances, mass, mass distribution.

- Hub: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution. - Pitch hydraulic system and pitch system controller: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Main shaft: temperature, pressure(s), stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution. - Gearbox: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Brake system: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Turbine system controller: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Generator: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Power factor control system and power factor system controller: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Transformer: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution. - Nacelle ambient conditions controller and nacelle ambient controller: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

- Ambient conditions: temperature, pressure(s), voltage, current, stress, load, vibration (amplitude and magnitude), displacement, rotation, fluid level, tolerances, mass, mass distribution.

The at least one wind turbine is typically included in a wind farm with a plurality of wind turbines. As discussed in detail below, the wind turbines of the wind farm may be divided into groups of wind turbines sharing one or more wind turbine characteristics, whereby average values are determined in each of the groups.

The predetermined deviation value may be a constant or a variable value and may be determined as a standard deviation. It may e.g. be a function of ambient parameters, such as wind velocity, air pressure, temperature, or light intensity, or it may be a function of other operating parameters of the wind turbine, such as stresses, loads, temperatures, pressures etc. of various components. For example, a deviation threshold in respect of oil pressure in a gearbox of the wind turbines may be set at a first, relatively low value, if the ambient wind velocity is relatively low and the gearbox input shaft rotates at a relatively low velocity, and at a second, relatively high value, if the ambient wind velocity is relatively high and the gearbox input shaft rotates at a relatively high velocity. In another example, the predetermined deviation threshold value is a function of the time lapse since last service interval, or a function of the age of a particular component relative to the component's expected life time.

In the present context, the term Output signal' should be understood to embrace any optical, audible, vibratory or any other signal, which occurs in case the difference between an individual parameter value and the average value exceeds the predetermined threshold value. In one embodiment, the output signal is provided by a continuous display on a monitor of the computer system. For example, values of an operating parameter in respect of each individual turbine may be continuously displayed in a chart, which compares the individual values to the average value. If the value of one turbine exceeds the predetermined deviation threshold value, the chart will indicate this, and an update of the chart to this effect may be regarded as emission of the output signal. Additionally, an operator may be notified, e.g. via an acoustic signal, e-mail, voice mail or any other appropriate communications signal. As previously stated, the emitted output signal may involve booking of a technician for example for scheduling service.

The method and system of the present invention may cause automatic action to be taken to remedy a condition causing the deviation of the first value from the second value. In case of a critical deviation, the computer system may cause the wind turbine in question to drill down or to shut down. In other embodiments, the action to be taken may be performed or initiated manually, i.e. by an operator.

The computer system may be located at a wind farm or in the vicinity thereof. For example, a wind farm operator may decide to erect a wind farm control shed in the vicinity of the wind turbine, which may typically house the computer system to carry out the method of the present invention. Alternatively, the computer system may be located at a remote location. If for example, the wind farm is located at an off-shore location, the computer system may be located on shore, there being provided a suitable data transmission system for transmitting the measurement data from the wind farm to the computer system and, possibly, for transmitting control signals back from the computer system to the wind turbines of the wind farm. The data may be transmitted via wireless and/or wired communication facilities, including e.g. the Internet. In one embodiment, a plurality of wind farms may be monitored by the computer system. Hence, a power company operating a plurality of wind farms may monitor the wind farms from a single location. Alternatively, a supplier of wind turbines may monitor a plurality of wind farms at the supplier's own premises.

The method of the present invention may comprises the further the steps of defining at least a first group of wind turbines, so that all wind turbines in the first group share at least one common wind turbine characteristic. In the first group, the first value of the operation parameter may be compared to the second value of the operation parameter, and the output signal may be emitted in case the first value deviates by more than a predetermined value from said second value. The computer system of the wind power production facility of the present invention may likewise be configured to perform the above steps.

Division of the wind turbines into one or more groups may ensure that each individual one of the wind turbines is compared to other wind turbines sharing at least one characteristic of that individual one turbine. The turbine characteristics of the wind turbines may for example include one or more of: a maximal nominal output power of the turbine, a blade length of blades of the turbine, a type identifier of the turbine, the blades and/or any other component of the wind turbine, a height of the wind turbine tower, a type identifier of the generator, gearbox, vibration damper, brake, clutch, transformer, frequency converter or of any other mechanical and/or electrical component of the turbine, an age of the turbine, a time lapse since last service check, and a physical location of the turbine. The characteristics of the wind turbine may also include one or more of the operation parameters.

A second group of wind turbines may be defined, so that all wind turbines in the second group share at least one common wind turbine characteristic. In the second group, a first value of said operation parameter may be compared to a second value of said operation parameter, and the output signal may be emitted in case the first value deviates by more than a predetermined value from said second value.

The groups may be pre-defined and constant, so that changes to the groups may only occur by operator-intervention. Alternatively, the method of the invention may comprise the steps of continuously determining whether each wind turbine of each group has characteristics similar to wind turbines of another one of the groups, and transferring one or more of the wind turbines from a current one of the groups to another one of the groups, if similarities between said one or more wind turbines with said other one of the groups prevail over similarities between said one ore more wind turbines with said current one of the groups. Likewise, the computer system of the invention may be configured to perform the aforementioned steps. Hence, the possibility of dynamic grouping is provided, ensuring that a maximum degree of resemblance exists between wind turbines of individual groups.

Description of the drawings

Embodiments of the invention will now be described with reference to the accompanying drawings, in which:

Fig. 1 illustrates an embodiment of a wind power production facility according to the present invention;

Fig. 2 illustrates a flowchart of an embodiment of a method according to the invention;

Figs. 3-6 illustrate output charts as produced by embodiments of the present invention.

For Figs. 3-6 there is the presentation of the data in time series graph where the Y-axis is assigned the metric of the data being analyzed and X-axis is presented in progression in time. Data set A always represents the average of the group whereas Data set B represents the subject or the individual.

There is then presentation of a probability distribution diagram (commonly referred to a histogram) which presents the same data from time series as distribution. The notations Data set A and Data set B also apply here. The Y-axis is the % probability of occurrence for the given X value represented on the X-axis. The X-axis is assigned the metric of data being analyzed.

Fig. 1 generally depicts an embodiment of a wind power production facility according to the present invention. The facility comprises a wind farm 100, including a plurality of wind turbines 102. Though Fig. 1 only illustrates six wind turbines 102, wind farms typically include more wind turbines, such as 10-100 or 10-50, typically around 20 or 30. Each wind turbine 102 comprises one more sensor elements (not shown) for measuring at least one operation parameter of the turbine. Possible operation parameters which can be measured in embodiments of the present invention are listed in the above summary of the invention. One, several or all of the may be measured. Further, each wind turbine 102 is connected to a computer system 106 via data transmission means, shown by dashed lines in Fig. 1. Data transmission may e.g. occur via a data network, such as a public data network, such as the Internet 104.

As shown in Fig. 2, group average data are collected, and an expectation (normal) is determined from performance of all other similar turbines. The data of the individual turbines are compared to the group data (or group average). As shown in Fig. 2, data may be represented graphically in charts, which, cf. the third chart in the TCM Decision' column of Fig. 2, may indicate an outlier, i.e. differences between average and one ore more individual turbine values beyond a threshold level. In an external process, the turbine may be drilled down, if the turbine is deviant from its specification. An operator may simultaneously be notified, and appropriate site reaction may be undertaken.

The upper chart of Fig. 3 shows a times series graphic presentation of temperature variation throughout 1 month period in a wind turbine parameter (this example is a generator bearing temperature). The grey line (Data set A) represents the average for a group of wind turbines having the same characteristics as the turbine under investigation which is represented by the darker line (series of data points) - Data set B.

The Group average temperature throughout thel month period varies from about 50 to about 70°C. The turbine under investigation peaks around 80°C but varies to as low as 30°C.

The lower chart of Fig. 3 shows a corresponding probability density function graph of the same data for the same time period. The Y axis (right hand axis) represents the % distribution of data measured at the temperature which is represented on the X axis. As shown, the group turbine data (grey bars to the left - Data set A) is below 70°C over 90% of the time measured. The turbine under investigation (bars to the right - Data set B) operated above 70°C over 50% of time measured. The bi-modal distribution (double hump) exhibited by the turbine under investigation also indicates at least 2 distinct operating conditions. Review of the time series clearly indicates the turbine under investigation shifted it's operating temperature range approximately half way through the measured period. Presentation of the time series data with the probability density function make it very easy for an observation to notice significant variation of an individual and identify when the change took place.

Fig. 4 represents distribution of events per a chosen denominator or averaging function. In this example the event analysis is high temperature faults averaged per week.

The Y axis (right hand axis) represents the % distribution of data measured at the given events per week which is represented on the X axis. As shown, the group turbine data (grey bars to the left - Data set A) is below 80 faults per week over 80% of the time observed. The turbine under investigation (bars to the right - Data set B) operated with over 80 faults per week over 80% of time observed.

The upper chart of Fig. 5 shows a times series graphic presentation of a calculated metric of turbulence intensity (it is a function of 10 minutes wind speed and wind variance values) throughout 1 year period. The grey line (Data set A) represents the average for a group of wind turbines similar to the turbine under investigation which is represented by the darker line/data points (series of data points) - Data set B.

The lower chart of Fig. 5 shows a corresponding probability density function graph of the same data for the same time period. The Y axis (right hand axis) represents the % distribution of data calculated at a given turbulence intensity which is represented on the X axis. As shown, the group turbine data (grey bars to the left - Data set A) is turbulence intensity is lower than 16 over 80% of the time measured. The turbine under investigation (bars to the right - Data set B) operated above 16 over 80% of time measured.

Fig. 6 shows the percentage distribution of GWh production. For example, 1% of the turbines have a total production of 7 GWh, and 4% of the turbines have a total production of 9 GWh. Three individual turbines are shown in reference to the others by vertical lines in Fig. 5. The turbine represented by the far right line is close to its wear out limit. The chart of Fig. 5 may represent complete turbines, or it may represent components of turbines. For example, the GWh production sum may be calculated since last inspection of the entire turbine. Alternatively, it may be calculated since last replacement of a component, e.g. gearbox.

Claims

1. A system for monitoring at least one wind turbine, the at least one wind turbine being characterised by a plurality of wind turbine characteristics, the system comprising at least one sensor element for obtaining measurement data of at least one operation parameter of the at least one wind turbine, the system further comprising:

2. A system according to claim 1, wherein said first value is an average of past values, such as past actual values, of a single wind turbine.

3. A system according to claim 1 or 2, wherein said second value is an average, a median, a distribution or a deviation of values of a plurality of wind turbines.

4. A system according to any of the preceding claims, wherein the at least one wind turbine comprises a plurality of wind turbines comprised in a wind farm, and wherein the computer system is located at the wind farm.

5. A system according to any of the preceding claims, wherein the at least one wind turbine comprises a plurality of wind turbines comprised in a wind farm, and wherein the computer system is located at a remote location.

6. A system according to any of the preceding claims, wherein the at least one sensor element of the at least one wind turbine is arranged to determine at least one of: a temperature of a component of the wind turbine, a temperature of a fluid in the wind turbine, a pressure of a fluid in a component of the wind turbine, a level of a fluid in a component of the wind turbine, a rotational speed of a component of the wind turbine, mechanical stress or load in a component of the wind turbine, a position or a displacement of a component of the wind turbine, a mass or a mass distribution of a component of the wind turbine, a frequency and/or an amplitude of vibrations in a component of the wind turbine, a frequency and/or amplitude of displacement in a component of the wind turbine, a power, voltage or current in a component of the wind turbine, a number of faults of a component of the wind turbine, and a number of predefined events determined by a control system of the wind turbine.

7. A system according to any of the preceding claims, wherein the output device includes a monitor configured to display a graphical representation of a distribution of said measurement data.

8. A system according to any of the preceding claims, wherein the computer system is configured to shut down one of the wind turbines, in case the first value deviates by more than a predetermined value from said second value.

9. A system according to any of the preceding claims, wherein the computer system is further configured to:

- define at least a first group of wind turbines, so that all wind turbines in the first group share at least one common wind turbine characteristic;

- in the first group, compare the first value of said operation parameter to the second value of said operation parameter;

- cause the output device to emit the output signal in case the first value deviates by more than a predetermined value from said second value.

10. A system according to claim 9, wherein the first value is obtained from one of the wind turbines of the first group, and wherein the second value is obtained from a plurality of wind turbines of the first group, such as all wind turbines of the first group.

11. A system according to claim 9 or 10, wherein the computer system is further configured to:

- define at least a second group of wind turbines, so that all wind turbines in the second group share at least one common wind turbine characteristic;

- in the second group, compare a first value of said operation parameter to a second value of said operation parameter;

- cause the output device to emit the output signal in case the first value deviates by more than a predetermined value from said second value;

- determine whether each wind turbine of each of the first and second groups has characteristics similar to wind turbines of another one of the groups; - transfer one or more of the wind turbines from a current one of the groups to another one of the groups, if similarities between said one or more wind turbines with said other one of the groups prevail over similarities between said one ore more wind turbines with said current one of the groups.

12. A system according to any of claims 9-11, wherein said wind turbine characteristics include said operation parameter.

13. A computer-implemented method of monitoring at least one operation parameter of at least one wind turbine, the wind turbine being characterised by a plurality of wind turbine characteristics, the method comprising the steps of:

- comparing a first value of said operation parameter to a second value of said operation parameter; - emitting an output signal in case the first value deviates by more than a predetermined value from said second value.

14. A method according to claim 13, wherein said first value is an average of past values, such as past actual values, of a single wind turbine.

15. A method according to claim 13 or 14, wherein said second value is an average, a median, a distribution or a deviation of values of a plurality of wind turbines.

16. A method according to any of claims 13-15, wherein the at least one sensor element of each wind turbine determines at least one of: a temperature of a component of the wind turbine, a temperature of a fluid in the wind turbine, a pressure of a fluid in a component of the wind turbine, a level of a fluid in a component of the wind turbine, a rotational speed of a component of the wind turbine, mechanical stress or load in a component of the wind turbine, a position or a displacement of a component of the wind turbine, a mass or a mass distribution of a component of the wind turbine, a frequency and/or an amplitude of vibrations in a component of the wind turbine, a frequency and/or amplitude of displacement in a component of the wind turbine, a power, voltage or current in a component of the wind turbine, a number of faults of a component of the wind turbine, and a number of predefined events determined by a control system of the wind turbine.

17. A method according to any of claims 13-16, further comprising the step of displaying a graphical representation of a distribution of said measurement data.

18. A method according to any of claims 13-17, further comprising the step of transmitting the measurement data from the at least one wind turbine to a computer system arranged at a remote location.

19. A method according to any of claims 13-18, further comprising the step of shutting down the at least one wind turbine, in case the first value deviates by more than a predetermined value from said second value.

20. A method according to any of claims 13-19, further comprising the steps of:

- defining at least a first group of wind turbines, so that all wind turbines in the first group share at least one common wind turbine characteristic;

- in the first group, comparing the first value of said operation parameter to the second value of said operation parameter; - emitting the output signal in case the first value deviates by more than a predetermined value from said second value.

21. A method according to claim 20, wherein said first value is an average of values of the wind turbines included in the first group of wind turbines.

22. A method according to claim 20 or 21, further comprising the steps of:

- defining a second group of wind turbines, so that all wind turbines in the second group share at least one common wind turbine characteristic;

- in the second group, comparing a first value of said operation parameter to a second value of said operation parameter;

- emitting the output signal in case the first value deviates by more than a predetermined value from said second value;

- determining whether each wind turbine of each of the first and second groups has characteristics similar to wind turbines of another one of the groups; - transferring one or more of the wind turbines from a current one of the groups to another one of the groups, if similarities between said one or more wind turbines with said other one of the groups prevail over similarities between said one ore more wind turbines with said current one of the groups.

23. A method according to any of claims 20-22, wherein said wind turbine characteristics include said operation parameter.