CN109655200B - Method and system for diagnosing unbalance of wind wheel of wind generating set - Google Patents
Method and system for diagnosing unbalance of wind wheel of wind generating set Download PDFInfo
- Publication number
- CN109655200B CN109655200B CN201710946675.9A CN201710946675A CN109655200B CN 109655200 B CN109655200 B CN 109655200B CN 201710946675 A CN201710946675 A CN 201710946675A CN 109655200 B CN109655200 B CN 109655200B
- Authority
- CN
- China
- Prior art keywords
- wind
- wind turbine
- diagnosing
- data
- shaking acceleration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M1/00—Testing static or dynamic balance of machines or structures
- G01M1/14—Determining unbalance
Abstract
The invention discloses a method and a system for diagnosing unbalance of a wind wheel of a wind generating set, wherein the method comprises the following steps: s1, acquiring state data of a plurality of wind generating sets according to a time sequence; the state data comprises wind speed, generating power and engine room shaking acceleration; s2, dispersing the wind speed and the generated power into independent intervals, and calculating data values of the shaking acceleration of the engine room in the intervals; and S3, performing outlier characteristic analysis on the data values of the wind generating sets, and judging the wind generating sets corresponding to the data values with the outlier characteristics as fault sets. The method has the advantages of simpler deployment, more stable diagnosis effect, less possibility of being influenced by short-term abnormal signals during the running of the unit and the like.
Description
Technical Field
The invention relates to the technical field of wind generating set fault diagnosis, in particular to a method and a system for diagnosing unbalance of a wind wheel of a wind generating set.
Background
In the annual operation process of the wind generating set, various faults can occur due to the influence of aerodynamic force and severe environment, and the state monitoring and fault diagnosis of the wind generating set become essential links. Most of blade faults such as damage, ice coating, lightning stroke, blade installation errors and the like are expressed as mass unbalance or pneumatic unbalance of the wind wheel, the unbalanced wind wheel rotation brings higher fatigue load to each component of the wind generating set, premature damage of the structure is caused, and even serious accidents such as tower collapse and the like can be induced under severe conditions. The mass unbalance fault of the wind wheel is caused by the inaccurate processing and manufacturing of the blades, the icing of the blades, the fatigue damage of the blades and the like; the aerodynamic imbalance of the wind wheel is caused by wind shear, airfoil change, blade installation angle error or variable pitch actuating mechanism misoperation and the like.
If the unbalance faults of the wind wheel during the operation of the unit can be accurately detected, various blade faults can be identified in time, and further fault diffusion is prevented to a great extent to avoid causing larger loss.
In recent years, research institutions and companies in the industry successively develop some wind power blade monitoring equipment, mainly adopt modes such as videos, strain gauges and sounds to detect, and wind turbine manufacturers also embed optical fiber sensors in advance when blades are produced so as to obtain running state information of the blades. However, most of the current detection means are expensive and complex to install, and are difficult to popularize and apply. In related research patents and papers, methods of extracting frequency components from a time sequence of operating parameters of a wind generating set and then comparing the frequency components with rotor frequency are almost all used for diagnosing the problem of imbalance of a wind wheel, the difference is only in the difference of selected specific operating parameters and signal processing algorithm details, the methods need to synchronously acquire the operating parameters of the wind generating set at a higher frequency, such as the rotating speed, the power and the like, and need to extract data of specific operating conditions meeting strict screening conditions to carry out analysis and calculation, a complicated pre-model or calculation algorithm is often needed, obvious limiting conditions exist during actual deployment, and the diagnosis effect of the methods also often causes false alarm or false alarm due to over-sensitivity to some abnormal conditions or failure to comprehensively consider some special conditions.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides a method and a system for diagnosing the unbalance of the wind wheel of the wind generating set, which are simpler to deploy, more stable in diagnosis effect and not easily influenced by short-term abnormal signals during the running of the set.
In order to solve the technical problems, the technical scheme provided by the invention is as follows: a method for diagnosing imbalance of a wind wheel of a wind generating set comprises the following steps:
s1, acquiring state data of the wind generating sets according to the time sequence; the state data comprises wind speed, generating power and engine room shaking acceleration;
s2, discretizing the wind speed and the generated power into independent intervals, and calculating the data value of the nacelle shaking acceleration in the intervals;
and S3, performing outlier characteristic analysis on the data values of the wind generating sets, and judging the wind generating sets corresponding to the data values with the outlier characteristics as fault units.
Further, the nacelle shaking acceleration comprises a nacelle shaking acceleration in the front-back direction and a nacelle shaking acceleration in the left-right direction.
Further, the time-series status data is preferably data of more than one day; further preferred are data of more than one day and less than three months.
Further, the state data in time series may be data in a preset time period in all the state data; the state data according to the time sequence can be data under a preset working condition in all the state data; and the state data according to the time sequence is data which meets the sampling requirement of a preset data sampling rate.
Further, the preset operating condition is an operating condition determined by at least one predetermined operating condition parameter, the predetermined operating condition parameter comprising: the state of a main controller, the power generation power, the rotating speed of a wind wheel and the rotating speed of a generator; the preset data sampling rate is preferably greater than 0.1 hertz.
Further, in step S1, smoothing the data of wind speed and generated power to eliminate high frequency fluctuation inherent in the operation of the wind turbine generator system; the smoothing processing method comprises a smoothing algorithm of sliding window average; preferably, the smoothing processing method is a time scale average aggregation processing algorithm.
Further, in step S1, performing time domain transformation on the nacelle shake acceleration; preferably, the time domain transform method includes: downscaling calculates the effective value or standard deviation per minute of cabin sway acceleration at second-level granularity.
Further, the wind speed and the generated power after the smoothing processing in step S1 are aligned with the time stamp of the nacelle sway acceleration after the time domain transformation.
Further, the discretization in the step S2 adopts an equidistant binning algorithm; the data value of the nacelle sway acceleration in the interval is preferably an average value within the interval.
Further, the specific step of step S3 includes:
s3.1, drawing the wind speed, the generating power and the engine room shaking acceleration data of each wind generating set into a curve chart;
and S3.2, analyzing the outlier characteristics of the corresponding curves of the wind generating sets in the curve chart, and determining whether the wind generating sets have the pneumatic unbalance faults or the mass unbalance faults according to the outlier characteristics and the variation trend of the curves.
Further, the graph in step S3.1 includes: a wind speed interval-cabin front and rear direction shaking acceleration curve chart, a wind speed interval-cabin left and right direction shaking acceleration curve chart, a power generation interval-cabin front and rear direction shaking acceleration curve chart, and a power generation interval-cabin left and right direction shaking acceleration curve chart.
Further, the specific steps of step S3.2 include: and analyzing the outlier characteristics of the curve graphs, and judging that the wind generating set has an unbalanced fault when the curve of the wind generating set in the curve graphs has the outlier characteristics.
Further, the specific steps of step S3.2 include: analyzing the outlier characteristics of the graphs, and judging that the wind generating set has a pneumatic unbalance fault when the curves of the wind generating set in the wind speed interval-cabin front-rear direction shaking acceleration graph and/or the generating power interval-cabin front-rear direction shaking acceleration graph have the outlier characteristics and the cabin front-rear direction shaking acceleration shows a continuously increasing trend along with the increase of the wind speed and/or the generating power; and when the curves of the wind generating set in the wind speed interval-cabin left and right direction shaking acceleration curve and/or the generating power interval-cabin left and right direction shaking acceleration curve have the outlier characteristic, and after the wind speed and/or the generating power are increased to a certain value, the cabin left and right direction shaking acceleration shows a relatively stable trend, and the wind generating set is judged to have the mass unbalance fault. Further, the analysis of the outlier features comprises: and adopting a fitting algorithm for each curve, describing each curve by using the characteristic parameters of the fitting model, carrying out cluster analysis on the fitted curves, and judging whether the curves which do not belong to the cluster set in the cluster analysis result have the outlier characteristic.
Further, the fitting algorithm is preferably a linear fitting algorithm or a polynomial fitting algorithm; the algorithm of the cluster analysis is preferably a k-Mean algorithm or a SOM algorithm.
A system for diagnosing wind turbine imbalance of a wind turbine generator system, comprising a diagnostic computer loaded with a program for carrying out the method according to any one of the preceding claims.
Compared with the prior art, the invention has the advantages that:
1. the time domain analysis method of the time sequence data is used for fault diagnosis instead of the frequency analysis method, the method is simpler to deploy, the diagnosis effect is more stable, and the method is not easily influenced by a small amount of abnormal signals in a short time;
2. compared with a frequency analysis method, the method disclosed by the invention has the advantages that the requirement on the required input operation data is lower, the specific operation working condition is not required, and the sampling rate higher than 1Hz is not required, so that the deployment difficulty is reduced, the requirements can be met by the common configuration of the SCADA of the existing wind generating set, and the additional investment is not required.
3. The algorithm (such as downscaling average aggregation, effective value, standard deviation, equidistant binning, linear fitting, k-means clustering and the like) used by the method is simple and easy to implement, has small calculated amount, and is favorable for deployment in production environments with limited computing resources, such as PLC (programmable logic controller) and the like.
4. The invention constructs the quantitative evaluation index of the effective value (or standard deviation) of the shaking acceleration of the engine room corresponding to the interval of the wind speed (or the generated power), fully and obviously quantifies and displays the unbalance problem of the wind wheel, and has obvious identification and diagnosis effects and very high interpretability.
5. The quantitative index of the invention can be combined with machine learning algorithms such as k-means or self-organizing mapping and the like to realize the online monitoring of the fault.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention.
FIG. 2 is raw time series data at 1 second granularity for a specific embodiment of the present invention.
FIG. 3 is data of a time series aggregated to a 1 minute particle size for a specific embodiment of the present invention.
Fig. 4 is a diagram illustrating discretization of generated power according to an embodiment of the present invention.
FIG. 5 is a diagram illustrating a method for representing the nacelle sloshing level for each power stage in an embodiment of the present invention.
FIG. 6 is a comparison chart of the nacelle sway at each power level for 23 units in an embodiment of the present invention.
FIG. 7 is a linear fit chart of 23 modules in an embodiment of the present invention.
Fig. 8 is a result of k-means clustering separating an abnormal unit in the embodiment of the present invention.
Fig. 9 is a diagram of a data analysis of 25 sets of machines in accordance with an embodiment of the present invention.
Detailed Description
The invention is further described below with reference to the drawings and specific preferred embodiments of the description, without thereby limiting the scope of protection of the invention.
The first embodiment is as follows: as shown in fig. 1, the method for diagnosing imbalance of a wind wheel of a wind turbine generator system of the present embodiment includes: s1, acquiring state data of the wind generating sets according to the time sequence; the state data comprises wind speed, generating power and engine room shaking acceleration; s2, discretizing the wind speed and the generated power into independent intervals, and calculating the data value of the nacelle shaking acceleration in the intervals; and S3, performing outlier characteristic analysis on the data values of the wind generating sets, and judging the wind generating sets corresponding to the data values with the outlier characteristics as fault units. The time-series status data of the plurality of wind turbine generators in this embodiment is defined as: the state data of different wind generating sets at the same or different time according to the time sequence, or the state data of the same wind generating set obtained for multiple times at different time according to the time sequence. In this embodiment, the wind wheel rotation speed and other parameters that can be directly or indirectly converted into the wind speed can be used as the replacement variables of the wind speed parameters, the generator rotation speed and other parameters that can be directly or indirectly converted into the generator power can be used as the replacement variables of the generator power parameters, and all should be regarded as the equivalent technical features of the present invention, and fall into the protection scope of the present invention.
In this embodiment, the nacelle sway acceleration includes a nacelle front-rear direction sway acceleration, and a nacelle left-right direction sway acceleration. The time-series status data is preferably data of more than one day; further preferred are data of more than one day and less than three months. The increased amount of data may improve the accuracy of the diagnosis, but the increased amount of data may also increase the complexity of the calculation.
In this embodiment, the time-series status data may be data having a preset time period in all status data; the state data according to the time sequence can be data under a preset working condition in all the state data; and the state data according to the time sequence is data which meets the sampling requirement of a preset data sampling rate.
In this embodiment, the preset operating condition is an operating condition determined by at least one predetermined operating condition parameter, and the predetermined operating condition parameter includes: the state of a main controller, the power generation power, the rotating speed of a wind wheel and the rotating speed of a generator; the preset data sampling rate is preferably greater than 0.1 hertz.
In this embodiment, in step S1, the method further includes smoothing the data of the wind speed and the generated power to eliminate high-frequency fluctuations inherent in the operation of the wind turbine generator system; the smoothing processing method comprises a smoothing algorithm of sliding window average; preferably, the smoothing processing method is a time scale average aggregation processing algorithm. Step S1, further includes performing time domain transformation on the nacelle shaking acceleration; preferably, the time domain transform method includes: downscaling calculates the effective value or standard deviation per minute of cabin sway acceleration at second-level granularity. And aligning the wind speed and the generated power after the smoothing processing in the step S1 with the time stamp of the nacelle shaking acceleration after the time domain transformation.
In this embodiment, the time-series status data of the plurality of wind turbine generators is acquired at a granularity of 1 second, the originally acquired data is shown in fig. 2, and the time-series status data aggregated to a granularity of 1 minute through smoothing processing is shown in fig. 3. It can be seen from the time sequence diagram that the wind input into the whole system changes constantly, and the wind generating set responds to and controls and adjusts the structure of the wind generating set, so that the running state of the wind generating set is time-varying, and a large section of stable working condition is not easy to separate from the time dimension. The abscissa of fig. 2 is a time stamp, and 5 curves are, from top to bottom, the wind speed, the power, the rotational speed of the wind wheel, the forward and backward shaking acceleration of the nacelle, and the left and right shaking acceleration of the nacelle in sequence. The abscissa of fig. 3 is a time stamp, and 5 curves are, from top to bottom, the smoothed wind speed, power, wind wheel rotational speed, nacelle forward and backward shaking acceleration, and nacelle left and right shaking acceleration in sequence.
In this embodiment, the discretization in step S2 adopts an equidistant binning algorithm; the data value of the nacelle sway acceleration in the interval is preferably an average value within the interval. In order to simplify the data structure, the continuous valued variables such as wind speed and generated power are discretized. In this embodiment, the power generation power is discretized by dividing the power generation power into one box by 50kW, as shown in fig. 4. The generated power (ActPowBin) and the effective value of the nacelle forward-backward shaking acceleration (nadrimms) are plotted on the horizontal axis and the vertical axis, respectively, as shown in fig. 5. The upper half of fig. 5 is plotted as a boxed graph with nadrimms in each bin, the middle point of the worker-word line in each bin in the lower half represents the average value of nadrimms in the bin, and the upper and lower vertices of the h-line represent the positive and negative one-fold standard deviation of nadrimms in the bin, respectively. As can be seen from the figure, the average value of NADriRMS within each power bin can be used to reflect the level of nacelle sloshing at that power level.
In this embodiment, the specific step of step S3 includes: s3.1, drawing the wind speed, the generating power and the engine room shaking acceleration data of each wind generating set into a curve chart; and S3.2, analyzing the outlier characteristics of the corresponding curves of the wind generating sets in the curve chart, and determining whether the wind generating sets have the pneumatic unbalance faults or the mass unbalance faults according to the outlier characteristics and the variation trend of the curves. The graph in step S3.1 includes: a wind speed interval-cabin front and rear direction shaking acceleration curve chart, a wind speed interval-cabin left and right direction shaking acceleration curve chart, a power generation interval-cabin front and rear direction shaking acceleration curve chart, and a power generation interval-cabin left and right direction shaking acceleration curve chart. The specific steps of step S3.2 include: and analyzing the outlier characteristics of the curve graphs, and judging that the wind generating set has an unbalanced fault when the curve of the wind generating set in the curve graphs has the outlier characteristics. Analysis of the outlier features comprising: and adopting a fitting algorithm for each curve, describing each curve by using the characteristic parameters of the fitting model, carrying out cluster analysis on the fitted curves, and judging whether the curves which do not belong to the cluster set in the cluster analysis result have the outlier characteristic. The fitting algorithm is preferably a linear fitting algorithm or a polynomial fitting algorithm; the algorithm of the cluster analysis is preferably a k-Mean algorithm, or a SOM (self-organizing map) algorithm.
In the present embodiment, an average value of the effective value of the nacelle front-rear direction rocking acceleration (nadrinms) and the effective value of the nacelle left-right direction rocking acceleration (NANonRMS) in the power bins of 23 wind turbine generators (from a to W) is plotted as shown in fig. 6. It can be seen from the figure that the shaking acceleration of the A-type and B-type wind generating sets in the front and back directions of the engine room in all power intervals is obviously higher than that of other sets, the shaking acceleration of the left and right directions of the engine room is also obviously higher than that of other sets in a plurality of power intervals, and the shaking acceleration of the front and back directions of the engine room is continuously increased along with the increase of the power, so that the wind turbine pneumatic imbalance fault of the A-type and B-type wind generating sets can be judged. Through field inspection, the fact that the installation angle of each of the A unit and the B unit is about 5 degrees, which causes obvious pneumatic imbalance of the wind wheel, is found, and the unit obviously shakes greatly during operation. However, the vibration of the magnitude does not reach the safety threshold value of the vibration protection of the wind generating set, so that the fault alarm is not triggered, and the wind generating set belongs to a more serious hidden fault.
In this embodiment, the result of linear fitting of the average data of effective values (nadrimms) of forward and backward shaking acceleration in the cabin in each sub-box of each power generation of 23 wind turbine generators is shown in fig. 7, and the slope and intercept values of each wind turbine generator set are obtained, and it can be seen that the slope values of the a-type and B-type wind turbine generators show obvious outlier characteristics. In this embodiment, clustering analysis is performed through a k-means algorithm (the number of given classes is 2), and the clustering result is shown in fig. 8, which shows that the unit a and the unit B do not belong to the clustering set of the majority of normal units, and have an obvious outlier characteristic.
Example two: the method is basically the same as the first embodiment, except that in step S3.2, by analyzing the outlier characteristics of each graph, when the curves of the wind generating set in the wind speed interval-nacelle front-rear direction shaking acceleration graph and/or the power generation interval-nacelle front-rear direction shaking acceleration graph have the outlier characteristics, and the nacelle front-rear direction shaking acceleration shows a continuously increasing trend along with the increase of the wind speed and/or the power generation power, it is determined that the wind generating set has the aerodynamic imbalance fault; and when the curves of the wind generating set in the wind speed interval-cabin left and right direction shaking acceleration curve and/or the generating power interval-cabin left and right direction shaking acceleration curve have the outlier characteristic, and after the wind speed and/or the generating power are increased to a certain value, the cabin left and right direction shaking acceleration shows a relatively stable trend, and the wind generating set is judged to have the mass unbalance fault. The embodiment is more detailed for the fault analysis of embodiment one, not only can judge whether wind generating set has unbalanced fault, and further divide unbalanced fault into pneumatic unbalanced fault and the unbalanced mass trouble, and guidance nature is stronger.
In this embodiment, data of 25 wind turbine generators are analyzed, standard deviations of generated power, rotational speed, and forward and backward shaking acceleration of the nacelle in the wind speed sub-box of the 25 wind turbine generators are sequentially shown in three sub-diagrams from left to right in fig. 9, average data of standard deviations of generated power, rotational speed, and left and right shaking acceleration of the nacelle in the wind speed sub-box of the 25 wind turbine generators is shown in three sub-diagrams from left to right in fig. 9, and for better comparison, the six sub-diagrams are displayed together. It can be seen from the figure that the left and right shaking acceleration of the nacelle of one unit is significantly higher than that of other units, and after the generated power (and the wind speed) reaches a certain value, the left and right shaking acceleration of the nacelle shows a relatively stable trend, so that the unit can be judged to have the fault of unbalanced wind wheel mass. And the field inspection verifies that the fault of the wind generating set really exists.
The wind turbine imbalance diagnosis system of the wind turbine generator system of the embodiment comprises a diagnosis computer loaded with a program capable of executing the diagnosis method.
The foregoing is considered as illustrative of the preferred embodiments of the invention and is not to be construed as limiting the invention in any way. Although the present invention has been described with reference to the preferred embodiments, it is not intended to be limited thereto. Therefore, any simple modification, equivalent change and modification made to the above embodiments according to the technical spirit of the present invention should fall within the protection scope of the technical scheme of the present invention, unless the technical spirit of the present invention departs from the content of the technical scheme of the present invention.
Claims (16)
1. A method for diagnosing imbalance of a wind wheel of a wind generating set is characterized by comprising the following steps:
s1, acquiring state data of a plurality of wind generating sets according to a time sequence; the state data comprises wind speed, generating power and engine room shaking acceleration;
s2, dispersing the wind speed and the generated power into independent intervals, and calculating data values of the shaking acceleration of the engine room in the intervals;
and S3, performing outlier characteristic analysis on the data values of the wind generating sets, and judging the wind generating sets corresponding to the data values with the outlier characteristics as fault sets.
2. The method of diagnosing wind turbine imbalance of a wind turbine generator set according to claim 1, wherein: the cabin shaking acceleration comprises a front and rear cabin shaking acceleration and a left and right cabin shaking acceleration.
3. The method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 2, wherein: the time-series status data is data of more than one day.
4. A method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 3, characterised in that: the state data according to the time sequence is data with a preset time period in all the state data;
or:
the state data according to the time sequence is data under a preset working condition in all the state data;
or: and the state data according to the time sequence is data which meets the sampling requirement of a preset data sampling rate.
5. The method of diagnosing wind turbine imbalance of a wind turbine generator set according to claim 4, wherein: the predetermined operating condition is an operating condition determined by at least one predetermined operating condition parameter, the predetermined operating condition parameter comprising: the state of a main controller, the power generation power, the rotating speed of a wind wheel and the rotating speed of a generator; the preset data sampling rate is greater than 0.1 hertz.
6. The method of diagnosing wind turbine imbalance of a wind turbine generator set according to claim 5, wherein: in step S1, smoothing the data of wind speed and generated power to eliminate high frequency fluctuation inherent in the operation of the wind turbine generator system; the smoothing processing method comprises a smoothing algorithm of sliding window average; or: the smoothing processing method is a time scale reduction average aggregation processing algorithm.
7. The method of diagnosing wind turbine imbalance of a wind turbine generator set according to claim 6, wherein: step S1, further includes performing time domain transformation on the nacelle shaking acceleration; the time domain transformation method comprises the following steps: downscaling calculates the effective value or standard deviation per minute of cabin sway acceleration at second-level granularity.
8. The method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 7, wherein: and aligning the wind speed and the generated power after the smoothing processing in the step S1 with the time stamp of the nacelle shaking acceleration after the time domain transformation.
9. The method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 8, wherein: in the step S2, the discretization adopts an equidistant binning algorithm; and the data value of the cabin shaking acceleration in the interval is the average value in the interval.
10. Method for diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 9, characterized in that: the specific steps of step S3 include:
s3.1, drawing the wind speed, the generating power and the engine room shaking acceleration data of each wind generating set into a curve chart;
and S3.2, analyzing the outlier characteristics of the curves corresponding to the wind generating sets in the curve chart, and determining whether the wind generating sets have the pneumatic unbalance faults or the mass unbalance faults according to the outlier characteristics and the variation trend of the curves.
11. Method for diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 10, characterized in that: the graph in step S3.1 includes: a wind speed interval-cabin front and rear direction shaking acceleration curve chart, a wind speed interval-cabin left and right direction shaking acceleration curve chart, a power generation interval-cabin front and rear direction shaking acceleration curve chart, and a power generation interval-cabin left and right direction shaking acceleration curve chart.
12. The method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 11, wherein: the specific steps of step S3.2 include: and analyzing the outlier characteristics of the curve graphs, and judging that the wind generating set has an unbalanced fault when the curve of the wind generating set in the curve graphs has the outlier characteristics.
13. The method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 11, wherein: the specific steps of step S3.2 include: analyzing the outlier characteristics of the graphs, and judging that the wind generating set has a pneumatic unbalance fault when the curves of the wind generating set in the wind speed interval-cabin front-rear direction shaking acceleration graph and/or the generating power interval-cabin front-rear direction shaking acceleration graph have the outlier characteristics and the cabin front-rear direction shaking acceleration shows a continuously increasing trend along with the increase of the wind speed and/or the generating power; and when the curves of the wind generating set in the wind speed interval-cabin left and right direction shaking acceleration curve and/or the generating power interval-cabin left and right direction shaking acceleration curve have the outlier characteristic, and after the wind speed and/or the generating power are increased to a certain value, the cabin left and right direction shaking acceleration shows a relatively stable trend, and the wind generating set is judged to have the mass unbalance fault.
14. Method for diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 12 or 13, characterized in that: analysis of the outlier features comprising: and adopting a fitting algorithm for each curve, describing each curve by using the characteristic parameters of the fitting model, carrying out cluster analysis on the fitted curves, and judging whether the curves which do not belong to the cluster set in the cluster analysis result have the outlier characteristic.
15. The method of diagnosing a wind turbine imbalance of a wind turbine generator set according to claim 14, wherein: the fitting algorithm is a linear fitting algorithm or a polynomial fitting algorithm; the algorithm of the cluster analysis is a k-Mean algorithm or an SOM algorithm.
16. A diagnosis system for unbalance of a wind wheel of a wind generating set is characterized in that: comprising a diagnostic computer loaded with a program for performing the diagnostic method according to any one of claims 1 to 15.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710946675.9A CN109655200B (en) | 2017-10-12 | 2017-10-12 | Method and system for diagnosing unbalance of wind wheel of wind generating set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710946675.9A CN109655200B (en) | 2017-10-12 | 2017-10-12 | Method and system for diagnosing unbalance of wind wheel of wind generating set |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109655200A CN109655200A (en) | 2019-04-19 |
| CN109655200B true CN109655200B (en) | 2021-01-29 |
Family
ID=66109192
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710946675.9A Active CN109655200B (en) | 2017-10-12 | 2017-10-12 | Method and system for diagnosing unbalance of wind wheel of wind generating set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109655200B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112014593B (en) * | 2019-05-28 | 2022-05-17 | 浙江德盛铁路器材股份有限公司 | Device and method for monitoring and evaluating quality condition of railway track basic equipment |
| CN112555101B (en) * | 2019-09-26 | 2023-03-31 | 北京金风科创风电设备有限公司 | Method and device for identifying impeller aerodynamic state of wind generating set |
| CN112983750B (en) * | 2019-12-13 | 2022-07-19 | 中车株洲电力机车研究所有限公司 | Method and device for diagnosing mounting dislocation of blades of wind turbine generator |
| CN111289179B (en) * | 2020-03-04 | 2022-06-03 | 中国船舶重工集团海装风电股份有限公司 | Method for detecting unbalanced fusion of impellers of wind generating set |
| CN112083347B (en) * | 2020-08-20 | 2023-06-30 | 上海电享信息科技有限公司 | Screening method for power batteries of electric vehicles |
| CN112267979B (en) * | 2020-10-26 | 2021-07-23 | 积成电子股份有限公司 | Early warning method and system for judging failure of yaw bearing |
| CN112834224B (en) * | 2021-01-05 | 2023-05-23 | 广东核电合营有限公司 | Nuclear power steam turbine generator health state assessment method and system |
| CN113297291A (en) * | 2021-05-08 | 2021-08-24 | 上海电气风电集团股份有限公司 | Monitoring method, monitoring system, readable storage medium and wind driven generator |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100959543B1 (en) * | 2008-08-10 | 2010-05-27 | 주식회사 서부에너지기술 | Method and system for safely managing structure |
| CN101871844A (en) * | 2010-06-13 | 2010-10-27 | 清华大学 | Performance analysis and fault simulation experiment system of wind machine |
| CN103234702A (en) * | 2013-04-11 | 2013-08-07 | 东南大学 | Method for diagnosing imbalance faults of blades |
| CN105569932A (en) * | 2016-01-08 | 2016-05-11 | 新疆金风科技股份有限公司 | Dynamic unbalance online testing and fault identification method and system for wind turbine generators |
| CN107191339A (en) * | 2017-07-31 | 2017-09-22 | 上海电气风电集团有限公司 | Wind-driven generator group wind-wheel imbalance monitoring method |
-
2017
- 2017-10-12 CN CN201710946675.9A patent/CN109655200B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100959543B1 (en) * | 2008-08-10 | 2010-05-27 | 주식회사 서부에너지기술 | Method and system for safely managing structure |
| CN101871844A (en) * | 2010-06-13 | 2010-10-27 | 清华大学 | Performance analysis and fault simulation experiment system of wind machine |
| CN103234702A (en) * | 2013-04-11 | 2013-08-07 | 东南大学 | Method for diagnosing imbalance faults of blades |
| CN105569932A (en) * | 2016-01-08 | 2016-05-11 | 新疆金风科技股份有限公司 | Dynamic unbalance online testing and fault identification method and system for wind turbine generators |
| CN107191339A (en) * | 2017-07-31 | 2017-09-22 | 上海电气风电集团有限公司 | Wind-driven generator group wind-wheel imbalance monitoring method |
Non-Patent Citations (2)
| Title |
|---|
| 大型风力机组远程智能监测与诊断系统的研究与开发;杨文广 等;《中国工程科学》;20151231;第17卷(第3期);第24-29页 * |
| 风电机组气动不平衡分析及诊断;王千 等;《风能》;20141231(第10期);第108-110页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109655200A (en) | 2019-04-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109655200B (en) | Method and system for diagnosing unbalance of wind wheel of wind generating set | |
| EP3221579B1 (en) | Wind turbine condition monitoring method and system | |
| US10725439B2 (en) | Apparatus and method for monitoring a device having a movable part | |
| CN102870057B (en) | Plant diagnosis device, diagnosis method, and diagnosis program | |
| EP3478960B1 (en) | Diagnostic system, wind turbine system, method for use in a wind turbine and computer program product | |
| US20130060524A1 (en) | Machine Anomaly Detection and Diagnosis Incorporating Operational Data | |
| WO2016086360A1 (en) | Wind farm condition monitoring method and system | |
| US11761427B2 (en) | Method and system for building prescriptive analytics to prevent wind turbine failures | |
| CN107247230A (en) | A kind of electric rotating machine state monitoring method based on SVMs and data-driven | |
| JP6849446B2 (en) | Vibration monitoring system | |
| CN110905732B (en) | Method and system for identifying unbalance of wind wheel of wind turbine generator and storage medium | |
| CN103674234A (en) | State early warning method and system for abnormal vibration of wind generating set | |
| CN103925155A (en) | Self-adaptive detection method for abnormal wind turbine output power | |
| CN109840666A (en) | A kind of model building method and system for predicting that the following Wind turbines break down | |
| CN113383215A (en) | System and process for mode-matched bearing vibration diagnostics | |
| Koukoura et al. | Wind turbine intelligent gear fault identification | |
| EP4045791B1 (en) | Method and an apparatus for computer-implemented monitoring of a wind turbine | |
| WO2022064038A1 (en) | Method and system for wind speed determination using vibration data | |
| EP3104152B1 (en) | Method and controller for determining an undesired condition in an electrical drive system | |
| JP2021076597A (en) | Detecting rotor anomalies by determining vibration trends during transient speed operation | |
| An et al. | Fault diagnosis of direct-drive wind turbine based on support vector machine | |
| TWI762101B (en) | Abnormal Diagnosis Device and Abnormal Diagnosis Program | |
| CN110836786B (en) | Mechanical fault monitoring method, device, system, medium and computing equipment | |
| CN117128143A (en) | Blade fault detection method and related device | |
| CN117028164A (en) | Vibration detection method and equipment for wind turbine generator |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |