CN114065842A - Wind turbine generator fault self-diagnosis method and system - Google Patents
Wind turbine generator fault self-diagnosis method and system Download PDFInfo
- Publication number
- CN114065842A CN114065842A CN202111248577.0A CN202111248577A CN114065842A CN 114065842 A CN114065842 A CN 114065842A CN 202111248577 A CN202111248577 A CN 202111248577A CN 114065842 A CN114065842 A CN 114065842A
- Authority
- CN
- China
- Prior art keywords
- wind turbine
- data
- model
- sample
- parameter
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Wind Motors (AREA)
Abstract
The disclosure provides a wind turbine generator fault self-diagnosis method and system. The method comprises the following steps: acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises actual measured values of at least one parameter to be predicted; determining a first wind turbine group where a wind turbine is located; determining a first working condition type of the wind turbine generator; according to the first wind turbine group and the first working condition type, determining a trained fault diagnosis model corresponding to the first working condition type in the first wind turbine group; inputting the operation parameter data into the trained fault diagnosis model to obtain a model prediction value of the parameter to be predicted; and generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted. Therefore, the method for self-diagnosing the faults of the wind turbine generator under all working conditions is provided, and the corresponding diagnosis strategy is provided according to the model prediction result to realize the self-diagnosis function.
Description
Technical Field
The disclosure relates to the technical field of energy, in particular to a wind turbine generator fault self-diagnosis method and system for a wind turbine generator gearbox.
Background
At present, with the aggravation of the problem of energy shortage, wind energy is taken as a very important clean energy and will play an irreplaceable role in the future low-carbon times. Wind power generation has the advantages of being renewable, environment-friendly and the like, and is widely applied, and a wind turbine generator is an important part of wind power generation, can convert wind energy into alternating current electric energy, and is large-scale rotating equipment which operates under variable working conditions. With the continuous rising of installed capacity, how to reduce the operation and maintenance cost of the wind turbine generator and increase the generation benefit are more and more concerned.
In the operation and maintenance cost of the wind turbine generator, huge operation and maintenance cost is consumed due to the complete machine safety and the failure of key parts. How to quickly and effectively diagnose the fault of the wind turbine generator is a technical problem which needs to be solved urgently.
Disclosure of Invention
The present disclosure is directed to solving, at least to some extent, one of the technical problems in the above-described technology.
Therefore, a first object of the present disclosure is to provide a wind turbine generator fault self-diagnosis method, which can realize wind turbine generator fault self-diagnosis under all operating conditions.
A second objective of the present disclosure is to provide a wind turbine generator fault self-diagnosis system.
A third object of the present disclosure is to provide a wind turbine.
A fourth object of the present disclosure is to provide an electronic device.
A fifth object of the present disclosure is to propose a computer-readable storage medium.
An embodiment of the first aspect of the disclosure provides a wind turbine generator fault self-diagnosis method, which includes: acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises true measurements of at least one parameter to be predicted; determining a first wind turbine group where the wind turbines are located according to the state parameter data and the operation parameter data; determining a first working condition type of the wind turbine generator according to the operation parameter data and the first wind turbine generator group; according to the first wind turbine group and the first working condition category, determining a trained fault diagnosis model corresponding to the first working condition category in the first wind turbine group; inputting the operation parameter data into a trained fault diagnosis model to obtain a model prediction value of the parameter to be predicted; and generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted.
In some embodiments, the determining, according to the state parameter data and the operation parameter data, the first wind turbine group in which the wind turbine is located includes: acquiring state parameter sample data of a plurality of sample wind turbine generators and operating parameter sample data within a second preset time length; according to the state parameter sample data and the operation parameter sample data, performing cluster analysis on the plurality of sample wind turbine generators, and dividing the cluster analysis into a plurality of sample wind turbine generator groups; and determining a first wind turbine group where the wind turbine is located by comparing the state parameter data and the operation parameter data with the state parameter sample data and the operation parameter sample data of different wind turbine groups.
In some embodiments, the performing, according to the state parameter sample data and the operation parameter sample data, cluster analysis on a plurality of sample wind turbines to divide the sample wind turbines into a plurality of sample wind turbine groups includes: performing characteristic quantization processing on the state parameter sample data and the operation parameter sample data to generate quantized state parameter sample data and quantized operation parameter sample data; performing cluster analysis based on a cluster analysis algorithm according to the quantized state parameter sample data and the quantized operation parameter sample data to obtain a classification result; and dividing the plurality of sample wind generation sets into a plurality of sample wind generation sets according to the classification result.
In some embodiments, the determining a first operating condition category of the wind turbine generator according to the operating parameter data and the first wind turbine group includes: dividing the operating parameter sample data of at least one sample wind turbine generator in the sample wind turbine generator group into sample working condition data under different working conditions; and determining the first working condition type of the wind turbine generator by comparing the operation parameter data with sample working condition data under different working conditions in the first wind turbine generator group.
In some embodiments, the dividing the operating parameter sample data of at least one of the sample wind turbines in the sample wind turbine group into sample operating condition data under different operating conditions includes: extracting working condition data from the operating parameter sample data; dividing each working condition data by adopting a K-fold equal probability method to obtain a plurality of different working condition groups; wherein K is an integer greater than 1; and dividing the operation parameter sample data according to the working condition groups to obtain sample working condition data under different working conditions.
In some embodiments, the determining, according to the first wind turbine group and the first operating condition category, a trained first fault diagnosis model corresponding to the first operating condition category in the first wind turbine group includes: determining at least one fault diagnosis model applicable to each working condition according to sample working condition data under different working conditions; inputting the sample data under each working condition to at least one fault diagnosis model applicable to the working condition, and training the fault diagnosis model to obtain a trained fault diagnosis model; and determining at least one trained fault diagnosis model corresponding to the first working condition type in the first wind turbine group according to the first wind turbine group and the first working condition type.
In some embodiments, the fault diagnosis model is a random forest model and/or an extreme gradient boosting tree Xgboost model and/or a neural network model.
In some embodiments, the method further comprises: judging whether to trigger model updating or not according to a first state parameter and a first preset model updating parameter in the state parameter data; and under the condition that the first state parameter is the first preset model updating parameter, triggering model updating, and updating the trained first fault model.
In some embodiments, the first state parameter comprises at least one of: unit maintenance information; unit fault information; regular maintenance records; key component maintenance information; key component replacement information; adding lubrication to key parts; the model is used for time.
In some embodiments, the generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted includes: calculating a difference value between the model predicted value and the real measured value, and comparing the difference value with a residual threshold value or a change rate or a change quantity to generate a fault diagnosis result; or generating a fault diagnosis result by comparing the probability distribution of the model predicted value and the real measured value.
An embodiment of a second aspect of the present disclosure provides a wind turbine generator system fault self-diagnosis system, including: the data acquisition module is used for acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises actual measurements of the parameter to be predicted; the first determining module is used for determining a first wind turbine group where the wind turbine is located according to the state parameter data and the operation parameter data; the second determining module is used for determining the first working condition category of the wind turbine generator according to the operation parameter data and the first wind turbine generator group; the third determining module is used for determining a trained first fault diagnosis model corresponding to the first working condition type in the first wind turbine group according to the first wind turbine group and the first working condition type; the model prediction module is used for inputting the state parameter data into the trained first fault diagnosis model to obtain a model prediction value of the parameter to be predicted; and the diagnosis processing module is used for generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted.
An embodiment of a third aspect of the present disclosure provides a wind turbine generator, including: the wind turbine generator fault self-diagnosis system is disclosed as an embodiment of the second aspect of the disclosure.
An embodiment of a fourth aspect of the present disclosure provides an electronic device, including: the self-diagnosis method for the fault of the wind turbine generator set comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein when the processor executes the program, the self-diagnosis method for the fault of the wind turbine generator set is realized.
An embodiment of a fifth aspect of the present disclosure provides a computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for self-diagnosing a fault of a wind turbine generator according to the embodiment of the first aspect of the present disclosure.
Additional aspects and advantages of the disclosure will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the disclosure.
Drawings
The foregoing and/or additional aspects and advantages of the present disclosure will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
fig. 1 is a flowchart of a method for self-diagnosing a fault of a wind turbine generator according to an embodiment of the present disclosure;
fig. 2 is a flowchart of a sub-step S2 in a method for self-diagnosing a fault of a wind turbine generator according to an embodiment of the present disclosure;
fig. 3 is a flowchart of a sub-step S22 in a method for self-diagnosing a fault of a wind turbine generator according to an embodiment of the present disclosure;
fig. 4 is a flowchart of a sub-step S3 in a method for self-diagnosing a fault of a wind turbine generator according to an embodiment of the present disclosure;
fig. 5 is a flowchart of a sub-step S4 in a method for self-diagnosing a fault of a wind turbine generator according to an embodiment of the present disclosure;
fig. 6 is a structural diagram of a wind turbine generator fault self-diagnosis system provided in the embodiment of the present disclosure;
fig. 7 is a structural diagram of a wind turbine provided according to an embodiment of the present disclosure;
fig. 8 is a block diagram of an electronic device provided in accordance with an embodiment of the present disclosure.
Detailed Description
Reference will now be made in detail to the embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and intended to be illustrative of the present disclosure, and should not be construed as limiting the present disclosure.
The following describes a wind turbine generator fault self-diagnosis method and system according to embodiments of the present disclosure with reference to the accompanying drawings.
Fig. 1 is a flowchart of a wind turbine generator fault self-diagnosis method provided by an embodiment of the present disclosure.
As shown in fig. 1, the method for diagnosing a fault of a wind turbine gearbox according to the embodiment of the present disclosure includes, but is not limited to, the following steps:
s1: acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises actual measured values of at least one parameter to be predicted.
The state parameter data are inherent parameter data of the wind turbine generator, exemplarily, the blade length of the wind turbine generator, the altitude, the longitude and the latitude and the like of the wind turbine generator, and the state parameter data are fixed parameters and do not change along with time.
The operation parameter data is parameter data detected in the operation process of the wind turbine generator, and illustratively, the environment temperature, the wind frequency, the power, the rotating speed, the wind speed, the pitch angle, various temperature information, the oil pressure of the gear box and the like can change along with time, and the detected operation parameter data can be different according to different working conditions of the operation of the wind turbine generator.
It should be noted that the above examples of the state parameter data and the operation parameter data are only used for illustration, and are not intended to be specific limitations on the embodiments of the present disclosure, and the state parameter data and the operation parameter data in the embodiments of the present disclosure may also be other parameter data in the above examples, and are not specifically limited by the embodiments of the present disclosure.
In the embodiment of the present disclosure, in the process of the fault self-diagnosis of the wind turbine, the state parameter data of the wind turbine and the operation parameter data within the first preset time period are obtained, corresponding parameter data may be obtained by setting a corresponding parameter measurement detection device, and the specific measurement device and method may use an existing related device to perform automatic measurement or manual measurement, and the like, which is not specifically limited in the embodiment of the present disclosure.
The operating parameter data in the first preset time period includes a real measured value of at least one parameter to be predicted, may include a real measured value of one parameter to be predicted, or may further include two or more real measured values of the parameter to be predicted.
In the embodiment of the present disclosure, the first preset time period may be 1 day, 10 days, 15 days, 30 days, or the like, or the first preset time period may also be one year, which is not specifically limited by the embodiment of the present disclosure.
S2: and determining a first wind turbine group where the wind turbine is located according to the state parameter data and the operation parameter data.
It can be understood that the state parameter data of the wind turbine generator is inherent parameter data of the wind turbine generator, and the operation parameter data is parameter data detected in the operation process of the wind turbine generator.
S3: and determining a first working condition type of the wind turbine generator according to the operation parameter data and the first wind turbine generator group.
It can be understood that the operating parameter data of the wind turbines under different working conditions may be different, and the operating parameter data of the wind turbines under the same working conditions of different wind turbine groups may also be different, further, after the corresponding relation between the operating parameter data of the wind turbines in each wind turbine group and the whole working conditions is determined, the working condition where the wind turbines are located can be determined according to the operating parameter data of the wind turbines, that is, the first working condition category of the wind turbines can be determined according to the operating parameter data and the first wind turbine group.
In an exemplary embodiment, the operation parameter data of the wind turbines includes power, in the first wind turbine group, the full operating condition of the power includes multiple operating condition categories (50-100 kW), (100-.
It can be understood that the above example is merely an illustration, and when the operation parameter data is less and is discontinuous data, the discontinuous single-point value may be further divided into multiple operating condition categories, or a first operating condition category of the wind turbine may also be determined by other parameters besides power, or multiple different operating condition categories of the wind turbine may also be determined according to different parameters, which is not specifically limited in the embodiment of the present disclosure.
S4: and determining a trained fault diagnosis model corresponding to the first working condition type in the first wind turbine group according to the first wind turbine group and the first working condition type.
In the embodiment of the disclosure, the fault diagnosis models can be respectively set for different working condition categories according to different wind turbine generator groups in advance, and after the fault diagnosis models are trained, the trained fault diagnosis models are generated so as to perform fault diagnosis on the working condition categories.
It can be understood that each wind turbine group may include more than one operating condition type, and one fault diagnosis model may be set for different operating conditions according to needs, and of course, each wind turbine group may include many operating condition types.
S5: and inputting the operation parameter data into the trained fault diagnosis model to obtain a model prediction value of the parameter to be predicted.
It can be understood that the trained fault diagnosis model can predict the model prediction value of the parameter to be predicted according to the operation parameter data.
It should be noted that the operation parameter data includes a plurality of data, and when a certain parameter to be predicted is predicted, the data input into the trained fault diagnosis model may be one or more of the operation parameter data, which is not specifically limited by the embodiment of the present disclosure.
S6: and generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted.
In the embodiment of the disclosure, after the operation parameter data is input to the trained fault diagnosis model to obtain the model prediction value of the parameter to be predicted, the model prediction value and the real measurement value can be compared or calculated to generate a fault diagnosis result.
It can be understood that the fault diagnosis result may be no fault or may be a fault, and when the generated fault diagnosis result is a fault, the wind turbine generator may send early warning information to the control center or other monitoring platforms monitoring the wind turbine generator to prompt technicians to overhaul or monitor the wind turbine generator.
The embodiment of the disclosure provides a wind turbine generator fault self-diagnosis method, which includes: acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises actual measured values of at least one parameter to be predicted; determining a first wind turbine group where the wind turbines are located according to the state parameter data and the operation parameter data; determining a first working condition type of the wind turbine generator according to the operation parameter data and the first wind turbine generator group; according to the first wind turbine group and the first working condition type, determining a trained fault diagnosis model corresponding to the first working condition type in the first wind turbine group; inputting the operation parameter data into the trained fault diagnosis model to obtain a model prediction value of the parameter to be predicted; and generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted. Therefore, the method for self-diagnosing the faults of the wind turbine generator under all working conditions is provided, and the corresponding diagnosis strategy is provided according to the model prediction result to realize the self-diagnosis function.
As shown in fig. 2, in some embodiments, in the disclosed embodiments, S2 includes, but is not limited to, the following sub-steps:
s21: and acquiring state parameter sample data of the wind turbine generators and operation parameter sample data within a second preset time length.
The second preset time period may be 1 day, 10 days, 15 days, 30 days, or the like, or the second preset time period may also be one year, which is not specifically limited by the embodiment of the present disclosure.
In the embodiment of the present disclosure, the second preset duration may be equal to the first preset duration.
S22: and performing cluster analysis on the plurality of sample wind turbines according to the state parameter sample data and the operation parameter sample data, and dividing the plurality of sample wind turbines into a plurality of sample wind turbine groups.
In the embodiment of the disclosure, a plurality of wind turbine generator groups can be divided in advance according to different state parameter data and operation parameter sample data, and then the first wind turbine generator group where the wind turbine generator is located can be determined according to the state parameter data and the operation parameter data of the wind turbine generator.
Specifically, state parameter sample data and operation parameter sample data of a plurality of sample wind turbines can be obtained in advance, and the sample wind turbines are subjected to cluster analysis according to the state parameter sample data and the operation parameter sample data and are divided into a plurality of sample wind turbine groups.
As shown in fig. 3, in some embodiments, in the disclosed embodiments, S22 includes, but is not limited to, the following sub-steps:
s221: and performing characteristic quantization processing on the state parameter sample data and the operation parameter sample data to generate quantized state parameter sample data and quantized operation parameter sample data.
In an exemplary embodiment, the state parameter sample data and the operation parameter sample data include a unit parameter, wind resource information, and the like, and are subjected to feature quantization processing, for example: the unit parameters are subjected to numerical coding, and wind resource information is subjected to wind frequency characteristic extraction, turbulence characteristic extraction and the like to generate quantized state parameter sample data and quantized operation parameter sample data.
It should be noted that the above example is only an illustration, and the state parameter sample data and the operation parameter sample data in the embodiment of the present disclosure further include other parameters in the above example, and the above example is not a specific limitation to the embodiment of the present disclosure, and the embodiment of the present disclosure does not specifically limit this.
S222: and performing cluster analysis based on a cluster analysis algorithm according to the quantized state parameter sample data and the quantized operation parameter sample data to obtain a classification result.
In the embodiment of the disclosure, the adopted clustering analysis algorithm can be a Gaussian mixture model, an AP (access point) neighbor propagation clustering algorithm and the like.
Illustratively, in the embodiment of the present disclosure, clustering is performed based on a gaussian mixture analysis algorithm according to quantized state parameter sample data and quantized operating parameter sample data to obtain a classification result.
In the embodiment of the disclosure, based on a cluster analysis algorithm, cluster analysis is performed according to quantized state parameter sample data and quantized operation parameter sample data to obtain a classification result, and a cluster result of a plurality of sample wind turbine generators is obtained and stored in a database for use in subsequent steps.
S223: and dividing the plurality of sample wind turbines into a plurality of sample wind turbine groups according to the classification result.
It can be understood that the classification result may divide the plurality of sample wind turbines into at least two categories, for example, divide the plurality of sample wind turbines into two categories, or three categories or more than three categories, and the like, which is not specifically limited by the embodiment of the present disclosure.
S23: and determining a first wind turbine group where the wind turbines are located by comparing the state parameter data and the operation parameter data with state parameter sample data and operation parameter sample data of different wind turbine groups.
In the embodiment of the disclosure, the first wind turbine group where the wind turbine is located can be determined by comparing the state parameter data and the operation parameter data of the wind turbine with the state sample parameter data and the operation parameter sample data of a plurality of sample wind turbine groups with different classification results stored in the database.
As shown in fig. 4, in some embodiments, in the disclosed embodiments, S3 includes, but is not limited to, the following sub-steps:
s31: and dividing the operation parameter sample data of at least one sample wind turbine generator in the sample wind turbine generator group into sample working condition data under different working conditions.
In some embodiments, S31 specifically includes: extracting working condition data from the operating parameter sample data; dividing each working condition data by adopting a K-fold equal probability method to obtain a plurality of different working condition groups; wherein K is an integer greater than 1; and dividing the operation parameter sample data according to the working condition groups to obtain sample working condition data under different working conditions.
Illustratively, the working condition is power, and multiple working condition data such as (50-100kW ], (100-150 kW) and the like are obtained by selecting every 50kW as one working condition data, then, k-fold equal probability division is carried out on each working condition data, such as 5-fold equal probability division, the division result is represented by labels 1, 2, 3, 4 and 5, namely, the equal probability is divided into 5 equal parts under each working condition, numbers from 1 to 5 are used as working condition groups, and finally, the working condition groups are obtained and stored in a database to prepare for multi-model training.
S32: and determining the first working condition type of the wind turbine generator by comparing the operation parameter data with sample working condition data under different working conditions in the first wind turbine generator group.
In the embodiment of the disclosure, the first working condition category of the wind turbine generator can be determined according to the operation parameter data of the wind turbine generator and the first wind turbine group in which the wind turbine generator is located.
For example, according to the determined sample wind turbine group, the operation parameter sample data of at least one sample wind turbine in the sample wind turbine group is divided into sample working condition data under different working conditions, after the first wind turbine group where the wind turbine is located is determined, the operation parameter data of the wind turbine and the sample working condition data of the sample wind turbine under different working condition categories in the first wind turbine group are compared, and the first working condition category of the wind turbine can be determined.
As shown in fig. 5, in some embodiments, in the disclosed embodiments, S4 includes, but is not limited to, the following sub-steps:
s41: and determining at least one fault diagnosis model suitable for each working condition according to the sample working condition data under different working conditions.
It can be understood that, in the embodiment of the present disclosure, different fault diagnosis models may be selected according to sample working condition data under different working conditions, and multiple fault diagnosis models may also be selected for the working condition sample data of the sample of each working condition, which is not specifically limited by the embodiment of the present disclosure.
S42: and inputting the sample data under each working condition into at least one fault diagnosis model applicable to the working condition, and training the fault diagnosis model to obtain a trained fault diagnosis model.
In the embodiment of the disclosure, after at least one fault diagnosis model used in each working condition is determined, the determined fault diagnosis model used in the working condition is trained according to the sample working condition data in the working condition, so that the trained fault diagnosis model can be obtained.
It can be understood that, under the condition that a plurality of fault diagnosis models are determined under the working condition, each fault diagnosis model is trained to generate a trained fault diagnosis model, and the trained fault diagnosis model with the highest accuracy is screened by comparing the accuracy of the diagnosis results of the plurality of trained fault diagnosis models to serve as the trained fault diagnosis model under the working condition.
In some embodiments, the fault diagnosis model in the embodiments of the present disclosure is a random forest model and/or an extreme gradient boosting tree Xgboost model and/or a neural network model.
It is to be understood that the fault diagnosis model may also be a random forest model, an extreme gradient boosting tree Xgboost model, or other models besides a neural network model, and the above examples are not intended to be specific limitations on the embodiments of the present disclosure, which are not specifically limited by the embodiments of the present disclosure.
S43: and determining at least one trained fault diagnosis model corresponding to the first working condition type in the first wind turbine group according to the first wind turbine group and the first working condition type.
It can be understood that after the first wind turbine group of the wind turbine and the corresponding first working condition type of the first wind turbine group are determined, at least one trained fault diagnosis model corresponding to the first working condition type is used as the trained fault diagnosis model of the wind turbine.
In some embodiments, the method for self-diagnosing the fault of the wind turbine generator provided by the embodiment of the present disclosure further includes: judging whether to trigger model updating or not according to a first state parameter and a first preset model updating parameter in the state parameter data; and under the condition that the first state parameter is a first preset model updating parameter, triggering model updating, and updating the trained fault diagnosis model.
In some embodiments, the first state parameter comprises at least one of:
unit maintenance information; unit fault information; regular maintenance records; key component maintenance information; key component replacement information; adding lubrication to key parts; the model is used for time.
Illustratively, when the first state parameter is the service time of the model, and when the first state parameter is the service time of the model of 2 months and the first preset model update parameter is the service time of the model of 2 months, the first state parameter is the first preset model update parameter, at this time, the model update is triggered, and the trained fault diagnosis model is updated.
It should be noted that, the above example is taken as an illustration, in the embodiment of the present disclosure, the first preset model update parameter may also be that the usage time of the model is 1 month, or 3 months, or 6 months, and the like, which is not limited in the embodiment of the present disclosure.
According to the method and the device for detecting the fault of the wind turbine generator, the trained fault diagnosis model can be retrained under the condition that the model is triggered to be updated according to the first state parameter and the first preset model updating parameter, and the method and the device can be adaptive, self-learned and self-updated, so that the fault detection of the wind turbine generator is more accurate.
In some embodiments, generating a fault diagnosis result according to a model predicted value and a true measured value of a parameter to be predicted in the embodiments of the present disclosure includes: calculating a difference value between the model predicted value and the real measured value, and comparing the difference value with a residual threshold value or a change rate or a change quantity to generate a fault diagnosis result; or generating a fault diagnosis result by comparing the probability distribution of the model predicted value and the real measured value.
Illustratively, in the embodiment of the present disclosure, the fault diagnosis model is to perform fault diagnosis on the temperature of the generator driving-end bearing, select a suitable fault diagnosis model according to the relationship between the power, the rotation speed, the temperature, the pressure, and the like, and the temperature of the generator driving-end bearing, train the fault diagnosis model according to multiple groups of data of the power, the rotation speed, the temperature, the pressure, and the like, and the temperature data of the generator driving-end bearing, generate a trained fault diagnosis model, predict a model predicted value of the temperature of the generator driving-end bearing according to the data of the real-time measurement of the power, the rotation speed, the temperature, the pressure, and the like during the use of the trained fault diagnosis model, compare the model predicted value with the measured value of the temperature of the generator driving-end bearing, obtain whether the temperature of the generator driving-end bearing is abnormal, and further determine whether the generator driving-end bearing is in fault, and obtaining a fault diagnosis result.
Further, the model predicted value is compared with a measured actual measured value of the temperature of the bearing at the driving end of the generator, and a fault diagnosis result can be generated by calculating a difference value between the model predicted value and the actual measured value and comparing the difference value with a residual threshold value or a change rate or a change quantity; alternatively, the fault diagnosis result is generated by comparing the probability distribution of the model predicted value and the true measured value, which is not specifically limited by the embodiment of the present disclosure.
In order to realize the embodiment, the disclosure further provides a fault diagnosis device of the wind turbine gearbox.
Fig. 6 is a structural diagram of a wind turbine generator fault self-diagnosis system provided in the embodiment of the present disclosure.
As shown in fig. 6, the wind turbine generator fault self-diagnosis system 10 according to the embodiment of the present disclosure includes: a data acquisition module 11, a first determination module 12, a second determination module 13, a third determination module 14, a model prediction module 15, and a diagnostic processing module 16.
The data acquisition module 11 is used for acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises actual measured values of the parameter to be predicted.
And the first determining module 12 is configured to determine, according to the state parameter data and the operation parameter data, a first wind turbine group in which the wind turbine is located.
And a second determining module 13, configured to determine a first working condition category of the wind turbine generator according to the operating parameter data and the first wind turbine group.
And a third determining module 14, configured to determine, according to the first wind turbine group and the first working condition category, a trained first fault diagnosis model corresponding to the first working condition category in the first wind turbine group.
And the model prediction module 15 is configured to input the state parameter data to the trained first fault diagnosis model to obtain a model prediction value of the parameter to be predicted.
And the diagnosis processing module 16 is configured to generate a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted.
It should be noted that details which are not disclosed in the wind turbine generator fault self-diagnosis system according to the embodiment of the present disclosure refer to details which are disclosed in the wind turbine generator fault self-diagnosis method according to the embodiment of the present disclosure, and are not described herein again.
In summary, the wind turbine generator fault self-diagnosis system of the embodiment of the disclosure can provide a wind turbine generator fault self-diagnosis method under all working conditions, and provide corresponding diagnosis strategies according to model prediction results to realize a self-diagnosis function.
In order to realize the embodiment, the disclosure further provides a wind turbine generator.
Fig. 7 is a structural diagram of a wind turbine provided according to an embodiment of the present disclosure.
As shown in fig. 7, the wind turbine generator 100 according to the embodiment of the present disclosure includes the wind turbine generator fault self-diagnosis system 10.
The wind turbine generator set provided by the embodiment of the disclosure can provide a method for self-diagnosing the fault of the wind turbine generator set under all working conditions, and provides a corresponding diagnosis strategy according to a model prediction result to realize a self-diagnosis function.
In order to implement the above embodiments, as shown in fig. 8, an embodiment of the present disclosure provides an electronic device 1200, including: the wind turbine generator system fault self-diagnosis method comprises a memory 1201, a processor 1202 and a computer program which is stored in the memory 1201 and can run on the processor 1202, wherein when the processor 1202 executes the program, the wind turbine generator system fault self-diagnosis method is achieved.
In order to implement the above embodiments, the present disclosure provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method for self-diagnosing the fault of the wind turbine generator is implemented.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present disclosure, "a plurality" means two or more unless specifically limited otherwise.
In the present disclosure, unless otherwise expressly stated or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integral; can be mechanically or electrically connected; either directly or indirectly through intervening media, either internally or in any other relationship. The specific meaning of the above terms in the present disclosure can be understood by those of ordinary skill in the art as appropriate.
In the present disclosure, unless otherwise expressly stated or limited, the first feature "on" or "under" the second feature may be directly contacting the first and second features or indirectly contacting the first and second features through an intermediate. Also, a first feature "on," "over," and "above" a second feature may be directly or diagonally above the second feature, or may simply indicate that the first feature is at a higher level than the second feature. A first feature being "under," "below," and "beneath" a second feature may be directly under or obliquely under the first feature, or may simply mean that the first feature is at a lesser elevation than the second feature.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Although embodiments of the present disclosure have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure, and that changes, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present disclosure.
Claims (14)
1. A wind turbine generator fault self-diagnosis method is characterized by comprising the following steps:
acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises true measurements of at least one parameter to be predicted;
determining a first wind turbine group where the wind turbines are located according to the state parameter data and the operation parameter data;
determining a first working condition type of the wind turbine generator according to the operation parameter data and the first wind turbine generator group;
according to the first wind turbine group and the first working condition category, determining a trained fault diagnosis model corresponding to the first working condition category in the first wind turbine group;
inputting the operation parameter data into a trained fault diagnosis model to obtain a model prediction value of the parameter to be predicted;
and generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted.
2. The method of claim 1, wherein determining the first wind turbine group in which the wind turbine is located according to the state parameter data and the operation parameter data comprises:
acquiring state parameter sample data of a plurality of sample wind turbine generators and operating parameter sample data within a second preset time length;
according to the state parameter sample data and the operation parameter sample data, performing cluster analysis on the plurality of sample wind turbine generators, and dividing the cluster analysis into a plurality of sample wind turbine generator groups;
and determining a first wind turbine group where the wind turbine is located by comparing the state parameter data and the operation parameter data with the state parameter sample data and the operation parameter sample data of different wind turbine groups.
3. The method according to claim 2, wherein the performing cluster analysis on the plurality of sample wind turbines according to the state parameter sample data and the operation parameter sample data to divide the plurality of sample wind turbines into a plurality of sample wind turbine groups comprises:
performing characteristic quantization processing on the state parameter sample data and the operation parameter sample data to generate quantized state parameter sample data and quantized operation parameter sample data;
performing cluster analysis based on a cluster analysis algorithm according to the quantized state parameter sample data and the quantized operation parameter sample data to obtain a classification result;
and dividing the plurality of sample wind generation sets into a plurality of sample wind generation sets according to the classification result.
4. The method of claim 3, wherein determining the first operating condition category for the wind turbine based on the operating parameter data and the first group of wind turbines comprises:
dividing the operating parameter sample data of at least one sample wind turbine generator in the sample wind turbine generator group into sample working condition data under different working conditions;
and determining the first working condition type of the wind turbine generator by comparing the operation parameter data with sample working condition data under different working conditions in the first wind turbine generator group.
5. The method according to claim 4, wherein the dividing the operating parameter sample data of at least one of the sample wind turbines in the sample wind turbine group into sample operating condition data under different operating conditions comprises:
extracting working condition data from the operating parameter sample data;
dividing each working condition data by adopting a K-fold equal probability method to obtain a plurality of different working condition groups; wherein K is an integer greater than 1;
and dividing the operation parameter sample data according to the working condition groups to obtain sample working condition data under different working conditions.
6. The method of claim 5, wherein the determining a trained first fault diagnosis model corresponding to the first operating condition category in the first wind turbine group according to the first wind turbine group and the first operating condition category comprises:
determining at least one fault diagnosis model applicable to each working condition according to sample working condition data under different working conditions;
inputting the sample data under each working condition to at least one fault diagnosis model applicable to the working condition, and training the fault diagnosis model to obtain a trained fault diagnosis model;
and determining at least one trained fault diagnosis model corresponding to the first working condition type in the first wind turbine group according to the first wind turbine group and the first working condition type.
7. The method according to claim 6, characterized in that the fault diagnosis model is a random forest model and/or an extreme gradient boosting tree Xgboost model and/or a neural network model.
8. The method of claim 1, further comprising:
judging whether to trigger model updating or not according to a first state parameter and a first preset model updating parameter in the state parameter data;
and under the condition that the first state parameter is the first preset model updating parameter, triggering model updating, and updating the trained first fault model.
9. The method of claim 8, wherein the first state parameter comprises at least one of:
unit maintenance information;
unit fault information;
regular maintenance records;
key component maintenance information;
key component replacement information;
adding lubrication to key parts;
the model is used for time.
10. The method according to claim 1, wherein the generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted comprises:
calculating a difference value between the model predicted value and the real measured value, and comparing the difference value with a residual threshold value or a change rate or a change quantity to generate a fault diagnosis result;
or generating a fault diagnosis result by comparing the probability distribution of the model predicted value and the real measured value.
11. The utility model provides a wind turbine generator system fault self-diagnosis system which characterized in that includes:
the data acquisition module is used for acquiring state parameter data of the wind turbine generator and operation parameter data within a first preset time length; wherein the operating parameter data comprises actual measurements of the parameter to be predicted;
the first determining module is used for determining a first wind turbine group where the wind turbine is located according to the state parameter data and the operation parameter data;
the second determining module is used for determining the first working condition category of the wind turbine generator according to the operation parameter data and the first wind turbine generator group;
the third determining module is used for determining a trained first fault diagnosis model corresponding to the first working condition type in the first wind turbine group according to the first wind turbine group and the first working condition type;
the model prediction module is used for inputting the state parameter data into the trained first fault diagnosis model to obtain a model prediction value of the parameter to be predicted;
and the diagnosis processing module is used for generating a fault diagnosis result according to the model predicted value and the real measured value of the parameter to be predicted.
12. A wind turbine, comprising: the wind turbine generator system fault self-diagnosis system according to claim 11.
13. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and operable on the processor, wherein the processor executes the program to realize the wind turbine generator fault self-diagnosis method according to any one of claims 1 to 10.
14. A computer-readable storage medium on which a computer program is stored, characterized in that the program, when executed by a processor, implements the wind turbine generator set fault self-diagnosis method according to any one of claims 1 to 10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111248577.0A CN114065842A (en) | 2021-10-26 | 2021-10-26 | Wind turbine generator fault self-diagnosis method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111248577.0A CN114065842A (en) | 2021-10-26 | 2021-10-26 | Wind turbine generator fault self-diagnosis method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114065842A true CN114065842A (en) | 2022-02-18 |
Family
ID=80235538
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111248577.0A Pending CN114065842A (en) | 2021-10-26 | 2021-10-26 | Wind turbine generator fault self-diagnosis method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114065842A (en) |
-
2021
- 2021-10-26 CN CN202111248577.0A patent/CN114065842A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bhardwaj et al. | Reliability prediction of an offshore wind turbine gearbox | |
| CN111597682B (en) | Method for predicting remaining life of bearing of gearbox of wind turbine | |
| CN109524139B (en) | Real-time equipment performance monitoring method based on equipment working condition change | |
| Artigao et al. | Wind turbine reliability: A comprehensive review towards effective condition monitoring development | |
| CN102022264B (en) | System and method for wind turbine health management | |
| Cui et al. | An anomaly detection approach based on machine learning and scada data for condition monitoring of wind turbines | |
| CN115013261B (en) | State monitoring method and system for offshore wind farm | |
| CN107291991B (en) | Early defect early warning method for wind turbine generator based on dynamic network sign | |
| CN110414154B (en) | Fan component temperature abnormity detection and alarm method with double measuring points | |
| CN105205569A (en) | Draught fan gear box state on-line evaluation model building method and on-line evaluation method | |
| CN111415070A (en) | Wind turbine generator gearbox oil temperature over-temperature fault early warning method based on SCADA data | |
| CN114036656B (en) | Fault diagnosis method and device for wind turbine generator gearbox | |
| CN107728059A (en) | A kind of pitch-controlled system state evaluating method | |
| CN111878322A (en) | Wind power generator device | |
| CN110969185A (en) | Equipment abnormal state detection method based on data reconstruction | |
| Shi et al. | Study of wind turbine fault diagnosis and early warning based on SCADA data | |
| Kim et al. | Design of wind turbine fault detection system based on performance curve | |
| CN112308391A (en) | Real-time monitoring and anomaly detection method for equipment state based on neural network | |
| Yongjie et al. | Research on early fault diagnostic method of wind turbines | |
| Yang et al. | Application research of arima model in wind turbine gearbox fault trend prediction | |
| CN117649044A (en) | Multi-dimensional quantitative state evaluation method for offshore wind turbine generator | |
| CN113158535A (en) | Wind turbine generator main bearing temperature prediction method based on deep learning | |
| CN114065842A (en) | Wind turbine generator fault self-diagnosis method and system | |
| CN114091197B (en) | Method for evaluating service life of whole bearing part of wind turbine generator | |
| CN113705076B (en) | Wind turbine generator system gearbox state monitoring method and control device based on partial least square method |
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 |