CN114037127A - Fault prediction method and device for wind turbine generator - Google Patents
Fault prediction method and device for wind turbine generator Download PDFInfo
- Publication number
- CN114037127A CN114037127A CN202111243323.XA CN202111243323A CN114037127A CN 114037127 A CN114037127 A CN 114037127A CN 202111243323 A CN202111243323 A CN 202111243323A CN 114037127 A CN114037127 A CN 114037127A
- Authority
- CN
- China
- Prior art keywords
- primary model
- sample
- parameters
- parameter
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000000605 extraction Methods 0.000 claims abstract description 24
- 238000012216 screening Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 20
- 230000002159 abnormal effect Effects 0.000 claims description 15
- 238000004590 computer program Methods 0.000 claims description 12
- 238000004140 cleaning Methods 0.000 claims description 11
- 238000012423 maintenance Methods 0.000 description 13
- 230000008569 process Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 6
- 238000010248 power generation Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 229910052799 carbon Inorganic materials 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Biophysics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computing Systems (AREA)
- Quality & Reliability (AREA)
- Game Theory and Decision Science (AREA)
- Biomedical Technology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Entrepreneurship & Innovation (AREA)
- Molecular Biology (AREA)
- Computational Linguistics (AREA)
- Evolutionary Biology (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
Abstract
The invention discloses a method and a device for predicting faults of a wind turbine generator. The method comprises the following steps: the method comprises the steps of collecting operation parameters of all subsystems of the wind turbine generator, determining target operation parameters in the operation parameters, taking one of the target operation parameters as an output parameter of a primary model, taking other parameters except one of the target operation parameters as input parameters of the primary model, obtaining a predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model, and obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and a secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
Description
Technical Field
The invention relates to the technical field of energy, in particular to a method and a device for predicting a fault of a wind turbine generator, the wind turbine generator, electronic equipment and a storage medium.
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 and can convert wind energy into alternating current energy. Because the operating environment of the wind turbine generator is severe and the terrain is complex, and the wind turbine generator is a complex and variable-working-condition large-scale rotating device, a large number of faults and maintenance conditions can occur in the operation. The current operation and maintenance mode of the wind power plant mainly takes passive maintenance and regular maintenance as main factors, so that the operation and maintenance cost is high. In order to reduce the operation and maintenance cost, predictive maintenance is the key research direction. The aim of predictive maintenance is to change passive maintenance into active maintenance, to eliminate hidden troubles before failure occurs and to reduce the downtime of the failure. Meanwhile, unnecessary regular maintenance is reduced, maintenance resources are allocated macroscopically, and operation and maintenance cost is reduced.
Therefore, how to enhance the accuracy and the universality of the fault prediction of the wind turbine generator system becomes a problem to be solved urgently in the field of wind power generation.
Disclosure of Invention
The present invention is directed to solving, at least to some extent, one of the technical problems in the art described above.
Therefore, a first object of the present invention is to provide a method for predicting a fault of a wind turbine generator, which includes collecting operating parameters of subsystems of the wind turbine generator, determining target operating parameters in the operating parameters, sequentially using one of the target operating parameters as an output parameter of a primary model, using other parameters except one of the target operating parameters as input parameters of the primary model, obtaining a predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model, and obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and a secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
The second purpose of the invention is to provide a fault prediction device of a wind turbine generator.
The third purpose of the invention is to provide a wind turbine generator.
A fourth object of the invention is to propose an electronic device.
A fifth object of the present invention is to propose a computer-readable storage medium.
An embodiment of a first aspect of the present invention provides a method for predicting a fault of a wind turbine generator, including: collecting operating parameters of subsystems of a wind turbine generator; determining a target operating parameter of the operating parameters; one parameter in the target operation parameters is used as an output parameter of a primary model in sequence, and other parameters except the one parameter in the target operation parameters are used as input parameters of the primary model; obtaining a predicted value of an output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model; and obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model.
According to the fault prediction method of the wind turbine generator, the operation parameters of each subsystem of the wind turbine generator are collected, the target operation parameters in the operation parameters are determined, one of the target operation parameters is used as the output parameter of the primary model in sequence, other parameters except one of the target operation parameters are used as the input parameters of the primary model, the predicted value of the output parameter of the primary model is obtained according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model, and the fault prediction result is obtained according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
In addition, the method for predicting the fault of the wind turbine generator set provided by the embodiment of the invention can also have the following additional technical characteristics:
in one embodiment of the invention, the method further comprises: the determining a target operating parameter of the operating parameters includes: cleaning abnormal data in the operation parameters; and determining the target operation parameters in the cleaned operation parameters according to the sample target operation parameters.
In one embodiment of the invention, the method further comprises: further comprising: collecting sample operation parameters of each subsystem of the wind turbine generator; cleaning abnormal data in the sample operation parameters; preliminarily screening the cleaned sample operation parameters according to expert experience; carrying out variance screening on the preliminarily screened sample operation parameters, and eliminating the sample operation parameters with the variance smaller than a preset variance threshold value; and performing correlation screening on the sample operation parameters subjected to the variance screening, and determining the sample operation parameters with the correlation number larger than a preset correlation coefficient threshold value as the sample target operation parameters.
In one embodiment of the invention, the method further comprises: one parameter in the sample target operation parameters is used as a sample output parameter of a primary model, and other parameters except the one parameter in the sample target operation parameters are used as sample input parameters of the primary model; carrying out feature extraction on the sample input parameters of the primary model to obtain the sample input parameter features of the primary model; and taking the sample input parameter characteristics as input, taking the sample output parameters as output, and training a primary model to be trained to obtain the primary model.
In one embodiment of the invention, the method further comprises: the obtaining a predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model comprises: performing feature extraction on the input parameters of the primary model to obtain the input parameter features of the primary model; and inputting the input parameter characteristics of the primary model into the primary model corresponding to the output parameters of the primary model to obtain the predicted value of the output parameters of the primary model.
In one embodiment of the invention, the method further comprises: the obtaining of the fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model comprises: calculating the difference value between the output parameter of the primary model and the predicted value of the output parameter of the primary model to obtain residual error characteristics; generating residual error state characteristics according to the residual error characteristics and the time information; and inputting the residual error characteristics, the residual error state characteristics and the working condition data into the secondary model to obtain the fault prediction result.
In one embodiment of the invention, the method further comprises: calculating the difference value between the sample output parameters of the primary model and the predicted values of the sample output parameters of the primary model in a preset number of time periods before the sample fault occurs to obtain sample residual error characteristics; generating a sample residual error state characteristic according to the sample residual error characteristic and the sample time information; and taking the sample residual error characteristics, the sample residual error state characteristics and the sample working condition data as input, taking the sample fault as output, and training a secondary model to be trained to obtain the secondary model.
An embodiment of a second aspect of the present invention provides a failure prediction apparatus for a wind turbine, including: the acquisition module is used for acquiring the operating parameters of each subsystem of the wind turbine generator; the first determination module is used for determining a target operation parameter in the operation parameters; the second determining module is used for sequentially taking one parameter in the target operation parameters as an output parameter of the primary model and taking other parameters except the one parameter in the target operation parameters as input parameters of the primary model; the first prediction module is used for obtaining a predicted value of an output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model; and the second prediction module is used for obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary models.
The fault prediction device of the wind turbine generator system acquires the operation parameters of each subsystem of the wind turbine generator system, determines the target operation parameters in the operation parameters, sequentially uses one of the target operation parameters as the output parameter of the primary model, uses other parameters except one of the target operation parameters as the input parameters of the primary model, obtains the predicted value of the output parameter of the primary model according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model, and obtains the fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
In addition, the failure prediction device for the wind turbine generator set provided by the above embodiment of the present invention may further have the following additional technical features:
in an embodiment of the present invention, the first determining module is specifically configured to: cleaning abnormal data in the operation parameters; and determining the target operation parameters in the cleaned operation parameters according to the sample target operation parameters.
In an embodiment of the present invention, the first determining module is further configured to: collecting sample operation parameters of each subsystem of the wind turbine generator; cleaning abnormal data in the sample operation parameters; preliminarily screening the cleaned sample operation parameters according to expert experience; carrying out variance screening on the preliminarily screened sample operation parameters, and eliminating the sample operation parameters with the variance smaller than a preset variance threshold value; and performing correlation screening on the sample operation parameters subjected to the variance screening, and determining the sample operation parameters with the correlation number larger than a preset correlation coefficient threshold value as the sample target operation parameters.
In one embodiment of the invention, the apparatus further comprises: a first training module to: one parameter in the sample target operation parameters is used as a sample output parameter of a primary model, and other parameters except the one parameter in the sample target operation parameters are used as sample input parameters of the primary model; carrying out feature extraction on the sample input parameters of the primary model to obtain the sample input parameter features of the primary model; and taking the sample input parameter characteristics as input, taking the sample output parameters as output, and training a primary model to be trained to obtain the primary model.
In an embodiment of the present invention, the first prediction module is specifically configured to: performing feature extraction on the input parameters of the primary model to obtain the input parameter features of the primary model; and inputting the input parameter characteristics of the primary model into the primary model corresponding to the output parameters of the primary model to obtain the predicted value of the output parameters of the primary model.
In an embodiment of the present invention, the second prediction module is specifically configured to: calculating the difference value between the output parameter of the primary model and the predicted value of the output parameter of the primary model to obtain residual error characteristics; generating residual error state characteristics according to the residual error characteristics and the time information; and inputting the residual error characteristics, the residual error state characteristics and the working condition data into the secondary model to obtain the fault prediction result.
In one embodiment of the invention, the apparatus further comprises: a second training module to: calculating the difference value between the sample output parameters of the primary model and the predicted values of the sample output parameters of the primary model in a preset number of time periods before the sample fault occurs to obtain sample residual error characteristics; generating a sample residual error state characteristic according to the sample residual error characteristic and the sample time information; and taking the sample residual error characteristics, the sample residual error state characteristics and the sample working condition data as input, taking the sample fault as output, and training a secondary model to be trained to obtain the secondary model.
An embodiment of a third aspect of the present invention provides a wind turbine generator, including: the failure prediction device for a wind turbine generator according to the embodiment of the second aspect of the present invention.
The wind turbine generator system of the embodiment of the invention acquires the operation parameters of each subsystem of the wind turbine generator system, determines the target operation parameters in the operation parameters, sequentially takes one of the target operation parameters as the output parameter of a primary model, takes other parameters except one of the target operation parameters as the input parameters of the primary model, obtains the predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model, and obtains the fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
A fourth aspect of the present invention provides an electronic device, including: the invention relates to a fault prediction method for a wind turbine generator, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein when the processor executes the program, the fault prediction method for the wind turbine generator is realized.
According to the electronic equipment provided by the embodiment of the invention, a processor executes a computer program stored on a memory, the operating parameters of each subsystem of the wind turbine generator are collected, the target operating parameters in the operating parameters are determined, one of the target operating parameters is sequentially used as the output parameter of a primary model, other parameters except one of the target operating parameters are used as the input parameters of the primary model, the predicted value of the output parameter of the primary model is obtained according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model, and the fault prediction result is obtained according to the predicted value of the output parameter of each primary model and a secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
An embodiment of a fifth aspect of the present application 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 predicting a fault of a wind turbine generator according to the embodiment of the first aspect of the present invention.
The computer-readable storage medium of the embodiment of the invention stores a computer program and is executed by a processor, acquires operating parameters of each subsystem of the wind turbine generator, determines target operating parameters in the operating parameters, sequentially takes one of the target operating parameters as an output parameter of a primary model, takes other parameters except one of the target operating parameters as input parameters of the primary model, obtains a predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model, and obtains a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
Additional aspects and advantages of the invention 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 invention.
Drawings
The foregoing and/or additional aspects and advantages of the present invention 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 schematic flow diagram of a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart illustrating a process of determining target operating parameters in a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart illustrating a process of determining a sample target operating parameter in a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention;
FIG. 4 is a schematic flow chart illustrating a predicted value of an output parameter for generating a primary model in a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention;
FIG. 5 is a schematic flow chart illustrating a training process of a primary model in a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention;
fig. 6 is a schematic flow chart illustrating a fault prediction result generated in the fault prediction method for a wind turbine generator according to a specific example of the present invention;
FIG. 7 is a schematic flow chart illustrating a training process of a secondary model in a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention;
fig. 8 is a schematic structural diagram of a failure prediction apparatus of a wind turbine generator according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of a wind turbine generator according to an embodiment of the present invention; and
fig. 10 is a schematic structural diagram of an electronic device according to an embodiment of the invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, 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 function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
The following describes a method and an apparatus for predicting a failure of a wind turbine generator, the wind turbine generator, an electronic device, and a storage medium according to an embodiment of the present invention with reference to the drawings.
Fig. 1 is a schematic flow chart of a method for predicting a fault of a wind turbine generator according to an embodiment of the present invention.
As shown in fig. 1, the method for predicting a fault of a wind turbine generator according to the embodiment of the present invention includes:
and S101, collecting the operation parameters of each subsystem of the wind turbine generator.
In the embodiment of the invention, the wind turbine generator is divided into a plurality of subsystems according to expert experience, for example: generator, gearbox, main bearing, pitch system and yaw system etc.. The method includes the steps of collecting operation parameters of each subsystem, And for example, obtaining the operation parameters of each subsystem of the wind turbine generator system through a Data collection And monitoring Control system (Scada for short).
And S102, determining a target operation parameter in the operation parameters.
In some embodiments, target operation parameters corresponding to each subsystem of the wind turbine are determined from the collected operation parameters of each subsystem of the wind turbine, so as to perform fault prediction on each subsystem of the wind turbine based on the target operation parameters, thereby realizing fault prediction on the wind turbine, where the target operation parameters may be parameters related to a normal working state of each subsystem of the wind turbine, for example, the target operation parameters corresponding to the gear box may be a high-speed shaft gear box inlet temperature, a gear box oil temperature, a gear box bearing drive end temperature, a gear box bearing non-drive end temperature, a gear box inlet oil pressure, a gear box outlet oil pressure, and the like.
S103, one of the target operation parameters is taken as an output parameter of the primary model, and other parameters except one of the target operation parameters are taken as input parameters of the primary model.
In some embodiments, for a subsystem of a wind turbine generator, one of a plurality of target operating parameters corresponding to the subsystem is sequentially used as an output parameter of a primary model, and other parameters except for one determined as the output parameter in the plurality of target operating parameters corresponding to the subsystem are used as input parameters of the primary model, that is: and each parameter in a plurality of target operation parameters corresponding to one subsystem is used as an output parameter of the primary model to be predicted in sequence. Therefore, a plurality of target operation parameters corresponding to each subsystem of the wind turbine generator are respectively predicted through the primary model.
For example, the plurality of target operation parameters corresponding to the gear box may be target operation parameters corresponding to the gear box, such as a high-speed shaft gear box inlet temperature, a gear box oil temperature, a gear box bearing drive end temperature, a gear box bearing non-drive end temperature, a gear box inlet oil pressure, and a gear box outlet oil pressure, where the high-speed shaft gear box inlet temperature is used as an output parameter of the primary model, and the gear box oil temperature, the gear box bearing drive end temperature, the gear box bearing non-drive end temperature, the gear box inlet oil pressure, and the gear box outlet oil pressure are used as input parameters of the gear box model; and thirdly, taking the oil temperature of the gearbox as an output parameter of the primary model, and taking the inlet temperature of the gearbox of the high-speed shaft, the bearing driving end temperature of the gearbox, the non-driving end temperature of the bearing of the gearbox, the inlet oil pressure of the gearbox and the outlet oil pressure of the gearbox as input parameters of the primary model, so as to sequentially realize the prediction of each target operation parameter.
And S104, obtaining a predicted value of the output parameter of the primary model according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model.
In some embodiments, the output parameter of each primary model determined in step S103 as the state characteristic of the wind turbine subsystem may correspond to one primary model, the primary model corresponding to the parameter is obtained according to the output parameter of the primary model, and the output parameter of the primary model is predicted according to the input parameter of the primary model corresponding to the output parameter of the primary model and the primary model corresponding to the output parameter of the primary model, so as to obtain a corresponding predicted value.
In some embodiments, the above procedure is circulated to sequentially obtain the prediction of the output parameter of each primary model determined in step S103, so as to obtain a corresponding predicted value, thereby obtaining the predicted value of the target operating parameter corresponding to each subsystem of the wind turbine.
In some embodiments, the primary model corresponding to the output parameters of the primary model may be obtained from a hard disk or a database, and may be a pre-trained primary model.
In some embodiments, the predicted values of the output parameters of the plurality of primary models may be stored in a database, so that the predicted values of the target operating parameters corresponding to the subsystems of the wind turbine generator may be obtained in time when predicting the fault of the wind turbine generator.
And S105, obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model.
In some embodiments, the wind turbine generator is subjected to fault prediction according to the predicted value of the output parameter of each primary model and the secondary model, and a fault prediction result is obtained.
In summary, according to the method for predicting the fault of the wind turbine generator, the operation parameters of each subsystem of the wind turbine generator are collected, the target operation parameter in the operation parameters is determined, one of the target operation parameters is sequentially used as the output parameter of the primary model, other parameters except one of the target operation parameters are used as the input parameters of the primary model, the predicted value of the output parameter of the primary model is obtained according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model, and the fault prediction result is obtained according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
On the basis of the embodiment shown in fig. 1, as shown in fig. 2, the "determining a target operating parameter" in the operating parameters in step S102 may specifically include the following steps:
s201, cleaning abnormal data in the operation parameters.
In some embodiments, the collected operation parameters may be preprocessed by data cleaning, for example, abnormal data in the operation parameters may be cleaned: judging the validity of the sensor data, and rejecting abnormal sensor data; according to the running condition information of the wind turbine generator, eliminating wind turbine generator shutdown data and fault shutdown data; according to the running state information of the wind turbine generator, abnormal power generation data of the wind turbine generator, such as limited power data, are removed, so that the accuracy of fault prediction is enhanced.
S202, determining target operation parameters in the cleaned operation parameters according to the sample target operation parameters.
In the embodiment of the invention, the target operation parameters can be determined from the cleaned operation parameters according to the pre-obtained sample target operation parameters.
On the basis of any of the above embodiments, as shown in fig. 3, the method for predicting a fault of a wind turbine generator according to the embodiment of the present invention may further include a process of determining a sample target operation parameter, and specifically may include the following steps:
s301, collecting sample operation parameters of each subsystem of the wind turbine generator.
In some embodiments, historical operating data of each subsystem of the wind turbine is collected as sample operating parameters based on Scada.
And S302, cleaning abnormal data in the sample operation parameters.
In some embodiments, the abnormal data in the sample operation parameters are cleaned to remove the abnormal data and retain valid data, and the specific implementation is similar to the above embodiments, and is not described herein again.
And S303, preliminarily screening the cleaned sample operation parameters according to expert experience.
In some embodiments, the cleaned sample operation parameters are preliminarily screened according to expert experience, and the operation parameters related to the fault state of the corresponding wind turbine subsystem in the sample operation parameters are determined.
S304, carrying out variance screening on the preliminarily screened sample operation parameters, and eliminating the sample operation parameters with the variance smaller than a preset variance threshold value.
In some embodiments, the variance screening may be continued on the preliminarily screened sample operating parameters, and the variance σ of each sample operating parameter may be calculated by the following variance calculation formulax：
Wherein x is a sample operating parameter, i is the number of the sample operating parameters,
is the mean of the sample run parameters.
And according to the variance of each sample operation parameter, eliminating the sample operation parameters of which the variance is smaller than a preset variance threshold, wherein the preset variance threshold can be set according to needs, and the method is not limited in the invention.
S305, carrying out relevance screening on the sample operation parameters subjected to the variance screening, and determining the sample operation parameters with the relevance number larger than a preset correlation coefficient threshold value as sample target operation parameters.
In some embodiments, the correlation coefficient of the variance-filtered sample operation parameter is calculated, the variance-filtered sample operation parameter is subjected to correlation degree screening based on the correlation coefficient, and the sample operation parameter of which the correlation number is greater than a preset correlation coefficient threshold is determined as the sample target operation parameter, where the preset correlation coefficient threshold may be set as required, which is not limited in the present invention.
Wherein, the correlation coefficient calculation formula is as follows:
where x represents a sample operating parameter and y represents the remaining sample operating parameters except for x.
On the basis of any of the above embodiments, as shown in fig. 4, in step S104, "obtaining a predicted value of an output parameter of a primary model according to a primary model corresponding to an input parameter of the primary model and an output parameter of the primary model" may specifically include the following steps:
s401, extracting the characteristics of the input parameters of the primary model to obtain the input parameter characteristics of the primary model.
In some embodiments, a Convolutional Neural Network (CNN) may be used to perform feature extraction on the input parameters of the primary model, so as to obtain input parameter features of the primary model.
In some embodiments, different input parameter characteristics may be obtained by multiple convolution-pooling operations.
In some embodiments, the input parameters of the primary model may be normalized before feature extraction.
S402, inputting the input parameter characteristics of the primary model into the primary model corresponding to the output parameters of the primary model to obtain the predicted value of the output parameters of the primary model.
In some embodiments, the extracted input parameter characteristics of the primary model are input into the primary model corresponding to the output parameters of the primary model, and output parameter prediction is performed to obtain the predicted value of the output parameters of the primary model.
On the basis of any of the above embodiments, as shown in fig. 5, the method for predicting a fault of a wind turbine generator according to the embodiment of the present invention may further include a training process of a primary model, and specifically may include the following steps:
s501, one parameter in the sample target operation parameters is taken as a sample output parameter of the primary model, and other parameters except the one parameter in the sample target operation parameters are taken as sample input parameters of the primary model.
It should be noted here that the determination manner of the sample output parameters and the input parameters of the primary model is similar to that of the above embodiment and is not repeated here.
And S502, performing feature extraction on the sample input parameters of the primary model to obtain the sample input parameter features of the primary model.
In some embodiments, feature extraction is performed on the sample input parameters of the primary model based on a convolutional neural network, so as to obtain sample input parameter features of the primary model.
And S503, training the primary model to be trained by taking the sample input parameter characteristics as input and the sample output parameters as output to obtain the primary model.
In some embodiments, a primary model can be constructed based on a random forest model, sample input parameter characteristics are used as input of the model, sample output parameters are used as output of the model, and the model is subjected to hyper-parameter optimization in a grid search mode, so that an optimal training model is obtained and used as the primary model for target operation parameter prediction.
Wherein, the mean square error can be used as a criterion to carry out the hyper-parameter optimization.
Therefore, the trained primary models corresponding to the sample output parameters can be generated through the process, and the primary models are stored in a hard disk or a database, so that the required primary models can be obtained in time to predict the target operation parameters.
On the basis of any of the above embodiments, as shown in fig. 6, the step S105 of "obtaining the failure prediction result according to the predicted value of the output parameter of each primary model and the secondary model" may specifically include the following steps:
s601, calculating the difference value between the output parameter of the primary model and the predicted value of the output parameter of the primary model to obtain residual error characteristics.
In some embodiments, the difference between the output parameter of each primary model and the predicted value of the output parameter of the primary model is calculated, so as to obtain the residual error characteristic of the output parameter of each primary model.
For example, a residual threshold, that is, a difference threshold between the output parameter of the primary model and the predicted value of the output parameter of the primary model is set, a state where the residual is equal to or greater than the residual threshold is set to 1, and a state where the residual is less than the residual threshold is set to 0, thereby obtaining the residual feature.
And S602, generating residual state characteristics according to the residual characteristics and the time information.
In some embodiments, the time corresponding to the target operating parameter is divided into N time periods according to the time information, the residual error characteristics in each time period are accumulated to obtain a window accumulated characteristic, and the residual error state characteristic is generated based on the window accumulated characteristic by setting a threshold V, for example, a state where the window accumulated characteristic in one time window is greater than or equal to the threshold V is set to 1, and a state where the window accumulated characteristic is less than the threshold V is set to 0, so as to obtain the residual error state characteristic.
And S603, inputting the residual error characteristics, the residual error state characteristics and the working condition data into a secondary model to obtain a fault prediction result.
In some embodiments, the residual error characteristics, the residual error state characteristics and the working condition data are input into the secondary model as input data of the model, and a fault prediction result is obtained. Wherein, the trained secondary model can be obtained from a database or a hard disk. The working condition parameters can be power, rotating speed, wind speed, pitch angle and the like.
On the basis of any of the above embodiments, as shown in fig. 7, the method for predicting a fault of a wind turbine generator according to the embodiment of the present invention may further include a training process of a secondary model, and specifically may include the following steps:
s701, calculating the difference value between the sample output parameters of the primary model and the predicted values of the sample output parameters of the primary model in a preset number of time periods before the sample fault occurs, and obtaining the residual error characteristics of the sample.
In some embodiments, a time period T is set, n time periods are preset, the output parameters of the primary model within n time periods T before the sample fault occurs are selected, the predicted values of the sample output parameters of the corresponding primary model are obtained from a database or a hard disk, and the difference between the sample output parameters of the primary model and the predicted values of the sample output parameters of the primary model is calculated to obtain the residual error characteristics of the sample within n time periods before the sample fault occurs.
And S702, generating a sample residual error state characteristic according to the sample residual error characteristic and the sample time information.
It should be noted that, the generation method of the sample residual error state feature is similar to the above embodiment, and is not described herein again.
And S703, training the secondary model to be trained by taking the sample residual error characteristics, the sample residual error state characteristics and the sample working condition data as input and taking the sample faults as output to obtain the secondary model.
In some embodiments, a secondary model is constructed based on Deep Neural Networks (DNNs) and expert experience, different failure modes of each subsystem of the wind turbine are combined, and a secondary model for failure prediction of the wind turbine is generated based on a prediction result of the primary model. And taking the sample residual error characteristics, the sample residual error state characteristics and the sample working condition data as the input of the model, taking the corresponding sample faults as the output of the model, training the secondary model to be trained to obtain the optimal parameters of the model, obtaining the secondary model, and storing the model into a database or a hard disk so as to predict the faults based on the secondary model.
In order to implement the above embodiment, the invention further provides a failure prediction device of the wind turbine.
Fig. 8 is a schematic structural diagram of a failure prediction apparatus for a wind turbine generator according to an embodiment of the present invention.
As shown in fig. 8, a failure prediction apparatus 1000 of a wind turbine generator according to an embodiment of the present invention includes: an acquisition module 1001, a first determination module 1002, a second determination module 1003, a first prediction module 1004, and a second prediction module 1005.
And the acquisition module 1001 is used for acquiring the operating parameters of each subsystem of the wind turbine generator.
A first determining module 1002, configured to determine a target operation parameter of the operation parameters.
A second determining module 1003, configured to sequentially use one of the target operating parameters as an output parameter of a primary model, and use other parameters than the one parameter in the target operating parameters as input parameters of the primary model.
The first prediction module 1004 is configured to obtain a predicted value of an output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model.
A second prediction module 1005, configured to obtain a failure prediction result according to the predicted value of the output parameter of each primary model and the secondary model.
In an embodiment of the present invention, the first determining module 1002 is specifically configured to: cleaning abnormal data in the operation parameters; and determining the target operation parameters in the cleaned operation parameters according to the sample target operation parameters.
In an embodiment of the present invention, the first determining module 1002 is further configured to: collecting sample operation parameters of each subsystem of the wind turbine generator; cleaning abnormal data in the sample operation parameters; preliminarily screening the cleaned sample operation parameters according to expert experience; carrying out variance screening on the preliminarily screened sample operation parameters, and eliminating the sample operation parameters with the variance smaller than a preset variance threshold value; and performing correlation screening on the sample operation parameters subjected to the variance screening, and determining the sample operation parameters with the correlation number larger than a preset correlation coefficient threshold value as the sample target operation parameters.
In an embodiment of the present invention, the failure prediction apparatus for a wind turbine further includes: a first training module to: one parameter in the sample target operation parameters is used as a sample output parameter of a primary model, and other parameters except the one parameter in the sample target operation parameters are used as sample input parameters of the primary model; carrying out feature extraction on the sample input parameters of the primary model to obtain the sample input parameter features of the primary model; and taking the sample input parameter characteristics as input, taking the sample output parameters as output, and training a primary model to be trained to obtain the primary model.
In an embodiment of the present invention, the first prediction module 1004 is specifically configured to: performing feature extraction on the input parameters of the primary model to obtain the input parameter features of the primary model; and inputting the input parameter characteristics of the primary model into the primary model corresponding to the output parameters of the primary model to obtain the predicted value of the output parameters of the primary model.
In an embodiment of the present invention, the second prediction module 1005 is specifically configured to: calculating the difference value between the output parameter of the primary model and the predicted value of the output parameter of the primary model to obtain residual error characteristics; generating residual error state characteristics according to the residual error characteristics and the time information; and inputting the residual error characteristics, the residual error state characteristics and the working condition data into the secondary model to obtain the fault prediction result.
In an embodiment of the present invention, the failure prediction apparatus for a wind turbine further includes: a second training module to: calculating the difference value between the sample output parameters of the primary model and the predicted values of the sample output parameters of the primary model in a preset number of time periods before the sample fault occurs to obtain sample residual error characteristics; generating a sample residual error state characteristic according to the sample residual error characteristic and the sample time information; and taking the sample residual error characteristics, the sample residual error state characteristics and the sample working condition data as input, taking the sample fault as output, and training a secondary model to be trained to obtain the secondary model.
It should be noted that details that are not disclosed in the fault prediction apparatus for a wind turbine generator according to the embodiment of the present invention refer to details that are disclosed in the fault prediction method for a wind turbine generator according to the embodiment of the present invention, and are not described herein again.
To sum up, the fault prediction apparatus for a wind turbine generator according to the embodiment of the present invention acquires operating parameters of each subsystem of the wind turbine generator, determines a target operating parameter of the operating parameters, sequentially uses one of the target operating parameters as an output parameter of a primary model, uses other parameters except one of the target operating parameters as input parameters of the primary model, obtains a predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model, and obtains a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
In order to realize the embodiment, the invention further provides a wind turbine generator.
Fig. 9 is a schematic structural diagram of a wind turbine generator according to an embodiment of the present invention.
As shown in fig. 9, a wind turbine 1100 according to an embodiment of the present invention includes the above-described failure prediction apparatus 1000 for a wind turbine.
According to the wind turbine generator, the target operation parameters in the operation parameters can be determined by collecting the operation parameters of each subsystem of the wind turbine generator, one of the target operation parameters is sequentially used as the output parameter of the primary model, other parameters except one parameter in the target operation parameters are used as the input parameters of the primary model, the predicted value of the output parameter of the primary model is obtained according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model, and the fault prediction result is obtained according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
In order to implement the above embodiment, as shown in fig. 10, an embodiment of the present invention provides an electronic device 1200, including: the wind turbine generator system fault prediction 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 prediction method is realized.
According to the electronic equipment provided by the embodiment of the invention, a processor executes a computer program stored on a memory, the operating parameters of each subsystem of the wind turbine generator are collected, the target operating parameters in the operating parameters are determined, one of the target operating parameters is sequentially used as the output parameter of a primary model, other parameters except one of the target operating parameters are used as the input parameters of the primary model, the predicted value of the output parameter of the primary model is obtained according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model, and the fault prediction result is obtained according to the predicted value of the output parameter of each primary model and a secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
In order to implement the above embodiments, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the above-mentioned method for predicting a fault of a wind turbine generator.
The computer-readable storage medium of the embodiment of the invention stores a computer program and is executed by a processor, acquires operating parameters of each subsystem of the wind turbine generator, determines target operating parameters in the operating parameters, sequentially takes one of the target operating parameters as an output parameter of a primary model, takes other parameters except one of the target operating parameters as input parameters of the primary model, obtains a predicted value of the output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model, and obtains a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model. According to the embodiment of the invention, the target operation parameters of each subsystem of the wind turbine generator are predicted based on the primary model, the secondary extraction of the fault characteristics of each subsystem is realized, the fault prediction is carried out on the wind turbine generator through the secondary model based on the predicted value of the primary model, and the accuracy and the universality of the fault prediction are enhanced.
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 invention, "a plurality" means two or more unless specifically defined otherwise.
In the present invention, 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 integrally formed; can be mechanically or electrically connected; either directly or indirectly through intervening media, either internally or in any other relationship. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the present invention, 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 invention. 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 invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (11)
1. A fault prediction method for a wind turbine generator is characterized by comprising the following steps:
collecting operating parameters of subsystems of a wind turbine generator;
determining a target operating parameter of the operating parameters;
one parameter in the target operation parameters is used as an output parameter of a primary model in sequence, and other parameters except the one parameter in the target operation parameters are used as input parameters of the primary model;
obtaining a predicted value of an output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model;
and obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model.
2. The fault prediction method of claim 1, wherein the determining a target one of the operating parameters comprises:
cleaning abnormal data in the operation parameters;
and determining the target operation parameters in the cleaned operation parameters according to the sample target operation parameters.
3. The failure prediction method of claim 2, further comprising:
collecting sample operation parameters of each subsystem of the wind turbine generator;
cleaning abnormal data in the sample operation parameters;
preliminarily screening the cleaned sample operation parameters according to expert experience;
carrying out variance screening on the preliminarily screened sample operation parameters, and eliminating the sample operation parameters with the variance smaller than a preset variance threshold value;
and performing correlation screening on the sample operation parameters subjected to the variance screening, and determining the sample operation parameters with the correlation number larger than a preset correlation coefficient threshold value as the sample target operation parameters.
4. The failure prediction method of claim 3, further comprising:
one parameter in the sample target operation parameters is used as a sample output parameter of a primary model, and other parameters except the one parameter in the sample target operation parameters are used as sample input parameters of the primary model;
carrying out feature extraction on the sample input parameters of the primary model to obtain the sample input parameter features of the primary model;
and taking the sample input parameter characteristics as input, taking the sample output parameters as output, and training a primary model to be trained to obtain the primary model.
5. The method according to claim 1, wherein obtaining the predicted value of the output parameter of the primary model according to the primary model corresponding to the input parameter of the primary model and the output parameter of the primary model comprises:
performing feature extraction on the input parameters of the primary model to obtain the input parameter features of the primary model;
and inputting the input parameter characteristics of the primary model into the primary model corresponding to the output parameters of the primary model to obtain the predicted value of the output parameters of the primary model.
6. The method according to claim 1, wherein obtaining the fault prediction result according to the predicted value of the output parameter of each primary model and the secondary model comprises:
calculating the difference value between the output parameter of the primary model and the predicted value of the output parameter of the primary model to obtain residual error characteristics;
generating residual error state characteristics according to the residual error characteristics and the time information;
and inputting the residual error characteristics, the residual error state characteristics and the working condition data into the secondary model to obtain the fault prediction result.
7. The failure prediction method of claim 1, further comprising:
calculating the difference value between the sample output parameters of the primary model and the predicted values of the sample output parameters of the primary model in a preset number of time periods before the sample fault occurs to obtain sample residual error characteristics;
generating a sample residual error state characteristic according to the sample residual error characteristic and the sample time information;
and taking the sample residual error characteristics, the sample residual error state characteristics and the sample working condition data as input, taking the sample fault as output, and training a secondary model to be trained to obtain the secondary model.
8. A failure prediction device for a wind turbine generator, comprising:
the acquisition module is used for acquiring the operating parameters of each subsystem of the wind turbine generator;
the first determination module is used for determining a target operation parameter in the operation parameters;
the second determining module is used for sequentially taking one parameter in the target operation parameters as an output parameter of the primary model and taking other parameters except the one parameter in the target operation parameters as input parameters of the primary model;
the first prediction module is used for obtaining a predicted value of an output parameter of the primary model according to the input parameter of the primary model and the primary model corresponding to the output parameter of the primary model;
and the second prediction module is used for obtaining a fault prediction result according to the predicted value of the output parameter of each primary model and the secondary models.
9. A wind turbine, comprising: the failure prediction apparatus of a wind turbine according to claim 8.
10. An electronic device, comprising: memory, processor and computer program stored on the memory and executable on the processor, which when executing the program, implements the method for fault prediction of a wind turbine as claimed in any of claims 1 to 7.
11. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out a method for fault prediction of a wind turbine according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111243323.XA CN114037127A (en) | 2021-10-25 | 2021-10-25 | Fault prediction method and device for wind turbine generator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111243323.XA CN114037127A (en) | 2021-10-25 | 2021-10-25 | Fault prediction method and device for wind turbine generator |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114037127A true CN114037127A (en) | 2022-02-11 |
Family
ID=80135300
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111243323.XA Pending CN114037127A (en) | 2021-10-25 | 2021-10-25 | Fault prediction method and device for wind turbine generator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114037127A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719002A (en) * | 2016-01-18 | 2016-06-29 | 重庆大学 | Wind turbine generator state parameter abnormity identification method based on combination prediction |
| CN105760617A (en) * | 2016-03-07 | 2016-07-13 | 华北电力大学（保定） | Calculation method applied to multi-parameter fault prediction and judgment indexes of wind generating set |
| CN105787561A (en) * | 2016-03-22 | 2016-07-20 | 新疆金风科技股份有限公司 | Recurrent neural network model construction method and gearbox fault detection method and device |
| CN107977508A (en) * | 2017-11-29 | 2018-05-01 | 北京优利康达科技股份有限公司 | A kind of dynamo bearing failure prediction method |
| CN108680358A (en) * | 2018-03-23 | 2018-10-19 | 河海大学 | A kind of Wind turbines failure prediction method based on bearing temperature model |
| CN110134919A (en) * | 2019-04-30 | 2019-08-16 | 华北电力大学 | Wind turbines abnormal data cleaning method |
| CN110766175A (en) * | 2019-10-25 | 2020-02-07 | 长沙理工大学 | Pitch system fault detection method and device based on optimal interval distribution machine |
| WO2021107866A1 (en) * | 2019-11-25 | 2021-06-03 | Envision Digital International Pte. Ltd. | Method and apparatus for detecting fault, method and apparatus for training model, and device and storage medium |
-
2021
- 2021-10-25 CN CN202111243323.XA patent/CN114037127A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719002A (en) * | 2016-01-18 | 2016-06-29 | 重庆大学 | Wind turbine generator state parameter abnormity identification method based on combination prediction |
| CN105760617A (en) * | 2016-03-07 | 2016-07-13 | 华北电力大学（保定） | Calculation method applied to multi-parameter fault prediction and judgment indexes of wind generating set |
| CN105787561A (en) * | 2016-03-22 | 2016-07-20 | 新疆金风科技股份有限公司 | Recurrent neural network model construction method and gearbox fault detection method and device |
| CN107977508A (en) * | 2017-11-29 | 2018-05-01 | 北京优利康达科技股份有限公司 | A kind of dynamo bearing failure prediction method |
| CN108680358A (en) * | 2018-03-23 | 2018-10-19 | 河海大学 | A kind of Wind turbines failure prediction method based on bearing temperature model |
| CN110134919A (en) * | 2019-04-30 | 2019-08-16 | 华北电力大学 | Wind turbines abnormal data cleaning method |
| CN110766175A (en) * | 2019-10-25 | 2020-02-07 | 长沙理工大学 | Pitch system fault detection method and device based on optimal interval distribution machine |
| WO2021107866A1 (en) * | 2019-11-25 | 2021-06-03 | Envision Digital International Pte. Ltd. | Method and apparatus for detecting fault, method and apparatus for training model, and device and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| ""海上风电机组智能故障预警***研究"", 《南方能源建设》, vol. 5, no. 2, 25 June 2018 (2018-06-25), pages 133 - 137 * |
| 张静等: ""基于数据的风电机组发电机健康状况评估"", 《信息与控制》, vol. 47, no. 6, 15 December 2018 (2018-12-15), pages 694 - 701 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111597682B (en) | Method for predicting remaining life of bearing of gearbox of wind turbine | |
| Garga et al. | Hybrid reasoning for prognostic learning in CBM systems | |
| Zaher et al. | A multi-agent fault detection system for wind turbine defect recognition and diagnosis | |
| CN114036656B (en) | Fault diagnosis method and device for wind turbine generator gearbox | |
| CN112418277A (en) | Method, system, medium, and apparatus for predicting remaining life of rotating machine component | |
| CN108287317B (en) | Battery power prediction model generation method and system and power prediction method and system | |
| Butler et al. | A feasibility study into prognostics for the main bearing of a wind turbine | |
| Zhan et al. | Health assessment methods for wind turbines based on power prediction and mahalanobis distance | |
| CN108304350A (en) | Wind turbine index prediction based on large data sets neighbour's strategy and fault early warning method | |
| Shi et al. | Study of wind turbine fault diagnosis and early warning based on SCADA data | |
| CN114004991A (en) | Fault identification method and device for wind turbine generator | |
| CN114037127A (en) | Fault prediction method and device for wind turbine generator | |
| CN113505448A (en) | Wind turbine generator dynamic reliability evaluation method based on improved Bayesian network | |
| WO2022102036A1 (en) | Machining diagnosis device, learning device, inference device, machining diagnosis method and program | |
| Wang et al. | Remaining useful life prediction based on deep residual attention network | |
| CN114201825A (en) | Method and system for evaluating equipment performance degradation state based on combination characteristics | |
| CN114442543A (en) | Computer monitoring method suitable for early warning of hydropower station fault | |
| Joshi et al. | Soft computing methods and its applications in condition monitoring of dgs—a review | |
| CN117273702A (en) | Gear box maintenance scheme prediction and recommendation method | |
| CN112699598A (en) | Intelligent diagnosis method and device for abnormal oil temperature of gear box | |
| Bo et al. | Multi-fault Diagnosis of Wind Turbine Based on Imprecise Probability | |
| Che et al. | Battery Lifetime Prediction and Degradation Reconstruction based on Probabilistic Convolutional Neural Network | |
| CN116429240A (en) | Rotating equipment vibration signal prediction system based on deep Boltzmann machine | |
| CN114065842A (en) | Wind turbine generator fault self-diagnosis method and system | |
| CN113447813A (en) | Fault diagnosis method and equipment for offshore wind generating set |
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 |