CN115754714A - Wind turbine generator fault positioning method, electronic device and storage medium - Google Patents
Wind turbine generator fault positioning method, electronic device and storage medium Download PDFInfo
- Publication number
- CN115754714A CN115754714A CN202211297008.XA CN202211297008A CN115754714A CN 115754714 A CN115754714 A CN 115754714A CN 202211297008 A CN202211297008 A CN 202211297008A CN 115754714 A CN115754714 A CN 115754714A
- Authority
- CN
- China
- Prior art keywords
- generator
- wind turbine
- fault
- turbine generator
- reconstruction error
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/70—Wind energy
- Y02E10/72—Wind turbines with rotation axis in wind direction
Landscapes
- Wind Motors (AREA)
Abstract
The invention relates to the technical field of wind turbine generator fault location, and particularly provides a wind turbine generator fault location method, electronic equipment and a storage medium, aiming at solving the technical problem that in the prior art, most workers locate faults, so that the accuracy of the fault location method is low. For the purpose, the method for positioning the generator fault of the wind turbine generator comprises the following steps: acquiring input variables related to a generator of a wind turbine generator; inputting the measured value of the input variable into the trained generator state monitoring model to obtain a reconstructed value of the input variable; determining a reconstruction error based on the reconstructed value and the measured value of the input variable; and carrying out fault positioning on the generator of the wind turbine generator based on the reconstruction error. Therefore, the accuracy of fault location of the generator of the wind turbine generator is improved, and the stability of the generator of the wind turbine generator is ensured.
Description
Technical Field
The invention relates to the technical field of wind turbine generator fault location, and particularly provides a wind turbine generator fault location method, electronic equipment and a storage medium.
Background
Wind turbine generators are mostly installed in remote areas such as high mountains, deserts and the sea, the operation environment is severe, the stress is complex and changeable, and when the wind turbine generators are operated in a variable-speed variable-load working condition for a long time, key parts such as generators are prone to failure, and shutdown is caused. The generator is used as one of indispensable key parts of a transmission chain system of the wind turbine generator, and the performance of the generator directly influences the performance and reliability of the transmission chain and even the whole machine. Therefore, the generator is timely and effectively positioned to ensure the safe, reliable and efficient operation of the unit, and the method is an effective way for reducing the failure outage time, reducing the operation and maintenance cost of the unit and prolonging the service life of the unit. However, in the prior art, most of the workers locate the fault, so that the accuracy of the fault locating method is low, and the actual requirement is difficult to meet.
Accordingly, there is a need in the art for a new wind turbine generator fault location solution to address the above-mentioned problems.
Disclosure of Invention
The present invention has been made to overcome the above-mentioned drawbacks, and aims to provide a solution or at least a partial solution to the above-mentioned technical problem. The invention provides a wind turbine generator fault positioning method, electronic equipment and a storage medium.
In a first aspect, the present invention provides a method for locating a fault of a generator of a wind turbine, the method comprising: acquiring input variables related to a generator of a wind turbine generator; inputting the measured value of the input variable into a trained generator state monitoring model to obtain a reconstructed value of the input variable; determining a reconstruction error based on the reconstructed value and the measured value of the input variable; and carrying out fault positioning on the generator of the wind turbine generator based on the reconstruction error.
In one embodiment, the input variables related to the wind turbine generator include wind speed, generator active power, generator reactive power, generator speed, actual torque, generator drive-end bearing temperature, generator non-drive-end bearing temperature, generator stator U-phase coil temperature, generator stator V-phase coil temperature, generator stator W-phase coil temperature, grid-side three-phase voltage, and grid-side three-phase current.
In one embodiment, the reconstruction error comprises a reconstruction error of the input variable and an overall reconstruction error of the generator state monitoring model; determining a reconstruction error based on the reconstructed value and the measured value of the input variable, comprising: determining a reconstruction error of the input variable based on an absolute value of a residual between the reconstructed value and an observed value; and determining the overall reconstruction error of the generator state monitoring model based on the reconstruction error of the input variable.
In one embodiment, fault locating the wind turbine generator based on the reconstruction error includes: calculating a correlation coefficient between the overall reconstruction error of the generator state monitoring model and the reconstruction error of the input variable; determining abnormal state parameters of the generator of the wind turbine generator based on the correlation coefficients; and positioning the fault position of the generator based on the abnormal state parameters of the generator of the wind turbine generator.
In one embodiment, determining the abnormal state parameter of the wind turbine generator based on the correlation coefficient comprises: and taking the input variable corresponding to the maximum value of the absolute value of the correlation coefficient as the abnormal state parameter of the generator of the wind turbine generator.
In one embodiment, the method further comprises: monitoring the generator fault of the wind turbine generator based on the reconstruction error and the alarm threshold value, and extracting a fault sample; and diagnosing the early fault of the wind turbine generator based on the fault sample.
In one embodiment, the extracting the fault sample includes: and under the condition that the reconstruction error exceeds an alarm threshold value, acquiring state data after alarm and taking the state data as a fault sample.
In one embodiment, diagnosing an early failure of a wind turbine generator based on the failure sample includes: and inputting the fault sample into a trained generator fault diagnosis model to obtain a final fault diagnosis result.
In a second aspect, an electronic device is provided, comprising a processor and a storage device, the storage device being adapted to store a plurality of program codes, the program codes being adapted to be loaded and run by the processor to perform the wind turbine generator fault localization method of any of the preceding claims.
In a third aspect, a computer readable storage medium is provided, having stored therein a plurality of program codes adapted to be loaded and run by a processor to perform the wind turbine generator fault location method of any of the preceding claims.
One or more technical schemes of the invention at least have one or more of the following beneficial effects:
the method for positioning the fault of the wind turbine generator comprises the steps of firstly obtaining input variables related to the wind turbine generator, then inputting measured values of the input variables into a trained generator state monitoring model to obtain reconstructed values of the input variables, secondly determining reconstruction errors based on the reconstructed values and the measured values of the input variables, and finally positioning the fault of the wind turbine generator based on the reconstruction errors. Therefore, the reconstruction error can be determined by using the reconstruction value output by the generator state monitoring model, the early fault position of the wind turbine generator is further positioned, and the accuracy of fault positioning is further improved by using the network model as a medium for determining the fault position.
Drawings
The disclosure of the present invention will become more readily understood with reference to the accompanying drawings. As is readily understood by those skilled in the art: these drawings are for illustrative purposes only and are not intended to be a limitation on the scope of the present disclosure. Moreover, in the drawings, like numerals are used to indicate like parts, and in which:
FIG. 1 is a flow chart illustrating the main steps of a wind turbine generator fault location method according to one embodiment of the present invention;
FIG. 2 is a schematic flow chart of a method for fault monitoring and fault diagnosis of a wind turbine generator according to one embodiment;
FIG. 3 is a wind speed-power plot of a wind turbine under normal operating conditions, according to one embodiment;
FIG. 4 is a test set reconstruction error and a reconstructed error frequency histogram of a generator condition monitoring model (LSTM-DAE) in one embodiment;
FIG. 5 is a schematic diagram of a generator early fault detection method according to one embodiment;
FIG. 6 is a schematic diagram of a confusion matrix of early generator fault diagnosis results based on a generator fault diagnosis model (XGboost model) in one embodiment;
FIG. 7 is a schematic diagram of the structure of an electronic device in one embodiment.
Detailed Description
Some embodiments of the invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and are not intended to limit the scope of the present invention.
In the description of the present invention, a "module" or "processor" may include hardware, software, or a combination of both. A module may comprise hardware circuitry, various suitable sensors, communication ports, memory, may comprise software components such as program code, and may be a combination of software and hardware. The processor may be a central processing unit, microprocessor, image processor, digital signal processor, or any other suitable processor. The processor has data and/or signal processing functionality. The processor may be implemented in software, hardware, or a combination thereof. Non-transitory computer readable storage media include any suitable medium that can store program code, such as magnetic disks, hard disks, optical disks, flash memory, read-only memory, random-access memory, and the like. The term "A and/or B" denotes all possible combinations of A and B, such as only A, only B or both A and B. The term "at least one of A or B" or "at least one of A and B" means similar to "A and/or B" and may include only A, only B, or both A and B. The singular forms "a", "an" and "the" may include plural forms as well.
At present, in the prior art, most workers are used for positioning the generator fault of the wind turbine generator, so that the accuracy of the fault positioning method is low, and the actual requirement is difficult to meet.
Therefore, the application provides a wind turbine generator fault positioning method, electronic equipment and a storage medium, firstly an input variable related to a wind turbine generator is obtained, then an actual measurement value of the input variable is input into a trained generator state monitoring model, a reconstruction value of the input variable is obtained, secondly a reconstruction error is determined based on the reconstruction value and the actual measurement value of the input variable, and finally fault positioning is carried out on the wind turbine generator based on the reconstruction error. Therefore, the reconstruction error can be determined by using the reconstruction value output by the generator state monitoring model, the early fault part of the wind turbine generator is further positioned, and the accuracy of the determined fault part is further improved by using the network model as a medium for determining the fault part.
Referring to fig. 1, fig. 1 is a schematic flow chart of main steps of a wind turbine generator fault location method according to an embodiment of the invention.
As shown in fig. 1, the method for locating a fault of a generator of a wind turbine generator in the embodiment of the present invention mainly includes the following steps S101 to S103.
Step S101: an input variable associated with a wind turbine generator is obtained.
In one embodiment, the input variables related to the wind turbine generator include wind speed, generator active power, generator reactive power, generator speed, actual torque, generator drive-end bearing temperature, generator non-drive-end bearing temperature, generator stator U-phase coil temperature, generator stator V-phase coil temperature, generator stator W-phase coil temperature, grid-side three-phase voltage, and grid-side three-phase current.
Step S102: and inputting the measured value of the input variable into the trained generator state monitoring model to obtain a reconstructed value of the input variable.
In one embodiment, an LSTM Seq2Seq network is introduced on the basis of a noise reduction self-encoder, and a generator state monitoring model based on a long-short-term memory noise reduction self-encoder (LSTM-DAE) is constructed, but is not limited to the method.
After the generator state monitoring model is constructed, the generator state monitoring model is further trained based on a training data set and a testing data set, and therefore the trained generator state monitoring model is obtained.
And inputting the measured value of the input variable into the trained generator state monitoring model to obtain the reconstructed value of the input variable.
Step S103: a reconstruction error is determined based on the reconstructed value and the measured value of the input variable.
In a specific embodiment, the reconstruction error comprises a reconstruction error of the input variable and an overall reconstruction error of the generator state monitoring model; determining a reconstruction error based on the reconstructed value and the measured value of the input variable, comprising: determining a reconstruction error of the input variable based on an absolute value of a residual between the reconstructed value and an actual value; and determining the overall reconstruction error of the generator state monitoring model based on the reconstruction error of the input variable.
Specifically, the actual measurement value of the input variable is input to the generator state monitoring model, and then the reconstructed value is obtained, and the absolute value of the residual between the reconstructed value and the actual measurement value is determined and used as the reconstruction error of the input variable. And further, taking the value obtained by square evolution of the reconstruction error of each input variable as the overall reconstruction error of the generator state monitoring model.
Step S104: and carrying out fault positioning on the wind turbine generator based on the reconstruction error.
In a specific embodiment, the fault locating the wind turbine generator based on the reconstruction error includes: calculating a correlation coefficient between the overall reconstruction error of the generator state monitoring model and the reconstruction error of the input variable; determining abnormal state parameters of the generator of the wind turbine generator based on the correlation coefficients; and positioning the fault position of the generator based on the abnormal state parameters of the generator of the wind turbine generator.
Specifically, a correlation coefficient between the overall reconstruction error of the generator state monitoring model and the reconstruction error of the input variable may be calculated using a spearman correlation coefficient (spearman correlation coefficient).
Illustratively, when the input variables are 16 input variables such as wind speed, generator active power, generator reactive power, generator rotating speed, actual torque, generator driving end bearing temperature, generator non-driving end bearing temperature, generator stator U-phase coil temperature, generator stator V-phase coil temperature, generator stator W-phase coil temperature, grid side three-phase voltage and grid side three-phase current, the integral reconstruction error sigma of the generator state monitoring model and the reconstruction error sigma of the 16 input variables are calculated by using the Spireman correlation coefficient 1 ,σ 2 …,σ 16 The correlation coefficient between the two is calculated according to the following formula:
wherein n is the sequence length, d i And (4) grade difference. It is generally accepted that if 1 ≧ r s | ≧ 0.8, strongly correlated; 0.8>|r s | ≧ 0.5, moderate correlation; 0.5>|r s | ≧ 0.3, low degree of correlation; | r s |<0.3, weak correlation.
In one embodiment, the determining the abnormal state parameter of the wind turbine generator based on the correlation coefficient comprises: and taking the input variable corresponding to the maximum value of the absolute value of the correlation coefficient as the abnormal state parameter of the generator of the wind turbine generator.
Specifically, the input variable corresponding to the maximum value of the absolute value of the correlation coefficient corresponds to the abnormal state parameter of the generator of the wind turbine generator.
And further positioning the fault position of the generator based on the abnormal state parameters of the generator of the wind turbine generator. In one embodiment, the corresponding position of the generator corresponding to the abnormal state parameter of the generator of the wind turbine generator may be used as a fault position, but is not limited thereto. In some complex scenes, a user or a worker can further position or troubleshoot the generator fault part according to the abnormal state parameters of the generator of the wind turbine generator, so that the positioning accuracy of the generator fault part can be further improved, and the safety and the reliability of the generator of the wind turbine generator are improved.
Based on the steps S101 to S104, first, an input variable related to the wind turbine generator is obtained, then, an actual measurement value of the input variable is input into the trained generator state monitoring model to obtain a reconstructed value of the input variable, then, a reconstruction error is determined based on the reconstructed value and the actual measurement value of the input variable, and finally, fault location is performed on the wind turbine generator based on the reconstruction error. Therefore, the reconstruction error can be determined by using the reconstruction value output by the generator state monitoring model, the early fault part of the wind turbine generator is further positioned, and the accuracy of the determined fault part is further improved by using the network model as a medium for determining the fault part.
In addition, the state monitoring and fault diagnosis research of the key components of the existing wind turbine mainly comprises the following two methods: (1) physical model-based methods; (2) Based on the data driving method (vibration, oil, sound signal and SCADA operation data). However, as wind turbines tend to become larger and more complex, physical modeling for establishing key components of the wind turbines is difficult to achieve. Hardware equipment for collection, storage, transmission and the like is additionally required to be added based on vibration signals, oil and sound signals, so that the large-area popularization is difficult. On the basis of SCADA operation data, a method combining a normal behavior model of a wind turbine generator and prediction residual analysis is established by methods such as machine learning and the like to realize fault detection and fault diagnosis of the wind turbine generator is a hot point of current research. The students at home and abroad have developed a lot of researches from the aspects of state monitoring, fault diagnosis and the like aiming at the fault early warning and diagnosis problems of key equipment such as a generator, a gear box and the like, but still have the following problems:
(1) The existing state monitoring method considers various state parameters of the equipment but ignores the time sequence of SCADA data, or considers the time sequence but a single state parameter can not completely and accurately reflect the running state of the equipment, and can not perform sufficient feature fusion on the two.
(2) The fault diagnosis method based on a single model cannot fully mine fault information, so that the fault diagnosis precision is low, the generalization capability is weak and the stability is poor when the model is migrated.
Therefore, the generator can be subjected to fault detection and diagnosis through the following embodiments, and the safety and the reliability of the generator of the wind turbine generator are further improved.
In a specific embodiment, the method further comprises: monitoring the generator fault of the wind turbine generator based on the reconstruction error and the alarm threshold value, and extracting a fault sample; and diagnosing the early fault of the wind turbine generator based on the fault sample.
In one embodiment, the overall reconstruction error of the generator state monitoring model may be taken as an example of the illustrated reconstruction error, but is not limited thereto.
Specifically, fault detection and fault diagnosis are carried out on the wind turbine generator through the reconstruction error and the alarm threshold. Specifically, the method comprises the steps of monitoring the generator fault of the wind turbine generator according to the reconstruction error and the alarm threshold, extracting a fault sample, and further diagnosing the early fault of the generator of the wind turbine generator based on the fault sample.
In a specific embodiment, the extracting the fault sample includes: and under the condition that the reconstruction error exceeds an alarm threshold, acquiring state data after alarm and taking the state data as a fault sample.
Specifically, in the case where it is detected that the reconstruction error exceeds the alarm threshold, the state data after the alarm is acquired and taken as a failure sample.
In one embodiment, a probability density function of the reconstruction error can be calculated by adopting a nuclear density estimation method, and an alarm threshold value of the generator fault is obtained according to the probability density function and a set confidence level. Since the method for further determining the alarm threshold by using the kernel density estimation method is a conventional technique, it is not described herein again.
In one embodiment, diagnosing an early failure of a wind turbine generator based on the failure sample includes: and inputting the fault sample into a trained generator fault diagnosis model to obtain a final fault diagnosis result.
In one embodiment, an XGboost integrated learning algorithm can be adopted to construct an XGboost multi-classification-based generator fault diagnosis model, and the XGboost multi-classification-based generator fault diagnosis model is trained, so that the trained generator fault diagnosis model is obtained. And further inputting the fault sample into the trained generator fault diagnosis model to obtain a final fault diagnosis result.
By the method, the performance of a wind turbine generator state monitoring model and the accuracy and reliability of fault detection can be remarkably improved, the problems of low early warning accuracy, insufficient early warning time and difficulty in obtaining fault samples in the existing research can be effectively solved, and the method has high reconstruction precision, excellent fault early warning capability and better diagnosis precision.
In an embodiment, a flow of a wind turbine generator fault detection method, a diagnosis method and a positioning method is shown in fig. 2, which will be described in detail through the following steps S201 to S209.
S201: based on the distribution characteristics of abnormal data in a wind speed-power scatter diagram of the wind turbine generator, abnormal data of the wind turbine generator, caused by working condition fluctuation, wind abandonment and power limitation, acquisition and transmission hardware equipment faults and the like, are removed by adopting a quartering bit point method. Firstly, directly removing obvious abnormal state data: 1. the wind turbine generator is in normal shutdown, idling or starting state data with the wind speed of less than 3m/s or more than 25 m/s; 2. and the wind turbine generator abnormal shutdown state data has the wind speed of more than 3m/s and the power of less than or equal to 0. And then, eliminating abnormal data caused by abandoned wind power limit, working condition fluctuation, acquisition and transmission hardware equipment failure and the like by adopting a quartile point method. Finally, the operation data of the wind turbine generator in the normal operation state (health state) is obtained through data cleaning, as shown in fig. 3.
S202: based on the obtained running data of the wind turbine generator in the normal running state (health state), 16 monitoring variables capable of reflecting the running state of the generator are selected according to the running principle of the wind turbine generator, wherein the monitoring variables are used as input variables of the model, and the monitoring variables can reflect the running state of the generator, such as wind speed, active power of the generator, reactive power of the generator, rotating speed of the generator, actual torque, bearing temperature of a driving end of the generator, bearing temperature of a non-driving end of the generator, U-phase coil temperature of a stator of the generator, V-phase coil temperature of a stator of the generator, W-phase coil temperature of the stator of the generator, network-side three-phase voltage, network-side three-phase current and the like.
S203: and obtaining a training set and a testing set of the generator state monitoring model through data cleaning and feature selection. And carrying out normalization processing on the data of the training set and the data of the test set so as to eliminate dimension influence between SCADA multi-source heterogeneous data of the wind turbine generator and reduce the difficulty of model training. The normalized formula is shown below:
s204: on the basis of the noise reduction self-encoder, an LSTM Seq2Seq network is introduced to construct a generator state monitoring model based on a long-short term memory noise reduction self-encoder (LSTM-DAE). Fig. 2 shows a training process of the proposed generator state monitoring model, which is trained and tested based on the training set and the testing set obtained in S203.
S205: based on the generator state monitoring model and the test set trained in step S204, a distribution rule of the reconstruction error under the normal operation condition (healthy state) of the generator is obtained. And calculating a probability density function of the reconstruction error of the test set by adopting a nuclear density estimation method, and solving a generator fault alarm threshold value according to the probability density function and a set confidence level. FIG. 4 shows reconstruction errors and a frequency histogram of the reconstruction errors of a generator state monitoring model (LSTM-DAE) test set.
S206: and (4) carrying out online state monitoring on the generator of the wind turbine generator based on the generator state monitoring model trained in the step (S204) and the fault alarm threshold calculated in the step (S205) so as to realize detection of early faults of the generator. And constructing a fault sample data set based on the detected generator fault for generator fault diagnosis of the next stage. The process of the online state monitoring of the proposed generator state monitoring model is shown in fig. 2. Fig. 5 shows an actual case based on a generator fault of a wind turbine generator, and a verification result of the proposed generator state monitoring model verifies that the proposed generator state monitoring model has early generator fault early warning capability, and can find a fault in advance and send an alarm.
S207: and constructing a fault diagnosis model based on the XGboost multi-classification generator by adopting an XGboost ensemble learning algorithm based on the extracted fault sample data set. The proposed generator fault diagnosis model training process is illustrated in fig. 2.
S208: and performing online fault diagnosis on the generator of the wind turbine generator based on the generator fault diagnosis model trained in the step S207 so as to realize the diagnosis of the early fault of the generator. Fig. 6 shows an actual case based on the generator fault of the wind turbine generator, and a verification result of the generator fault diagnosis model verifies that the generator fault diagnosis model has excellent fault diagnosis capability and has higher fault diagnosis accuracy and reliability.
S209: based on the generator state monitoring model (LSTM-DAE) of the long-short term memory noise reduction self-encoder (LSTM-DAE) trained in S204, model output (the reconstructed values of 16 input variables such as wind speed) of model input (16 input variables such as wind speed) is obtained, and further the overall reconstruction error sigma of the monitoring model (LSTM-DAE) and the reconstruction error sigma of each input variable can be calculated 1 ,σ 2 …,σ 16 . Calculating the overall reconstruction error sigma and each reconstruction component error sigma based on the spearman correlation coefficient 1 ,σ 2 …,σ 16 The input variable corresponding to the maximum value of the absolute value of the correlation coefficient and the absolute value of the correlation coefficient corresponds to the abnormal state parameter of the generator of the wind turbine generator, and the fault position of the generator can be further positioned. The Spireman correlation coefficient calculation formula is as follows:
wherein n is the sequence length, d i And (4) grade difference. It is generally accepted that if 1 ≧ r s | > 0.8, strong correlation; 0.8>|r s | > 0.5, moderately relevant; 0.5>|r s | ≧ 0.3, low degree of correlation; | r s |<0.3, weak correlation.
Abnormal operation state data such as outlier noise points, accumulation points and the like in the operation process of the wind turbine generator can be removed based on the quartile point method; based on the operation principle of the wind turbine generator, selecting a monitoring variable capable of reflecting the operation state of the generator as an input variable of a generator state monitoring model; constructing a normal behavior model of the generator based on the noise reduction self-encoder and the long-short term memory network, realizing the reconstruction of normal operation data of the generator and obtaining a reconstruction error of the generator in a healthy state; the method comprises the following steps of exploring the distribution rule of a reconstruction error in a healthy state of the generator by adopting a statistical method, calculating a probability density function of the reconstruction error based on a kernel density estimation method, and setting a fault alarm threshold value to realize the online detection of the early fault of the generator; constructing a fault sample data set based on the detected generator fault for the generator fault diagnosis of the next stage; and constructing a fault diagnosis model based on the XGboost multi-classification generator by adopting an XGboost ensemble learning algorithm based on the extracted fault sample data set. Based on the spearman correlation coefficient, the abnormal state parameters of the generator can be accurately judged, and the fault position of the generator can be further positioned. Therefore, the performance of the generator state monitoring model and the accuracy and reliability of fault detection can be remarkably improved, the problems of low early warning accuracy, insufficient early warning time and difficulty in obtaining fault samples in the existing research can be effectively solved, and the method has high reconstruction precision, excellent fault early warning capability and better diagnosis precision.
It should be noted that, although the foregoing embodiments describe each step in a specific sequence, those skilled in the art can understand that, in order to achieve the effect of the present invention, different steps do not have to be executed in such a sequence, and they may be executed simultaneously (in parallel) or in other sequences, and these changes are all within the scope of the present invention.
It will be understood by those skilled in the art that all or part of the flow of the method according to the above-described embodiment may be implemented by a computer program, which may be stored in a computer-readable storage medium and used to implement the steps of the above-described embodiments of the method when the computer program is executed by a processor. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable storage medium may include: any entity or device capable of carrying said computer program code, media, usb disk, removable hard disk, magnetic diskette, optical disk, computer memory, read-only memory, random access memory, electrical carrier wave signals, telecommunication signals, software distribution media, etc. It should be noted that the computer readable storage medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable storage media that does not include electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.
Furthermore, the invention also provides electronic equipment. In an embodiment of the electronic device according to the present invention, as shown in fig. 7, the electronic device includes a processor 71 and a storage device 72, the storage device may be configured to store a program for executing the wind turbine generator fault location method of the above-mentioned method embodiment, and the processor may be configured to execute a program in the storage device, the program including but not limited to a program for executing the wind turbine generator fault location method of the above-mentioned method embodiment. For convenience of explanation, only the parts related to the embodiments of the present invention are shown, and details of the specific techniques are not disclosed.
Further, the invention also provides a computer readable storage medium. In one computer-readable storage medium embodiment according to the present invention, the computer-readable storage medium may be configured to store a program for executing the wind turbine generator fault location method of the above-described method embodiment, and the program may be loaded and executed by a processor to implement the wind turbine generator fault location method described above. For convenience of explanation, only the parts related to the embodiments of the present invention are shown, and specific technical details are not disclosed. The computer readable storage medium may be a storage device formed by including various electronic devices, and optionally, the computer readable storage medium is a non-transitory computer readable storage medium in the embodiment of the present invention.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is apparent to those skilled in the art that the scope of the present invention is not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (10)
1. A method for positioning the fault of a generator of a wind turbine generator is characterized by comprising the following steps:
acquiring input variables related to a generator of a wind turbine generator;
inputting the measured value of the input variable into a trained generator state monitoring model to obtain a reconstructed value of the input variable;
determining a reconstruction error based on the reconstructed value and the measured value of the input variable;
and carrying out fault positioning on the generator of the wind turbine generator based on the reconstruction error.
2. The method according to claim 1, wherein the input variables related to the wind turbine generator comprise wind speed, generator active power, generator reactive power, generator speed, actual torque, generator drive-end bearing temperature, generator non-drive-end bearing temperature, generator stator U-phase coil temperature, generator stator V-phase coil temperature, generator stator W-phase coil temperature, grid-side three-phase voltage and grid-side three-phase current.
3. The wind turbine generator fault location method of claim 1, wherein the reconstruction error comprises a reconstruction error of the input variable and an overall reconstruction error of the generator state monitoring model; determining a reconstruction error based on the reconstructed value and the measured value of the input variable, comprising:
determining a reconstruction error of the input variable based on an absolute value of a residual between the reconstructed value and an observed value;
and determining the overall reconstruction error of the generator state monitoring model based on the reconstruction error of the input variable.
4. The wind turbine generator fault location method according to claim 3, wherein fault location of the wind turbine generator based on the reconstruction error includes:
calculating a correlation coefficient between the overall reconstruction error of the generator state monitoring model and the reconstruction error of the input variable;
determining abnormal state parameters of the generator of the wind turbine generator based on the correlation coefficients;
and positioning the fault position of the generator based on the abnormal state parameters of the generator of the wind turbine generator.
5. The wind turbine generator fault location method according to claim 4, wherein determining the abnormal state parameter of the wind turbine generator based on the correlation coefficient includes: and taking the input variable corresponding to the maximum value of the absolute value of the correlation coefficient as the abnormal state parameter of the generator of the wind turbine generator.
6. The wind turbine generator fault locating method according to claim 1, characterized in that the method further comprises:
monitoring the generator faults of the wind turbine generator based on the reconstruction errors and the alarm threshold value, and extracting fault samples;
and diagnosing the early fault of the wind turbine generator based on the fault sample.
7. The wind turbine generator fault location method of claim 6, wherein the extracting a fault sample comprises: and under the condition that the reconstruction error exceeds an alarm threshold, acquiring state data after alarm and taking the state data as the fault sample.
8. The wind turbine generator fault location method of claim 6, wherein diagnosing the early fault of the wind turbine generator based on the fault sample comprises: and inputting the fault sample into a trained generator fault diagnosis model to obtain a final fault diagnosis result.
9. An electronic device comprising a processor and a storage means adapted to store a plurality of program codes, characterized in that said program codes are adapted to be loaded and run by said processor to perform the wind turbine generator fault localization method according to any of claims 1 to 7.
10. A computer readable storage medium having a plurality of program codes stored therein, wherein the program codes are adapted to be loaded and run by a processor to perform the wind turbine generator fault localization method of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211297008.XA CN115754714A (en) | 2022-10-21 | 2022-10-21 | Wind turbine generator fault positioning method, electronic device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211297008.XA CN115754714A (en) | 2022-10-21 | 2022-10-21 | Wind turbine generator fault positioning method, electronic device and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115754714A true CN115754714A (en) | 2023-03-07 |
Family
ID=85352683
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211297008.XA Pending CN115754714A (en) | 2022-10-21 | 2022-10-21 | Wind turbine generator fault positioning method, electronic device and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115754714A (en) |
-
2022
- 2022-10-21 CN CN202211297008.XA patent/CN115754714A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Isham et al. | Variational mode decomposition: mode determination method for rotating machinery diagnosis | |
| CN115955161B (en) | Slip estimation method, device, equipment and medium for self-adaptive asynchronous induction motor | |
| Chen et al. | Hierarchical method for wind turbine prognosis using SCADA data | |
| CN111881594B (en) | Non-stationary signal state monitoring method and system for nuclear power equipment | |
| CN109725220B (en) | Detection method, system and device for transformer oil cooling loop | |
| CN113759876B (en) | Case-reasoning-based wind turbine generator fault diagnosis method and system | |
| EP3012643B1 (en) | Method and apparatus for identifying causes for cable overcurrent | |
| Sharp | Prognostic approaches using transient monitoring methods | |
| CN116466689B (en) | Fault diagnosis method and device | |
| CN115754714A (en) | Wind turbine generator fault positioning method, electronic device and storage medium | |
| CN117591924A (en) | Bearing fault detection method and device, storage medium and electronic equipment | |
| CN116226719A (en) | Bearing fault diagnosis method based on multidimensional steady-state vibration characteristics and related components | |
| CN111310959A (en) | Wind generating set damage prediction analysis method, device, equipment and storage medium | |
| CN115906942A (en) | Arc fault identification method, device, equipment and readable storage medium | |
| CN115561575A (en) | Method for distinguishing electrical abnormal state of offshore wind farm based on multi-dimensional matrix profile | |
| CN112906914B (en) | Rail transit IT equipment fault analysis method and device and electronic equipment | |
| CN114662977A (en) | Method and system for detecting abnormity of motion state of offshore drilling platform and electronic equipment | |
| CN112699598A (en) | Intelligent diagnosis method and device for abnormal oil temperature of gear box | |
| Zheng et al. | Wavelet packet decomposition and neural network based fault diagnosis for elevator excessive vibration | |
| CN114139440A (en) | Method and system for predicting residual service life of fan bearing and computer equipment | |
| CN113835029B (en) | Motor fault diagnosis evaluation standard construction method, system and computer storage medium | |
| CN113947248B (en) | Risk prediction method and related device for cable damp aging tripping | |
| CN114859226B (en) | Method and device for identifying turn-to-turn short circuit fault of rotor winding of offshore wind turbine generator system | |
| CN117035747B (en) | Multi-system fault diagnosis processing method, system, equipment and medium for machine room | |
| CN117627878A (en) | Fault diagnosis method, device, equipment and medium for land wind farm |
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 |