CN111709190A - Wind turbine generator operation data image identification method and device - Google Patents
Wind turbine generator operation data image identification method and device Download PDFInfo
- Publication number
- CN111709190A CN111709190A CN202010589896.7A CN202010589896A CN111709190A CN 111709190 A CN111709190 A CN 111709190A CN 202010589896 A CN202010589896 A CN 202010589896A CN 111709190 A CN111709190 A CN 111709190A
- Authority
- CN
- China
- Prior art keywords
- wind turbine
- turbine generator
- operation data
- generator operation
- neural network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- 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
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F03—MACHINES OR ENGINES FOR LIQUIDS; WIND, SPRING, OR WEIGHT MOTORS; PRODUCING MECHANICAL POWER OR A REACTIVE PROPULSIVE THRUST, NOT OTHERWISE PROVIDED FOR
- F03D—WIND MOTORS
- F03D17/00—Monitoring or testing of wind motors, e.g. diagnostics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
-
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/06—Wind turbines or wind farms
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Medical Informatics (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- Wind Motors (AREA)
Abstract
The invention provides a wind turbine generator operation data image identification method and device. The method comprises the following steps: drawing a trend graph of the operating data of the wind turbine generator; constructing a convolutional neural network for predicting the trend; training a convolutional neural network by utilizing the marked wind turbine generator operation data trend graph; and predicting data trends by using the trained convolutional neural network. The wind turbine generator operation data image recognition method and device provided by the invention improve the efficiency of data analysis in fault diagnosis and accurately diagnose the fault of the wind turbine generator.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a wind turbine generator operation data image identification method and device.
Background
The wind power plant relies on an SCADA (supervisory Control And Data acquisition) system to acquire information And monitor a wind generating set in real time for a long time, but with the development of the industry, only depending on Data monitoring, understanding of the 'surface' condition of the wind generating set is obviously insufficient, And the wind generating set needs to be deeply understood. Therefore, in recent years, in systems such as the series SCADA systems, expert fault diagnosis functions are added to perform application of manually extracted data analysis. The expert fault diagnosis function mostly depends on the experience of wind power enterprise experts on the wind power generator, which is accumulated for many years, and the wind power generator operation data are accurately analyzed. With the rapid development of the wind power industry in recent years, wind generating sets are spread over the river and mountain of the great Xinjiang, and the number of the wind generating sets is continuously increased. With the rapid development of the internet and information technology, wind turbine generator fault diagnosis experts cannot meet the fault data analysis of a plurality of wind turbine generators, and a digital and intelligent means needs to be innovated and applied to data analysis urgently.
Disclosure of Invention
The invention aims to provide a wind turbine generator operation data image recognition method and device, which improve the efficiency of data analysis in fault diagnosis and accurately diagnose the fault of the wind turbine generator.
In order to solve the technical problem, the invention provides a wind turbine generator operation data image identification method, which comprises the following steps: drawing a trend graph of the operating data of the wind turbine generator; constructing a convolutional neural network for predicting the trend; training a convolutional neural network by utilizing the marked wind turbine generator operation data trend graph; and predicting data trends by using the trained convolutional neural network.
In some embodiments, the method for drawing the trend graph of the wind turbine generator operating data comprises the following steps: importing historical operating data of the wind turbine generator; drawing a wind turbine generator operation data trend graph according to the imported historical operation data; and cleaning the drawn wind turbine generator operation data trend graph.
In some embodiments, the wind turbine components to which the historical operating data relates include: meteorological frame, gear box, generator, become oar, main bearing, converter, cabin, tower bottom.
In some embodiments, the constructed convolutional neural network for predicting trends comprises: five convolution layers, two full-connection layers and two dropout layers.
In some embodiments, five convolutional layers are connected in sequence, the fifth convolutional layer connects to the first fully-connected layer, the first fully-connected layer connects to the first dropout layer, the first dropout layer connects to the second fully-connected layer, and the second fully-connected layer connects to the second dropout layer.
In some embodiments, training the convolutional neural network by using the labeled wind turbine generator operation data trend graph comprises: loading a wind turbine generator operation data trend graph and labeled data; sequentially completing a first layer of convolution, a second layer of convolution, a third layer of convolution, a fourth layer of convolution, a fifth layer of convolution, a first full connection and a second full connection; adjusting network parameters to optimize the convolutional neural network; the trained convolutional neural network is saved.
In some embodiments, predicting data trends using a trained convolutional neural network comprises: loading a test set trend graph; carrying out batch normalization processing on the test set trend graph; loading the trained convolutional neural network; and outputting a trend prediction result.
In addition, the invention also provides a wind turbine generator operation data image recognition device, which comprises: one or more processors; the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the wind turbine generator operation data image identification method.
After adopting such design, the invention has at least the following advantages:
the method comprises the steps of modeling by adopting a Convolutional Neural Network (CNN) algorithm, importing expert base knowledge for judgment by combining a multivariate multidimensional array trend graph data set generated in the analysis of the operation data of the wind turbine generator, training a model by adopting a machine learning method, finally realizing batch identification of the operation state of the wind turbine generator, improving generalization of expert diagnosis and improving efficiency.
Drawings
The foregoing is only an overview of the technical solutions of the present invention, and in order to make the technical solutions of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and the detailed description.
FIG. 1 is a flowchart of an image recognition method for wind turbine operating data according to an embodiment of the present invention;
FIG. 2 is a normal data image provided by an embodiment of the present invention;
FIG. 3 is a data image of an anomaly provided by an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of a convolutional neural network structural model provided in an embodiment of the present invention;
fig. 5 is a structural diagram of an image recognition apparatus for wind turbine operating data according to an embodiment of the present invention.
Detailed Description
The preferred embodiments of the present invention will be described in conjunction with the accompanying drawings, and it will be understood that they are described herein for the purpose of illustration and explanation and not limitation.
Fig. 1 shows a flow chart of a wind turbine generator operation data image recognition method. Referring to fig. 1, the wind turbine generator operation data image recognition method includes:
and S11, drawing a wind turbine generator operation data trend graph.
S12, constructing a convolution neural network for predicting the trend.
And S13, training the convolutional neural network by using the marked wind turbine generator operation data trend graph.
And S14, predicting data trend by using the trained convolutional neural network.
Firstly, drawing and processing a wind turbine generator operation data graph data set.
Exporting historical operating data of the wind turbine generator, and classifying the data into categories according to wind turbine generator components, such as: meteorological frames (wind speed, wind direction), gear boxes (bearing temperature, oil pressure, etc.), generators (bearing temperature, cooling temperature, etc.), pitch (pitch blade position, pitch shaft temperature, driver temperature, etc.), main bearings (bearing temperature), frequency converters (IGBT temperature, stator temperature, rotor temperature, power, etc.), nacelles (nacelle temperature, etc.), tower bottoms (tower bottom temperature, etc.), and the like. And extracting data in the same time period and drawing a trend graph.
The size of the trend graph picture (carrying out matrix transformation of the image), the proportion of the labels and the verification set are set well and stored.
The trend plots relate to a large number of wind turbine components, but are of the same type that can be used for analysis, and therefore the entire process is explained below using only the meteorological frame-wind speed trend plots (this time using the wind speed 1-wind speed 2 trend plots). The normal and abnormal data images are shown in fig. 2 and 3, for example, according to the expert database judgment.
Secondly, a convolutional neural network structure model is established.
A 9-layer convolutional neural network model is built, and the structure sequentially comprises the following steps: the first convolution layer → the second convolution layer → the third convolution layer → the fourth convolution layer → the fifth convolution layer → the first fully connected layer → the first dropout layer → the second fully connected layer → the second dropout layer, and the network structure model is shown in fig. 4.
Third, a convolutional neural network is trained.
1. Loading training image data and label data;
2. starting a first convolution 401, a second convolution 402, a third convolution 403, a fourth convolution 404, a fifth convolution 405, a first full connection 406, a first dropout407, a second full connection 408 and a second dropout409 to finish the convolution process;
3. adjusting parameters to optimize the model;
4. and saving the model.
The following is a training process to train 444 anemometry image data, with the training step set to 4, and the model data saved.
Fourth, model prediction is performed.
1. Loading a test set trend graph;
2. carrying out batch normalization processing on the image set;
3. loading the trained model;
4. and outputting a prediction result.
And randomly selecting two wind speed trend graph images for testing, and obtaining a result after the above processing, wherein the result is consistent with the normal judgment.
With the increase of training data, model parameters are continuously adjusted, and the test accuracy is continuously improved.
Fig. 5 shows the structure of the wind turbine generator operation data image recognition apparatus. Referring to fig. 5, for example, the wind turbine operation data image recognition apparatus 500 may be used to serve as a fault diagnosis host in a wind turbine system. As described herein, the wind turbine operating data image recognition apparatus 500 may be used to implement a diagnostic function for faults in a wind turbine system. The wind turbine operation data image recognition apparatus 500 may be implemented in a single node, or the functions of the wind turbine operation data image recognition apparatus 500 may be implemented in a plurality of nodes in a network. Those skilled in the art will appreciate that the term wind turbine operation data image recognition means includes a broad sense of device, and the wind turbine operation data image recognition means 500 shown in fig. 5 is only one example thereof. The wind turbine operation data image recognition device 500 is included for clarity and is not intended to limit the application of the present invention to a particular wind turbine operation data image recognition device embodiment or to a certain class of wind turbine operation data image recognition device embodiments. At least some of the features/methods described herein may be implemented in a network device or component, such as the wind turbine generator operational data image recognition device 500. For example, the features/methods of the present invention may be implemented in hardware, firmware, and/or software running installed on hardware. The wind turbine operation data image recognition apparatus 500 may be any device that processes, stores and/or forwards data frames through a network, such as a server, a client, a data source, and the like. As shown in FIG. 5, the wind turbine operation data image recognition device 500 may include a transceiver (Tx/Rx)510, which may be a transmitter, a receiver, or a combination thereof. Tx/Rx510 may be coupled to a plurality of ports 550 (e.g., an uplink interface and/or a downlink interface) for transmitting and/or receiving frames from other nodes. Processor 530 may be coupled to Tx/Rx510 to process frames and/or determine to which nodes to send frames. Processor 530 may include one or more multi-core processors and/or memory devices 532, which may serve as data stores, buffers, and the like. Processor 530 may be implemented as a general-purpose processor, or may be part of one or more Application Specific Integrated Circuits (ASICs) and/or Digital Signal Processors (DSPs).
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the present invention in any way, and it will be apparent to those skilled in the art that the above description of the present invention can be applied to various modifications, equivalent variations or modifications without departing from the spirit and scope of the present invention.
Claims (8)
1. A wind turbine generator operation data image recognition method is characterized by comprising the following steps:
drawing a trend graph of the operating data of the wind turbine generator;
constructing a convolutional neural network for predicting the trend;
training a convolutional neural network by utilizing the marked wind turbine generator operation data trend graph;
and predicting data trends by using the trained convolutional neural network.
2. The wind turbine generator operation data image recognition method according to claim 1, wherein drawing a wind turbine generator operation data trend graph comprises:
importing historical operating data of the wind turbine generator;
drawing a wind turbine generator operation data trend graph according to the imported historical operation data;
and cleaning the drawn wind turbine generator operation data trend graph.
3. The image recognition method for the wind turbine generator operation data according to claim 2, wherein the wind turbine generator components related to the historical operation data comprise: meteorological frame, gear box, generator, become oar, main bearing, converter, cabin, tower bottom.
4. The wind turbine generator operation data image identification method according to claim 1, wherein the constructed convolutional neural network for predicting the trend comprises: five convolution layers, two full-connection layers and two dropout layers.
5. The wind turbine generator operation data image identification method according to claim 2, wherein five convolution layers are connected in sequence, the fifth convolution layer is connected with a first full connection layer, the first full connection layer is connected with a first dropout layer, the first dropout layer is connected with a second full connection layer, and the second full connection layer is connected with a second dropout layer.
6. The wind turbine generator operation data image recognition method according to claim 5, wherein training the convolutional neural network by using the labeled wind turbine generator operation data trend graph comprises:
loading a wind turbine generator operation data trend graph and labeled data;
sequentially completing a first layer of convolution, a second layer of convolution, a third layer of convolution, a fourth layer of convolution, a fifth layer of convolution, a first full connection and a second full connection;
adjusting network parameters to optimize the convolutional neural network;
the trained convolutional neural network is saved.
7. The wind turbine generator operation data image recognition method according to claim 1, wherein predicting data trends by using the trained convolutional neural network comprises:
loading a test set trend graph;
carrying out batch normalization processing on the test set trend graph;
loading the trained convolutional neural network;
and outputting a trend prediction result.
8. The utility model provides a wind turbine generator system operation data image recognition device which characterized in that includes:
one or more processors;
a storage device for storing one or more programs,
when executed by the one or more processors, the one or more programs cause the one or more processors to implement the wind turbine operation data image recognition method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010589896.7A CN111709190A (en) | 2020-06-24 | 2020-06-24 | Wind turbine generator operation data image identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010589896.7A CN111709190A (en) | 2020-06-24 | 2020-06-24 | Wind turbine generator operation data image identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111709190A true CN111709190A (en) | 2020-09-25 |
Family
ID=72542775
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010589896.7A Pending CN111709190A (en) | 2020-06-24 | 2020-06-24 | Wind turbine generator operation data image identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111709190A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112365399A (en) * | 2020-10-09 | 2021-02-12 | 北京星闪世图科技有限公司 | Fan blade image panoramic stitching method and system based on deep learning |
| CN112712126A (en) * | 2021-01-05 | 2021-04-27 | 南京大学 | Picture identification method |
| CN112832999A (en) * | 2021-01-08 | 2021-05-25 | 中国石油大学(北京) | Electric pump well working condition diagnosis system and method based on multi-sensor data fusion |
| CN113326879A (en) * | 2021-05-31 | 2021-08-31 | 深圳前海微众银行股份有限公司 | Service data monitoring method and device |
| CN115977874A (en) * | 2023-01-09 | 2023-04-18 | 中电投新疆能源化工集团木垒新能源有限公司 | Wind turbine generator yaw self-adaptive calibration method and system based on laser wind finding radar |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108830316A (en) * | 2018-06-05 | 2018-11-16 | 重庆大学 | The end-to-end fault diagnosis of wind electric converter based on convolutional neural networks |
| CN108896296A (en) * | 2018-04-18 | 2018-11-27 | 北京信息科技大学 | A kind of wind turbine gearbox method for diagnosing faults based on convolutional neural networks |
| CN109214356A (en) * | 2018-09-29 | 2019-01-15 | 南京东振测控技术有限公司 | A kind of fan transmission system intelligent fault diagnosis method based on DCNN model |
| US20190081476A1 (en) * | 2017-09-12 | 2019-03-14 | Sas Institute Inc. | Electric power grid supply and load prediction |
| CN110348513A (en) * | 2019-07-10 | 2019-10-18 | 北京华电天仁电力控制技术有限公司 | A kind of Wind turbines failure prediction method based on deep learning |
-
2020
- 2020-06-24 CN CN202010589896.7A patent/CN111709190A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190081476A1 (en) * | 2017-09-12 | 2019-03-14 | Sas Institute Inc. | Electric power grid supply and load prediction |
| CN108896296A (en) * | 2018-04-18 | 2018-11-27 | 北京信息科技大学 | A kind of wind turbine gearbox method for diagnosing faults based on convolutional neural networks |
| CN108830316A (en) * | 2018-06-05 | 2018-11-16 | 重庆大学 | The end-to-end fault diagnosis of wind electric converter based on convolutional neural networks |
| CN109214356A (en) * | 2018-09-29 | 2019-01-15 | 南京东振测控技术有限公司 | A kind of fan transmission system intelligent fault diagnosis method based on DCNN model |
| CN110348513A (en) * | 2019-07-10 | 2019-10-18 | 北京华电天仁电力控制技术有限公司 | A kind of Wind turbines failure prediction method based on deep learning |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112365399A (en) * | 2020-10-09 | 2021-02-12 | 北京星闪世图科技有限公司 | Fan blade image panoramic stitching method and system based on deep learning |
| CN112365399B (en) * | 2020-10-09 | 2024-05-03 | 江苏星闪世图科技(集团)有限公司 | Deep learning-based panoramic stitching method and system for fan blade images |
| CN112712126A (en) * | 2021-01-05 | 2021-04-27 | 南京大学 | Picture identification method |
| CN112712126B (en) * | 2021-01-05 | 2024-03-19 | 南京大学 | Picture identification method |
| CN112832999A (en) * | 2021-01-08 | 2021-05-25 | 中国石油大学(北京) | Electric pump well working condition diagnosis system and method based on multi-sensor data fusion |
| CN112832999B (en) * | 2021-01-08 | 2022-03-18 | 中国石油大学(北京) | Electric pump well working condition diagnosis system and method based on multi-sensor data fusion |
| CN113326879A (en) * | 2021-05-31 | 2021-08-31 | 深圳前海微众银行股份有限公司 | Service data monitoring method and device |
| CN115977874A (en) * | 2023-01-09 | 2023-04-18 | 中电投新疆能源化工集团木垒新能源有限公司 | Wind turbine generator yaw self-adaptive calibration method and system based on laser wind finding radar |
| CN115977874B (en) * | 2023-01-09 | 2024-03-19 | 中电投新疆能源化工集团木垒新能源有限公司 | Wind turbine generator yaw self-adaptive calibration method and system based on laser wind-finding radar |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111709190A (en) | Wind turbine generator operation data image identification method and device | |
| Guo et al. | Intelligent fault diagnosis method based on full 1-D convolutional generative adversarial network | |
| Liu et al. | A small-sample wind turbine fault detection method with synthetic fault data using generative adversarial nets | |
| Xu et al. | Online fault diagnosis method based on transfer convolutional neural networks | |
| Li et al. | Self-attention ConvLSTM and its application in RUL prediction of rolling bearings | |
| CN111538947B (en) | Method for constructing wind power generator bearing fault classification model | |
| CN112418277A (en) | Method, system, medium, and apparatus for predicting remaining life of rotating machine component | |
| Cao et al. | Fault diagnosis of wind turbine gearbox based on deep bi-directional long short-term memory under time-varying non-stationary operating conditions | |
| CN110008898A (en) | Industrial equipment data edges processing method based on symbol and convolutional neural networks | |
| CN113505664B (en) | Fault diagnosis method for planetary gear box of wind turbine generator | |
| Afrasiabi et al. | Wind turbine fault diagnosis with generative-temporal convolutional neural network | |
| CN111878322B (en) | Wind power generator device | |
| CN111198098A (en) | Wind power generator bearing fault prediction method based on artificial neural network | |
| CN114004512A (en) | Multi-unit operation state outlier analysis method and system based on density clustering | |
| CN113240022A (en) | Wind power gear box fault detection method of multi-scale single-classification convolutional network | |
| Marugán et al. | Multivariable analysis for advanced analytics of wind turbine management | |
| CN113758709A (en) | Rolling bearing fault diagnosis method and system combining edge calculation and deep learning | |
| CN111639852B (en) | Real-time evaluation method and system for vibration state of hydroelectric generating set based on wavelet singular value | |
| CN113673442A (en) | Variable working condition fault detection method based on semi-supervised single classification network | |
| CN112729825A (en) | Method for constructing bearing fault diagnosis model based on convolution cyclic neural network | |
| CN111539381A (en) | Construction method of wind turbine bearing fault classification diagnosis model | |
| Cao et al. | Remaining useful life prediction of wind turbine generator bearing based on EMD with an indicator | |
| CN115726935A (en) | Wind turbine generator abnormal state detection system and method based on artificial intelligence | |
| Yao et al. | Remaining useful life estimation by empirical mode decomposition and ensemble deep convolution neural networks | |
| Jiang et al. | An orbit-based encoder–forecaster deep learning method for condition monitoring of large turbomachines |
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 |