CN117828494A - Fan cable-releasing defect mode identification method, device and medium - Google Patents
Fan cable-releasing defect mode identification method, device and medium Download PDFInfo
- Publication number
- CN117828494A CN117828494A CN202311823721.8A CN202311823721A CN117828494A CN 117828494 A CN117828494 A CN 117828494A CN 202311823721 A CN202311823721 A CN 202311823721A CN 117828494 A CN117828494 A CN 117828494A
- Authority
- CN
- China
- Prior art keywords
- cable
- wind turbine
- untwisting
- wind
- turbine generator
- 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
- 230000007547 defect Effects 0.000 title claims abstract description 49
- 238000000034 method Methods 0.000 title claims abstract description 47
- 230000003578 releasing effect Effects 0.000 title claims description 47
- 238000003062 neural network model Methods 0.000 claims abstract description 37
- 230000001960 triggered effect Effects 0.000 claims abstract description 5
- 230000003068 static effect Effects 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 15
- 238000012549 training Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 10
- 230000009471 action Effects 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000002360 preparation method Methods 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012567 pattern recognition method Methods 0.000 claims description 3
- 230000002441 reversible effect Effects 0.000 claims description 3
- 238000003909 pattern recognition Methods 0.000 claims description 2
- 238000010248 power generation Methods 0.000 abstract description 20
- 230000006399 behavior Effects 0.000 description 13
- 230000008901 benefit Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000006872 improvement Effects 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011176 pooling Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- 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
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- Chemical & Material Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Biophysics (AREA)
- Sustainable Energy (AREA)
- Software Systems (AREA)
- Sustainable Development (AREA)
- Biomedical Technology (AREA)
- Health & Medical Sciences (AREA)
- Mechanical Engineering (AREA)
- Combustion & Propulsion (AREA)
- Wind Motors (AREA)
Abstract
The invention relates to a fan untwisting defect mode identification method, which comprises the following steps: acquiring sensing data of a discrete time point acquired by a wind turbine through a sensor, wherein the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic wind directionAnd the wind turbine generator set is orientedCalculating input data of a pre-trained neural network model according to the sensing data; inputting the input data into the pre-trained neural network model, and predicting a label corresponding to a defect mode of the input data. The method can identify the defect mode of the cable-disconnecting behavior triggered by the wind turbine based on the neural network model so as to improve the power generation efficiency of the wind turbine.
Description
Technical Field
The invention relates to the technical field of wind power generation, in particular to a fan untwisting defect mode identification method, a fan untwisting defect mode identification device and a computer readable storage medium.
Background
As the world demand for energy is increasing, environmental protection requirements are increasing. The development and utilization of energy resources are gradually turned to the wind energy field.
The yaw control system is arranged in the wind turbine, and is used for keeping the direction of the engine room parallel to the incoming wind direction in the power generation state of the wind turbine, so that the upper limit of wind energy which can be absorbed by the impeller of the wind turbine is increased, the side load caused by the incoming wind speed is reduced, and the purposes of improving the power of the wind turbine and guaranteeing safe operation are achieved. The yaw control algorithm mainly comprises a yaw algorithm based on a fuzzy controller, a yaw control algorithm based on a fixed time interval, a yaw control algorithm based on PID, and the like.
Under certain conditions, yaw can be continuously carried out towards a certain direction, so that cables connecting the nacelle and the tower bottom are twisted, and cable stretch-out can occur and cause the wind turbine to fall down in severe cases. Therefore, the control logic of the wind turbine generator system must be designed with a cable-releasing control logic, and it is easy to understand that when the cable-releasing control logic triggers the cable-releasing action of the wind turbine generator system, the cable-twisting angle of the cable is reversely reduced when the cable-twisting angle exceeds a threshold value in one direction, so that accidents in the operation process are avoided.
However, the inventors of the present application have found in research that some untwisting actions occur too frequently or not timely, for example, untwisting is performed in a period where wind conditions are suitable for power generation, which obviously reduces the efficiency of wind energy utilization. The cable-releasing behaviors can be considered to have certain defect modes, the defect modes of the cable-releasing behaviors are identified, improvement of cable-releasing control logic of the wind turbine generator is facilitated, and the power generation efficiency and economic benefit of the wind turbine generator are also facilitated to be improved.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a method and apparatus for identifying a defect mode of a wind turbine, and a computer readable storage medium, which can identify a defect mode of a wind turbine generator triggering cable-unwinding behavior based on a neural network model, so as to improve the power generation efficiency of the wind turbine generator.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
in a first aspect, the present application provides a method for identifying a fan untwisting defect mode, where the method includes:
acquiring sensing data of a discrete time point acquired by a wind turbine through a sensor, wherein the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
Calculating input data of a pre-trained neural network model according to the sensing data;
inputting the input data into the pre-trained neural network model, and predicting a label corresponding to a defect mode of the input data.
In one implementation manner of the present application, the calculating input data of the pre-trained neural network model according to the sensing data includes:
cable twisting angle according to discrete time pointsCalculating a torsion cable angle change value +.>
According to the wind turbine generator set orientation corresponding to each discrete time pointAnd the geographical wind direction->Calculating a wind deviation angle theta;
according to the wind speed v and the power P corresponding to each discrete time point act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->A torsion cable angle variation value +>And the wind deviation angle theta, and form 7-dimension input dataWhere t is the time sequence number of the discrete time point.
In one implementation of the present application, the set time interval is 10 minutes; the torsion cable angle change value of the set time intervalThe calculation formula of (2) is as follows:
in one implementation manner of the present application, the calculation formula of the wind deviation angle θ is:
wherein k is an integer, and the value of the integer satisfies θ in the range of [ -180 DEG, +180 DEG ].
In one implementation of the present application, the defect mode tag includes: 0. three classes 1 and 2, wherein:
class 0 labels, corresponding to normal untwisting behavior;
the class 1 labels are used for triggering cable disconnection abnormally in the starting process of the wind turbine generator;
and 2 types of labels correspond to static delay untwisting of the wind turbine generator.
In one implementation of the present application, the duration of the untwisting action is: and 60 minutes taking a cable release control logic trigger cable release action instruction of the wind turbine generator as a central time point.
In one implementation of the present application, the input data at 60 sets of discrete time points is input into the neural network model, and a label of a defect mode of a corresponding untwisting behavior is predicted
In one implementation manner of the present application, the abnormal triggering of the cable-disconnection during the starting process is defined as: in the starting process of the wind turbine, yaw is firstly carried out according to a first direction, then cable disconnection is triggered, and after the reverse cable disconnection is completed, secondary deviation is carried out according to the first direction to complete starting preparation;
the static delay untwisting is defined as: under the standby state, the wind turbine generator meets the static cable releasing condition, the cable releasing is not executed beyond the first predefined time, the wind condition meets the power generating condition, and the wind turbine generator does not enter the power generating state at the second predefined time.
In one implementation of the present application, the method further includes the step of training the neural network model;
a step of training the neural network model, comprising:
collecting sensing data of discrete time points of historical data, and packagingThe method comprises the following steps: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
Calculating a torsion angle change value of discrete time points of historical dataAnd a wind deviation angle θ;
according to the moment of triggering the cable-releasing action command every time in the historical data, correspondingly determining the discrete time points of 60 samples by taking the cable-releasing action command as a center;
labeling each sample of the determined discrete time points to finish the collection of sample data;
training of the neural network model is completed by using the collected sample data.
In a second aspect, the present application provides a fan untwisting defect mode identification device, the device comprising:
the sensing module is used for acquiring sensing data of a discrete time point acquired by the wind turbine generator through a sensor, and the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
The input data calculation module is used for calculating input data of a pre-trained neural network model according to the sensing data;
and the prediction module is used for inputting the input data into the pre-trained neural network model and predicting labels corresponding to defect modes of the input data.
In a third aspect, the present application provides a computer readable storage medium, where a computer program is stored, where the computer program controls, when running, a device where the computer readable storage medium is located to execute the fan untwisting defect mode identifying method according to the first aspect.
Due to the adoption of the technical scheme, the invention has the following advantages: according to the application scheme, the sensing data of the discrete time points of the wind turbine generator are acquired, calculation is performed based on the sensing data, the input data of the neural network model is obtained, and then the input data is input into the neural network model, so that the labels corresponding to the defect modes of the input data can be predicted, the defect modes of the cable releasing behavior of the wind turbine generator can be accurately identified, improvement of cable releasing control logic of the wind turbine generator is facilitated, and the power generation efficiency and economic benefit of the wind turbine generator are improved.
Drawings
FIG. 1 is a schematic flow diagram of the de-cabling control logic provided as a reference in an embodiment of the present application;
fig. 2 is a schematic diagram of a scenario corresponding to a defect mode of a cable-unwinding behavior in the embodiment of the present application;
FIG. 3 is a schematic view of a scenario of another defect mode of the untwisting behavior in an embodiment of the present application;
FIG. 4 is a flow chart of a fan untwisting defect pattern recognition method in an embodiment of the present application;
fig. 5 is a schematic architecture diagram of a neural network model.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, which are obtained by a person skilled in the art based on the described embodiments of the invention, fall within the scope of protection of the invention.
The angles involved in the embodiments of the present application will be described first. In the embodiment of the application, 0 DEG is that the fan is aligned with a certain reference direction. The positive angle of yaw of the blower represents that the blower yaw to the clockwise side of 0 degrees; negative angle, representing the fan yawing to the 0 counter-clockwise side. It will be readily appreciated that when the angle reaches 360 ° or an integer multiple thereof, it is indicated that the rotation has reached a certain number of turns.
One such control logic involves untwisting, as shown in FIG. 1.
The cable-releasing control logic comprises a design idea that the power generation efficiency is improved as much as possible, namely, the cable releasing is triggered when the wind condition is not suitable for power generation and the unit is in a standby state, or the cable twisting angle reaches a larger dynamic cable releasing threshold (for example +/-900 degrees) when the power generation state is achieved, and the machine is stopped for cable releasing.
Specifically, as shown in fig. 1, when the wind turbine generator is in the grid-connected power generation state, if the cable twisting angle reaches a larger dynamic cable untwisting threshold value, immediately triggering dynamic cable untwisting of the wind turbine generator, at this time, the wind turbine generator exits the grid-connected power generation state, and performing anticlockwise or clockwise yaw motion according to the positive and negative of the actual cable twisting angle until the wind deviation angle is 0 degree and the absolute value of the cable untwisting angle is smaller than 180 degrees;
when the wind turbine generator is in a standby state, firstly judging whether the current cable twisting angle reaches a small static cable untwisting threshold value (usually about +/-600 degrees), if so, judging whether the wind speed condition in the past period (usually 10 to 20 minutes) meets the power generation condition, and if not, performing anticlockwise or clockwise yaw motion for a plurality of circles according to the positive and negative of the actual cable twisting angle until the absolute value of the cable untwisting angle is smaller than 180 degrees.
According to the logic diagram in fig. 1. When the wind turbine generator system carries out dynamic cable untwisting, the cable twisting angle is in a relatively high value, so that the wind turbine generator system can safely run, the wind turbine generator system is immediately braked and stopped, and the dynamic cable untwisting is carried out, the logic judgment of the process is relatively clear, and defects are not easy to occur in practice. When the wind turbine generator system performs static cable-releasing, the following two yaw cable-releasing logic defects sometimes occur according to different yaw system designs of different host manufacturers.
A logic defect triggers a cable release for an abnormal start-up procedure. The definition is as follows: and in the starting process of the wind turbine, yaw is firstly carried out according to a first direction, then cable releasing is triggered, and after the reverse cable releasing is completed, secondary deflection is carried out according to the first direction to complete the preparation of starting.
A schematic diagram of a scenario in which the start-up procedure abnormally triggers the cable release is shown in fig. 2. After the wind turbine generator is stopped in a standby mode for a period of time and wind conditions meet the requirement that the wind turbine generator is started again to enter a power generation state, the wind direction can be changed greatly, and in order to ensure the running safety of the wind turbine generator and improve the generated energy, a yaw system of the wind turbine generator can automatically perform yaw motion through logic according to the wind turbine generator direction and the incoming wind direction, so that the wind turbine generator is opposite to the incoming wind direction. And yawing the yawing systems of part of wind turbines according to the yaw direction corresponding to the shortest yaw path by default, triggering a static cable-releasing threshold value in the yaw action, releasing cables under the condition that wind conditions meet grid-connected power generation, and finally completing the preparation of starting the wind turbines in the wind process by secondary yaw, wherein the wind turbines enter a grid-connected power generation state. In this process, since an ineffective yaw operation is performed for a certain period of time, the time for starting grid-connected power generation is delayed, and the power generation amount is lost.
Another logical defect, static delay untwisting, is defined as: under the standby state, the wind turbine generator meets the static cable releasing condition, the cable releasing is not executed beyond the first predefined time, the wind condition meets the power generating condition, and the wind turbine generator does not enter the power generating state at the second predefined time.
A schematic of static delay untwisting is shown in fig. 3. The static cable-releasing threshold is used for controlling the cable-releasing logic trigger of the wind turbine generator in a standby state, and cable releasing is performed in advance when the wind speed is low and the power generation condition is not met. When the ambient wind speed is reduced below the cut-in wind speed of the wind turbine, and the wind turbine enters a standby state from grid-connected operation, the cable twisting angle of the wind turbine may be higher than the static cable untwisting threshold. If the wind speed is lower than the cut-in wind speed in a period of time, the wind turbine generator set should trigger a static cable releasing logic to release the cable in advance. Due to the fact that the waiting time set value of part of wind turbines is high, wind conditions of the wind turbines in the static cable-releasing action process meet power generation conditions, and therefore a certain amount of generated energy is lost.
In order to accurately identify both defect modes. The embodiment of the application predicts based on a pre-trained neural network model. As shown in fig. 4, the specific method flow includes:
s41, acquiring sensing data of a discrete time point acquired by a wind turbine through a sensor, wherein the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
S42, calculating input data of a pre-trained neural network model according to the sensing data;
s43, inputting the input data into the pre-trained neural network model, and predicting labels corresponding to defect modes of the input data.
Further in a more detailed embodiment of the present application, the model training and data processing procedure of the above method is described.
A step of training the neural network model, comprising:
(1) Collecting sensing data at discrete time points of historical data, comprising: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
Specifically, historical operation data of wind turbines of 40 wind turbines in a wind farm can be collected, for example, the starting time is 2021, 1 month and 1 day, the ending time is 2021, 3 months and 31 days, and the time interval is 1 minute.
(2) Calculating a torsion angle change value of discrete time points of historical dataAnd a wind deviation angle θ;
calculating the torsion cable angle change value of the wind turbine generatorThe calculation formula is as follows: wherein->As the cable twisting angle at the current moment,the cable twisting angle of the wind turbine generator after 10 minutes; the wind deviation angle theta of the wind turbine generator is calculated, and the calculation formula is as follows: /> Wherein θ ranges from [ -180 °, +180°]The value of the integer k is adjusted to ensure that the wind deviation angle is within a specified range.
(3) According to the moment of triggering the cable-releasing action command every time in the historical data, correspondingly determining time points of 60 samples by taking the cable-releasing action command as a center;
the duration of the untwisting action is as follows: and 60 minutes taking a cable release control logic trigger cable release action instruction of the wind turbine generator as a central time point. I.e. the first 30 minutes to the last 30 minutes of the moment of triggering the untwisting action, total a period of 1 hour.
(4) Labeling each sample to finish the collection of sample data;
aiming at the cable-releasing behavior of each wind turbine generator, the data label y is manually marked, the labels are divided into three types of 0, 1 and 2, wherein 0 represents that the cable-releasing is normal, 1 represents that the cable-releasing is abnormal in the starting process, and 2 represents that the cable-releasing is static delay cable-releasing. In the embodiment, the data of the cable-releasing behavior of the 91 times of wind turbine generator are obtained through the data, and the data are judged by a manual expert, wherein 28 times of the data are abnormal triggering cable-releasing in the starting process, and 22 times of the data are static delay cable-releasing.
Sample data, including input data as And tag data y i 。
(5) Training of the neural network model is completed by using the collected sample data.
In this embodiment, training data X corresponding to each untwisting behavior i The dimension of (2) is 1 x 60 x 7, corresponding to 1 behavior, data at 60 time points, 7 features at each time point. The convolutional neural network structure is shown in fig. 5, the number of the first layer of convolutional kernels is 3, the number of the second layer of convolutional kernels is 6, and the output size and the input of each convolutional kernel are kept unchanged in the same padding mode; the downsampling proportion of the first layer of the pooling layer is 3 multiplied by 1, the output data dimension is 3 multiplied by 20 multiplied by 7, the downsampling proportion of the second layer of the pooling layer is 2 multiplied by 1, and the output data dimension is 6 multiplied by 20 multiplied by 7; the input dimension of the first full-connection layer is 840, and the output dimension is 64; of the second layer of all-connected layersThe input dimension is 64 and the output dimension is 3.
The trained neural network model can be applied to defect pattern recognition. The specific flow comprises the following steps: and acquiring sensing data in real time, calculating input data of the neural network model based on the sensing data, and inputting the input data into the neural network model to obtain a corresponding prediction label.
In an application example, the model obtained in the neural network training process is applied to data of wind turbines 2021, 5 months and 2021, 11 months of the same wind farm for identification, and the identification result is shown in table 1.
| Number of times | Model identification: 0 | Model identification: 1 | Model identification: 2 |
| And (3) manual judgment: 0 | 153 | 1 | 5 |
| And (3) manual judgment: 1 | 3 | 46 | 3 |
| And (3) manual judgment: 2 | 1 | 2 | 38 |
TABLE 1
As can be seen from the data in table 1, the method for detecting the cable-untwisting logic defect provided by the invention can obtain good effect in practical application. The method has good adaptability to other wind farms, and can be popularized and applied.
In summary, in the application scheme of the invention, by acquiring the sensing data of the discrete time points of the wind turbine, calculating based on the sensing data to obtain the input data of the neural network model, and inputting the input data into the neural network model, the label corresponding to the defect mode of the input data can be predicted, so that the defect mode of the cable releasing behavior of the wind turbine can be accurately identified, the improvement of the cable releasing control logic of the wind turbine is facilitated, and the improvement of the power generating efficiency and the economic benefit of the wind turbine is facilitated.
Embodiments of the present application provide a fan untwisting defect mode identification device, the device including:
the sensing module is used for acquiring sensing data of a discrete time point acquired by the wind turbine generator through a sensor, and the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
The input data calculation module is used for calculating input data of a pre-trained neural network model according to the sensing data;
and the prediction module is used for inputting the input data into the pre-trained neural network model and predicting labels corresponding to defect modes of the input data.
In an embodiment of the present application, there is correspondingly provided a computer readable storage medium, where a computer program is stored, and when the computer program is executed by a computer device, the method in the embodiment of the present application is implemented.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In the several embodiments provided in the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the above elements is merely a logical functional division, and there may be additional divisions in actual implementation, e.g., multiple elements or components may be combined or integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The integrated units implemented in the form of software functional units described above may be stored in a computer readable storage medium. The software functional unit is stored in a storage medium, and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a Processor (Processor) to perform part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (10)
1. A method for identifying a fan untwisting defect mode, the method comprising:
acquiring sensing data of a discrete time point acquired by a wind turbine through a sensor, wherein the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
Calculating input data of a pre-trained neural network model according to the sensing data;
inputting the input data into the pre-trained neural network model, and predicting a label corresponding to a defect mode of the input data.
2. The method for identifying a fan untwisting defect mode according to claim 1, wherein calculating input data of a pre-trained neural network model according to the sensing data comprises:
cable twisting angle according to discrete time pointsCalculating the torsion angle change value of the set time interval
According to the wind turbine generator set orientation corresponding to each discrete time pointAnd the geographic wind direction/>Calculating a wind deviation angle theta;
according to the wind speed v and the power P corresponding to each discrete time point act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->A torsion cable angle variation value +>And the wind deviation angle theta, and form 7-dimension input dataWhere t is the time sequence number of the discrete time point.
3. The fan untwisting defect pattern recognition method according to claim 2, wherein the set time interval is 10 minutes; the torsion cable angle change value of the set time intervalThe calculation formula of (2) is as follows:
the calculation formula of the wind deviation angle theta is as follows:
wherein k is an integer, and the value of the integer satisfies θ in the range of [ -180 DEG, +180 DEG ].
4. A fan untwisting defect pattern recognition method as claimed in claim 3, wherein the tag of the defect pattern comprises: 0. three classes 1 and 2, wherein:
class 0 labels, corresponding to normal untwisting behavior;
the class 1 labels are used for triggering cable disconnection abnormally in the starting process of the wind turbine generator;
and 2 types of labels correspond to static delay untwisting of the wind turbine generator.
5. The method for identifying a fan untwisting defect mode according to claim 4, wherein the duration period of the untwisting behavior is: and 60 minutes taking a cable release control logic trigger cable release action instruction of the wind turbine generator as a central time point.
6. The method for identifying a fan untwisting defect mode according to claim 5, wherein the input data of 60 groups of discrete time points is input into the neural network model to predict a label of a defect mode of a corresponding untwisting behavior.
7. The method for identifying a fan untwisting defect mode according to claim 6, wherein the starting process is abnormal to trigger untwisting, defined as: in the starting process of the wind turbine, yaw is firstly carried out according to a first direction, then cable disconnection is triggered, and after the reverse cable disconnection is completed, secondary deviation is carried out according to the first direction to complete starting preparation;
the static delay untwisting is defined as: under the standby state, the wind turbine generator meets the static cable releasing condition, the cable releasing is not executed beyond the first predefined time, the wind condition meets the power generating condition, and the wind turbine generator does not enter the power generating state at the second predefined time.
8. The method for identifying a fan untwisting defect pattern according to claim 7, further comprising the step of training the neural network model;
a step of training the neural network model, comprising:
collecting sensing data at discrete time points of historical data, comprising: wind speed v, power P act Angle of twisted cableGeographic wind direction->And wind turbine generator system orientation->
Calculating a torsion angle change value of discrete time points of historical dataAnd a wind deviation angle θ;
according to the moment of triggering the cable-releasing action command every time in the historical data, correspondingly determining the discrete time points of 60 samples by taking the cable-releasing action command as a center;
labeling each sample of the determined discrete time points to finish the collection of sample data;
training of the neural network model is completed by using the collected sample data.
9. A fan untwisting defect pattern recognition device, the device comprising:
the sensing module is used for acquiring sensing data of a discrete time point acquired by the wind turbine generator through a sensor, and the sensing data comprises: wind speed v, power P act Angle of twisted cableGeographic windTo->And wind turbine generator system orientation->
The input data calculation module is used for calculating input data of a pre-trained neural network model according to the sensing data;
and the prediction module is used for inputting the input data into the pre-trained neural network model and predicting labels corresponding to defect modes of the input data.
10. A computer readable storage medium, wherein a computer program is stored in the computer readable storage medium, and the computer program controls a device where the computer readable storage medium is located to execute the fan untwisting defect mode identification method according to any one of claims 1 to 8 when running.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311823721.8A CN117828494A (en) | 2023-12-27 | 2023-12-27 | Fan cable-releasing defect mode identification method, device and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311823721.8A CN117828494A (en) | 2023-12-27 | 2023-12-27 | Fan cable-releasing defect mode identification method, device and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117828494A true CN117828494A (en) | 2024-04-05 |
Family
ID=90522342
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311823721.8A Pending CN117828494A (en) | 2023-12-27 | 2023-12-27 | Fan cable-releasing defect mode identification method, device and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117828494A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118462471A (en) * | 2024-07-12 | 2024-08-09 | 长江三峡集团福建能源投资有限公司 | Control method and device for static cable-releasing of fan, controller and storage medium |
-
2023
- 2023-12-27 CN CN202311823721.8A patent/CN117828494A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118462471A (en) * | 2024-07-12 | 2024-08-09 | 长江三峡集团福建能源投资有限公司 | Control method and device for static cable-releasing of fan, controller and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105041570B (en) | Yaw control method and device for wind turbine generator | |
| CN113591359B (en) | Wind turbine generator set cut-in/cut-out wind speed adjusting and optimizing method, system and equipment medium | |
| CN112459965B (en) | Yaw optimization control method, device, equipment and medium considering wind power plant wake flow | |
| US20160084224A1 (en) | System and method for optimizing wind farm performance | |
| CN108072524B (en) | Wind turbine generator gearbox bearing fault early warning method | |
| CN108223269B (en) | Method and device for passing through overspeed fault of wind generating set | |
| CN117828494A (en) | Fan cable-releasing defect mode identification method, device and medium | |
| WO2019165752A1 (en) | Method and apparatus for dynamically determining yaw control precision | |
| AU2018334593B2 (en) | Method and system for controlling wind turbine based on sectors | |
| EP3608538A1 (en) | Model-based repowering solutions for wind turbines | |
| CN111287911A (en) | Wind turbine fatigue load early warning method and system | |
| CN117350159A (en) | Generator heat radiation performance early warning method, device, equipment and storage medium | |
| CN112598148A (en) | Fan variable pitch motor temperature fault early warning method based on collaborative expression and LightGBM algorithm | |
| JP2014202190A (en) | Control apparatus, control method, and program | |
| CN114778116A (en) | Fault early warning method and system for variable pitch bearing of wind generating set | |
| CN114139803A (en) | Method, system, equipment and medium for predicting short-term power of wind power station | |
| WO2019066934A1 (en) | Wind turbine performance monitoring | |
| CN111456899A (en) | Minimum headroom control system, method, electronic device, and storage medium | |
| CN116976650B (en) | Power grid lean management regulation and control method based on big data | |
| Tang et al. | Wind Turbine Pitch System Fault Detection Using ssODM-DSTA | |
| CN112287522B (en) | Optimization method and system for driving capability of variable pitch system of wind turbine generator and computer readable storage medium | |
| JP3184277U (en) | Power generation system | |
| CN116108989B (en) | Wind power ultra-short-term power prediction method, system, storage medium and device | |
| CN113468765B (en) | Component life determining method, system and readable storage medium | |
| CN114987728B (en) | Ship charging method, ship charging device, ship charging equipment and storage medium |
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 |