CN114439703A - Wind turbine generator blade imbalance fault monitoring method based on vehicle-mounted device - Google Patents
Wind turbine generator blade imbalance fault monitoring method based on vehicle-mounted device Download PDFInfo
- Publication number
- CN114439703A CN114439703A CN202210112470.1A CN202210112470A CN114439703A CN 114439703 A CN114439703 A CN 114439703A CN 202210112470 A CN202210112470 A CN 202210112470A CN 114439703 A CN114439703 A CN 114439703A
- Authority
- CN
- China
- Prior art keywords
- blade
- clearance
- vehicle
- wind turbine
- actual
- 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
-
- 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
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F05—INDEXING SCHEMES RELATING TO ENGINES OR PUMPS IN VARIOUS SUBCLASSES OF CLASSES F01-F04
- F05B—INDEXING SCHEME RELATING TO WIND, SPRING, WEIGHT, INERTIA OR LIKE MOTORS, TO MACHINES OR ENGINES FOR LIQUIDS COVERED BY SUBCLASSES F03B, F03D AND F03G
- F05B2260/00—Function
- F05B2260/83—Testing, e.g. methods, components or tools therefor
Landscapes
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Sustainable Development (AREA)
- Sustainable Energy (AREA)
- Chemical & Material Sciences (AREA)
- Combustion & Propulsion (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Wind Motors (AREA)
Abstract
The invention belongs to a blade imbalance fault monitoring method, and aims to solve the technical problems that a vibration sensor needs to be installed for acquiring vibration signals, the existing blade imbalance monitoring method needs to monitor and manually adjust based on a large amount of manual labor force, and is time-consuming and labor-consuming. Numerous fans are monitored in a circulating mode through the vehicle-mounted device, and the defects of time and labor consumption of a traditional manpower monitoring method are overcome.
Description
Technical Field
The invention belongs to a blade imbalance fault monitoring method, and particularly relates to a wind turbine generator blade imbalance fault monitoring method based on a vehicle-mounted device.
Background
Wind power generation is a renewable green energy which is being vigorously developed, can effectively save water and coal resources, lightens atmospheric pollution and has profound significance for protecting ecological environment. Wind farms are usually located in remote areas, where the technical and operational conditions are generally poor, and therefore the reliability and safety of the wind power system is of crucial importance. The blade is used as a key part for capturing wind energy in the wind turbine generator, and the performance of the blade directly influences the overall performance and the power generation quality of the fan.
Imbalance faults of a fan account for a relatively large proportion of all fan faults, and imbalance faults typically occur on blades or shafts. Among other things, imbalance failure of the blades is typically caused by deformation, wear, fatigue, icing, etc. caused by faults, aging during installation operations. Another common imbalance fault is aerodynamic asymmetry, which can be caused by high speed wind shear and errors in control mechanisms. If one blade is spaced slightly differently from the other two blades due to a mistake in the control mechanism, an imbalance in the shaft torque will occur, resulting in an asymmetric aerodynamic force. Because the structure of wind turbine generator system is exquisite and cost of maintenance is high, in time monitor and maintain the unbalance etc. trouble and can practice thrift the cost, bring huge economic benefits.
In recent years, researchers at home and abroad develop a blade imbalance fault monitoring technology. The relevant scholars studied the vibrations of the wind turbine generator set caused by the dynamic balance, but do not propose a specific blade imbalance determination method. In addition, aiming at the problem of unbalance of the fan blades, a student establishes a fan model and analyzes the vibration characteristics of the unit caused by the unbalance of the blades by using a finite element method; or the vibration signals of the unit are collected on a simulation experiment platform to carry out spectrum analysis, and the result shows that the vibration of the unit in the horizontal direction can be increased under the unbalanced condition. However, the vibration sensor is required to be installed for collecting the vibration signal, so that the monitoring cost is increased, and secondly, the real-time monitoring is difficult to be carried out on the cabin which is installed at the height of dozens of meters. Still someone has analyzed the relation of blade unbalance and generated power, because the calculation of power needs three-phase current and voltage, and it is troublesome to gather and calculate, consequently, the restriction analysis power effective value or single-phase power change come the analysis blade unbalance, and the analysis result is difficult to avoid partially. In addition, most of the existing blade balance monitoring methods are based on manual monitoring and manual adjustment of a large amount of manual labor force, and are time-consuming and labor-consuming.
Disclosure of Invention
The invention provides a wind turbine generator set blade imbalance fault monitoring method based on a vehicle-mounted device, and aims to solve the technical problems that a vibration sensor is required to be installed for acquiring vibration signals, manual monitoring and manual adjustment are required based on a large amount of manual labor force, and time and labor are consumed in the conventional blade imbalance monitoring method.
In order to achieve the purpose, the invention adopts the following technical scheme to realize the purpose:
a wind turbine generator blade imbalance fault monitoring method based on a vehicle-mounted device is characterized by comprising the following steps:
s1, driving to the side face of the blade of the wind turbine generator along with a vehicle through a vehicle-mounted image acquisition device, and acquiring blade image data and tower column images of the wind turbine generator;
s2, calibrating the blade image data and calibrating a blade area;
s3, extracting the characteristics of the blade area to obtain the tip coordinates of each blade;
s4, calculating the clearance pixel distance between the tip of each blade and the side wall of the tower column through the tip coordinates of each blade obtained in the step S3;
s5, calculating to obtain a clearance actual distance between each blade tip and the side wall of the tower column through a physical proportion formula according to a clearance pixel distance between each blade tip and the side wall of the tower column, an actual diameter of the tower column and a pixel diameter of the tower column in the tower column image;
s6, drawing a change curve of the actual clearance distances of all the blades along with time, and monitoring the overall balance condition of the blades of the wind turbine generator;
s7, calculating the average value of the actual clearance distance of each blade in N rotating cycles respectively, drawing the change curve of the actual clearance distance of each blade in the N rotating cycles along with time, and monitoring the regularity and the change characteristics of the balance state of each blade according to the change curve of the actual clearance distance of each blade in the N rotating cycles along with time and the average value of the actual clearance distance of each blade, wherein N is more than or equal to 2;
s8, taking the average value of the actual clearance distances of any blade in N rotation cycles as a first reference, and respectively calculating the difference value between the actual clearance distance of each blade and the first reference to obtain the first clearance difference value of each blade; taking the average value of the clearance actual distances of all the blades as a second reference of the corresponding blades, and respectively calculating the difference value of the clearance actual distance of each blade and the second reference to obtain a second clearance difference value of each blade; and monitoring and analyzing the balance state of each blade according to the first clearance difference value and the second clearance difference value of each blade.
Further, in step S2, the calibrating of the blade image data is specifically to draw a rectangular frame, so that the blades are all located in the rectangular frame, and the blade tips are located at the center of the lower edge of the rectangular frame.
Further, step S3 is specifically to input the blade image data into a Yolov5 target monitoring algorithm model, and perform feature extraction by combining the blade region obtained in step S2.
Further, in step S3, the Yolov5 target monitoring algorithm model is specifically obtained by dividing the blade image data into a training set, a testing set and a verification set, inputting the training set to the Yolov5 target monitoring algorithm model to train the Yolov5 target monitoring algorithm model, and then sequentially inputting the testing set and the verification set to the trained Yolov5 target monitoring algorithm model to perform testing and verification.
Further, step S0 is further included before step S1, a database is established, where the database includes corresponding blade model, blade pitch angle, difference between blade clearance value and clearance average value, and difference between blade clearance value and clearance reference value;
the method further comprises the step S9 of judging whether the balance state of the blade meets the preset requirement, if not, correcting the balance state of the blade, searching the closest corresponding blade pitch angle in a database according to the type of the blade and the first clearance difference value, adjusting the balance state of the blade according to the blade pitch angle, and repeatedly executing the step S1 to the step S9 until the balance state of the blade meets the preset requirement; and the first clearance difference value corresponds to the difference value between the blade clearance value and the clearance reference value in the database.
Further, step S10 is included, updating the database, and updating the blade pitch angle adjusted in step S9, the corresponding first clearance difference and the second clearance difference to the database.
Further, steps S2 to S8 are all completed on an on-board upper computer, and the upper computer and the image acquisition device are on the same vehicle;
in step S9, the adjusting of the blade balance state according to the blade pitch angle is specifically performed by the upper computer sending an adjustment command to a console of the wind turbine generator system according to the blade pitch angle, and adjusting the blade balance state through the console.
Compared with the prior art, the invention has the following beneficial effects:
1. the wind turbine generator blade imbalance fault monitoring method based on the vehicle-mounted device can be used for monitoring the imbalance fault of the blades in the wind power business in real time by using the vehicle-mounted image acquisition device, measuring the balance condition of the blades from multiple angles by combining simple and rapid mathematical operation, realizing the monitoring of the imbalance fault of the blades, monitoring and analyzing by using various visual curve data, and improving the intuitiveness and the accuracy.
2. The invention relates to a wind turbine generator blade imbalance fault monitoring method based on a vehicle-mounted device, which comprises the steps of utilizing a Yolov5 target monitoring algorithm model to calculate monitored blade information, obtaining a horizontal coordinate value of a blade tip and calculating the distance between the horizontal coordinate value and one side of a tower column, further taking the actual diameter and the pixel diameter of the tower column as theoretical basis, obtaining the clearance distance between the actual blade and the tower column through a physical proportion algorithm formula, comprehensively monitoring the balance state of the blade according to the clearance distance, and providing early warning basis and theoretical research basis for preventing the blade from sweeping the tower through the monitoring result.
3. According to the wind turbine generator blade imbalance fault monitoring method based on the vehicle-mounted device, the Yolov5 target monitoring algorithm model with real-time performance is combined with the rapid and efficient blade post-processing method, and the method is applied to the vehicle-mounted device due to the good real-time performance, so that vehicle-mounted automatic inspection and intelligent maintenance of a fan are realized through intelligent monitoring analysis and parameter adjustment, manpower and material resources are effectively reduced, and operation and maintenance of a wind power plant are facilitated.
4. In the invention, the Yolov5 target monitoring algorithm model is trained, tested and verified in sequence, so that the Yolov5 target monitoring algorithm model is more accurate.
5. The monitoring method of the invention also establishes a database, and corrects the balance state of the blade according to the monitoring result, so that the monitoring method of the invention is more accurate, and can quickly and accurately correct through the database.
6. The monitoring method of the invention can also continuously update the database along with the monitoring process, so that the monitoring result is accurate and stable.
7. The monitoring method is based on a vehicle, realizes a monitoring method combining vehicle-mounted automatic inspection and fan maintenance, is based on a deep learning technology, carries out corresponding monitoring processing on video data acquired by vehicle-mounted equipment, intelligently analyzes adjustment parameters of a fan, sends the parameters to a fan control console, is used for correcting the balance angle of the blade, and realizes intelligent operation and maintenance of a wind power station. In addition, numerous fans are monitored in a circulating mode through the vehicle-mounted device, the defects that a traditional manpower monitoring method consumes time and labor are overcome, manpower, material resources and cost are saved, and wind farm operation and maintenance management can be conducted more conveniently.
Drawings
FIG. 1 is a schematic connection diagram of a vehicle-mounted device in a wind turbine generator blade imbalance fault monitoring method based on the vehicle-mounted device according to the invention;
FIG. 2 is a schematic flow diagram of a wind turbine blade imbalance fault monitoring method based on a vehicle-mounted device according to the present invention;
FIG. 3 is a graph showing the clearance between the integral tower tip and the tower column as a function of time in an embodiment of the monitoring method of the present invention;
FIG. 4 is a graph of the individual clearance of three blades as a function of time in an embodiment of the monitoring method of the present invention;
FIG. 5 is a graph of headroom difference over time based on reference data in an embodiment of the monitoring method of the present invention;
FIG. 6 is a graph of the headroom difference over time based on mean data for an embodiment of the monitoring method of the present invention;
wherein: 1-an image acquisition device, 2-an upper computer and 3-a console.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations.
The traditional blade balance state monitoring method based on manpower is high in monitoring cost, time-consuming and labor-consuming, complex in data acquisition process and greatly influenced by installation environment, and therefore the blade balance state monitoring difficulty is high and the accuracy is low. Aiming at the problems, the invention utilizes the advantage of convenience of the vehicle-mounted device, takes the videos of the blades and the tower column shot from the side by the image acquisition device installed on the vehicle-mounted device as research data, sends the data into a network model constructed based on deep learning to perform feature extraction, acquires the position information of the blade tips, utilizes the coordinate information data of the blade tips of the three groups of blades changing along with time to draw a curve graph of the clearance between the blade tips and the tower column changing along with time, thereby analyzing and judging the blade balance state of the wind power plant unit at the current moment, obtaining the adjustment parameters of the fan, connecting the adjustment parameters to a control console of the fan through a wireless network, modifying the parameters, realizing the intelligentized and digitized operation and maintenance of the wind power plant unit, reducing the manpower and ensuring the normal and stable operation of the blades of the fan group.
The method for monitoring the imbalance fault of the blades of the wind turbine generator is deployed and tested on a vehicle-mounted device, and the mounting structure of the vehicle-mounted device is shown in figure 1. Firstly, an image acquisition device 1 is installed on a vehicle, real-time data of different fans are acquired through vehicle running, then video data are transmitted to an upper computer 2 deployed on the vehicle, whether blades are in an unbalanced state or not is judged through a monitoring algorithm, adjustment parameters of the fans are intelligently analyzed and given, finally the parameters are uploaded to a control console 3 of the fans, the control console 3 of the fans adjusts and corrects the blade balanced state, and intelligent operation and maintenance and automatic inspection of the fans are achieved.
As shown in fig. 2, a detailed process of the wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device of the present invention is as follows:
1) data processing and database establishment: the fan blade image data collected by the image collecting device 1 is calibrated and divided, a certain calibration rule is introduced in the data marking process, the calibration rule is that a blade area is marked firstly, a rectangular frame is used for framing the blade, and the central position of the blade tip at the lower edge of the rectangular frame is ensured, so that the monitored blade is positioned in the rectangular frame, the blade tip is positioned at the central position of the lower edge of the rectangular frame to the maximum extent, theoretical basis and support are provided for post-processing of a later model, the monitored central position of the bottom of the blade can be regarded as coordinate point information of the blade tip, and data processing is completed. And dividing the collected fan blade image data into a training set, a testing set and a verification set according to a preset proportion. In addition, a database is established, and the data in the database specifically includes different blade models, blade pitch angles corresponding to the blade models, differences between the blade clearance values corresponding to the blade pitch angles and the clearance average values, and differences between the blade clearance values corresponding to the blade pitch angles and the clearance reference values.
2) Detecting a tip coordinate point: sending a training set and a test set in fan blade image data into a Yolov5 target monitoring algorithm model for model training and testing to obtain characteristic information of the whole blade, verifying the Yolov5 target monitoring algorithm model through a verification set, utilizing positioning information obtained by the Yolov5 target monitoring algorithm model according to a unified rule when calibrating the data set to obtain coordinate information of the blade tip, and subsequently mainly utilizing the horizontal coordinate value of the blade tip of the unit blade;
3) and (3) calculating the clearance distance: the method comprises the steps of calculating the clearance distance from a blade tip to a tower column by using a physical proportion formula, providing reference basis for clearance distance information provided by a unit management system, obtaining two linear equations through a tower column schematic diagram which is actually marked, using the two linear equations for calculating the pixel diameter of the tower column subsequently, knowing the actual diameter R _ real from the left side of the tower column to the right side of the tower column and the pixel distance R _ px in a tower column image, calculating the final clearance actual distance D _ real according to an equal proportion formula through the pixel distance D _ px from the blade tip to the right side of the tower column which is obtained after model monitoring preprocessing.
4) Drawing a curve graph of the change of the various clearance distances: in order to judge the balance state of the final blade, the statistical analysis is carried out by dividing into two parts. The first part is: firstly, integrally counting the abscissa change graphs of the three blade tips of the blades, calculating the clearance distance between the blade tips and a tower column, drawing a change curve graph of the clearance distance of the blade tips and time, and integrally judging whether the blades have unbalanced faults or not; then, respectively counting the tip abscissa change values of the three blades, taking each rotation of the blade as a data period, calculating the clearance average value of the blade A in N data periods, the clearance average value of the blade B in N data periods, and the clearance average value of the blade C in N data periods, wherein N is an integer greater than or equal to 2, and the blade A, the blade B and the blade C are respectively three blades of the fan. On the basis, a tip clearance-time curve transformation graph is drawn by taking time as an independent variable, so that regularity and change characteristics of the balance state of the three blades are further analyzed and judged. The second part is: and respectively taking clearance data and integral mean value data of a certain blade as a reference, drawing a clearance difference value-time change curve graph, and if the blade A is taken as the reference blade, respectively calculating the differences between the clearance values of the blade A, the blade B and the blade C and the clearance mean value in each data period, thereby obtaining the differences delta A, delta B and delta C between the clearance values of the blade A, the blade B and the blade C and the clearance mean value in N data periods. Then, taking the mean value of the clearance actual distances of all the blades as a reference, respectively calculating the difference value between the clearance actual distance of each blade and the mean value of the clearance actual distances of all the blades, and more intuitively judging the balance state of the current blade through the two calculations; finally, the blade imbalance monitoring method is deployed in a vehicle-mounted system, so that the imbalance fault problem of the wind turbine generator blades can be monitored in real time more visually, more conveniently and more intelligently.
5) Parameter correction and blade balance judgment: in order to realize intelligent operation and maintenance of the fan, intelligent analysis and statistics are carried out on the unbalanced blades, and the pitch angles alpha, beta and delta corresponding to the differences delta A, delta B and delta C which are closest to the differences delta A, delta B and delta C are respectively searched in an established database,
Setting the pitch angles of the blades A, B and C as alpha and beta respectively,
and the control console 3 is connected to the fan through a wireless network, and parameters are repeatedly modified until the blades reach a relative balance state, so that the intelligent maintenance of the fan is realized.
6) Updating a database
And updating the database according to the corrected blade parameters which reach the relative balance state.
The following description is given by taking the blade imbalance fault monitoring of a wind turbine in a certain wind field as an example:
Step 1.1, installing an image acquisition device 1, such as a camera, at a certain vehicle fixed position, wherein the position of the image acquisition device 1 should meet the requirement that the shot image data comprises information of a tower column and a blade, transmitting the video data acquired by the image acquisition device 1 to an upper computer 2 on a vehicle, performing frame extraction on the video data by using a related algorithm to obtain a single-frame image for data division and use, and dividing the data into a training set, a test set and a verification set, wherein the data division condition specifically comprises 3398 training sets, 1203 test sets and 426 verification sets. When data calibration is carried out, the position of the blade tip is placed at the bottom center point of the rectangular frame as much as possible, so that post-processing is carried out after algorithm monitoring is facilitated;
step 1.2, a database is established, wherein data in the database comprise different blade models, blade pitch angles corresponding to the blade models, and differences between blade clearance values corresponding to the blade pitch angles and clearance average values and clearance reference values.
The model selects a Yolov5 target monitoring algorithm model with the advantages of real-time performance and accuracy as a basic model, the training data set and the calibration file in the step 1 are sent into a Yolov5 target monitoring algorithm model for network training feature learning, at the moment, the selected pre-training model is light Yolov5s which has a good monitoring rate advantage, and Yolov5s also shows excellent monitoring model accuracy under the condition that the data set is sufficient and the monitoring target is single, so that the wind turbine generator blade monitoring model trained and learned through Yolov5s has good accuracy and real-time performance advantages, and shows 95.31% of accuracy when a test set is tested, and then is verified through a verification set.
Drawing straight lines of the left side edge and the right side edge of the tower column through the interface, mainly obtaining the top point and the bottom point of the tower column on the left side and the top point and the bottom point of the tower column on the right side, and respectively calculating straight line equations of the two straight lines through the four points. For example: the top and bottom points of the left tower column are lftpoint ((a1, b1), (a2, b2)), and the top and bottom points of the right tower column are rightpoint ((c1, c1), (d2, d2)), so that two straight lines of lftline and rgtline are obtained, and the calculation formula is as follows:
(y-b1)/(b2-b1)＝(x-a1)/(a2-a1)
(y-d1)/(d2-d1)＝(x-c1)/(c2-c1)
meanwhile, configuring actual diameter information PillarWidth of the tower column, acquiring relative position coordinates of the blade tips through blade information monitored by a Yolov5 target monitoring algorithm model in the step 2, calculating pixel distances between horizontal coordinates of the blade tip points in the image and two linear equations of the tower column, wherein the distance to the right side is called the blade tip-tower column pixel clearance distance, obtaining a tower column radius pixel value corresponding to the blade tip at the moment, calculating an actual clearance distance value by using an equal proportion formula, and calculating D _ real through the following equal proportion formula:
R_real/D_real＝r_px/d_px
Step 4.1, drawing a curve graph of the clearance between the whole tower tip and the tower column along with the change of the clearance along with time through the actual clearance obtained in the step 3, as shown in fig. 3, counting the change trend of the whole clearance, obtaining mean data mean of the clearance, and integrally judging the balance state of the blades;
step 4.2, respectively drawing a clearance distance change curve graph between the blade tip and the tower column when the three blades run, as shown in fig. 4, at this time, numbering the three blades as blank 1, blank 2 and blank 3, defining each rotation of the blade as one data cycle, respectively calculating a clearance average value of the blade blank 1 in N data cycles, a clearance average value of the blade blank 2 in N data cycles, a clearance average value of the blade blank 3 in N data cycles, and N is an integer greater than or equal to 2;
and 4.3, combining the blade tip abscissa change value with the time change value to draw a clearance difference-time change curve chart based on the reference blade, selecting a blade blank 2 as a reference, calculating the difference between the blade tip points of the other two blades and the reference blade, drawing the curve chart, and judging the balance state of the blade, so as to more intuitively analyze and monitor the unbalance fault problem of the blade. In fig. 5, with the clearance distance of blade2 as a reference, in each data cycle, the difference between the clearance value of blade B and blade C and the reference is calculated respectively, so as to obtain the difference Δ a between the clearance reference value of blade2 and the clearance value of blade1, the difference Δ B between the clearance value of blade2 and the clearance reference value of blade2, and the difference Δ C between the clearance reference value of blade2 and the clearance value of blade3 in N data cycles, and respectively draw a transformation graph with the difference Δ a, Δ B, and Δ C;
and 4.4, combining the balance state of the whole blade with the balance states of the three independent blades, and drawing a clearance difference value-time change curve based on mean data mean, as shown in fig. 6. At this time, the overall mean value data mean in fig. 3 is taken as a reference, a transformation curve graph is respectively drawn by the blade1 and mean difference value Δ a ', the blade2 and mean difference value Δ B ', and the blade3 and mean difference value Δ C ', the balance state of the current unit blade is analyzed and judged, and the monitoring effect of the blade imbalance fault is realized.
Step 5, model deployment and blade balance correction
And (3) deploying the model with the best effect in the step (2) and the whole method of the invention to an upper computer on a vehicle-mounted device for monitoring the imbalance fault of the blade, intelligently analyzing and counting adjustment parameters of the fan, connecting the adjustment parameters to a fan control console by using a wireless network, and correcting the angles of the imbalance blade by modifying the parameters to realize intelligent operation and maintenance of the fan.
Step 5.1, respectively searching the pitch angles alpha and beta corresponding to the numerical values which are closest to the difference values delta A, delta B and delta C in the database established in the step 1.2,
step 5.2, set the pitch angles of the blades A, B, C to alpha, beta respectively,
repeating the step 2 to the step 5 to obtain difference values delta A, delta B and delta C; if the difference values delta A, delta B and delta C are within the error allowable range, the aerodynamic balance of the unit blade is represented; if one is not in the range, continuing to carry out the steps 4-5 until the difference values delta A, delta B and delta C are all in the range allowed by the error;
and 5.3, updating the database, and storing the blade type, the two groups of difference values and the pitch angle of the pneumatically balanced blades of the unit into the database.
As can be seen from the results shown in fig. 4 and 5, the three blades have an unbalanced state, and the clearance average difference distribution map and the clearance reference value difference distribution map have a large difference, and in order to adjust the blade balance parameter, the balance state of the three blades is adjusted multiple times according to the iteration operation in step 5 until the requirement that the clearance average difference and the clearance reference value difference are at the preset threshold value is met, that is, the relative balance state of the three blades is achieved.
According to the invention, through the image acquisition device 1 installed at the fixed position of the vehicle-mounted device, image data with lateral blade tips are acquired at fixed points along with the movement of the vehicle, the image data of the blades are analyzed and processed, and unbalanced faults of the blades are monitored and early warned .
The above is only a preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes will occur to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A wind turbine generator blade imbalance fault monitoring method based on a vehicle-mounted device is characterized by comprising the following steps:
s1, driving to the side face of the blade of the wind turbine generator along with a vehicle through the vehicle-mounted image acquisition device (1), and acquiring blade image data and tower column images of the wind turbine generator;
s2, calibrating the blade image data and calibrating a blade area;
s3, extracting the characteristics of the blade area to obtain the tip coordinates of each blade;
s4, calculating the clearance pixel distance between the tip of each blade and the side wall of the tower column through the tip coordinates of each blade obtained in the step S3;
s5, calculating to obtain a clearance actual distance between each blade tip and the side wall of the tower column through a physical proportion formula according to a clearance pixel distance between each blade tip and the side wall of the tower column, an actual diameter of the tower column and a pixel diameter of the tower column in the tower column image;
s6, drawing a change curve of the actual clearance distances of all the blades along with time, and monitoring the overall balance condition of the blades of the wind turbine generator;
s7, calculating the average value of the actual clearance distance of each blade in N rotating cycles respectively, drawing the change curve of the actual clearance distance of each blade in the N rotating cycles along with time, and monitoring the regularity and the change characteristics of the balance state of each blade according to the change curve of the actual clearance distance of each blade in the N rotating cycles along with time and the average value of the actual clearance distance of each blade, wherein N is more than or equal to 2;
s8, taking the average value of the actual clearance distances of any blade in N rotation cycles as a first reference, and respectively calculating the difference value between the actual clearance distance of each blade and the first reference to obtain the first clearance difference value of each blade; taking the average value of the clearance actual distances of all the blades as a second reference of the corresponding blades, and respectively calculating the difference value of the clearance actual distance of each blade and the second reference to obtain a second clearance difference value of each blade; and monitoring and analyzing the balance state of each blade according to the first clearance difference value and the second clearance difference value of each blade.
2. The wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device according to claim 1, characterized in that: in step S2, the step of calibrating the blade image data is to draw a rectangular frame so that all the blades are located in the rectangular frame and the blade tips are located at the center of the lower edge of the rectangular frame.
3. The wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device according to claim 2, characterized in that: step S3 is specifically to input the blade image data into a Yolov5 target monitoring algorithm model, and perform feature extraction by combining the blade region obtained in step S2.
4. The wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device is characterized in that: in step S3, the Yolov5 target monitoring algorithm model is specifically obtained by dividing the blade image data into a training set, a testing set, and a verification set, inputting the training set to the Yolov5 target monitoring algorithm model to train the Yolov5 target monitoring algorithm model, and then sequentially inputting the testing set and the verification set to the trained Yolov5 target monitoring algorithm model to perform testing and verification.
5. The wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device according to any one of claims 1 to 4, characterized in that:
step S0 is further included before the step S1, a database is established, and the database comprises corresponding blade models, blade pitch angles, differences between blade clearance values and clearance average values and differences between the blade clearance values and clearance reference values;
the method further comprises the step S9 of judging whether the balance state of the blade meets the preset requirement, if not, correcting the balance state of the blade, searching the closest corresponding blade pitch angle in a database according to the type of the blade and the first clearance difference value, adjusting the balance state of the blade according to the blade pitch angle, and repeatedly executing the step S1 to the step S9 until the balance state of the blade meets the preset requirement; and the first clearance difference value corresponds to the difference value between the blade clearance value and the clearance reference value in the database.
6. The wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device is characterized in that: the method further includes step S10, updating the database, and updating the blade pitch angle adjusted in step S9, the corresponding first clearance difference and the second clearance difference into the database.
7. The wind turbine blade imbalance fault monitoring method based on the vehicle-mounted device is characterized in that: the steps S2 to S8 are all completed on a vehicle-mounted upper computer (2), and the upper computer (2) and the image acquisition device (1) are on the same vehicle;
in step S9, the adjusting of the blade balance state according to the blade pitch angle is specifically performed by the upper computer (2) sending an adjustment command to the console (3) of the wind turbine generator system according to the blade pitch angle, and adjusting the blade balance state through the console (3).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210112470.1A CN114439703A (en) | 2022-01-29 | 2022-01-29 | Wind turbine generator blade imbalance fault monitoring method based on vehicle-mounted device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210112470.1A CN114439703A (en) | 2022-01-29 | 2022-01-29 | Wind turbine generator blade imbalance fault monitoring method based on vehicle-mounted device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114439703A true CN114439703A (en) | 2022-05-06 |
Family
ID=81372354
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210112470.1A Pending CN114439703A (en) | 2022-01-29 | 2022-01-29 | Wind turbine generator blade imbalance fault monitoring method based on vehicle-mounted device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114439703A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116107260A (en) * | 2023-04-13 | 2023-05-12 | 西安中科原子精密制造科技有限公司 | Method for reducing interference by sequential sampling of time sequence control three-channel radar |
-
2022
- 2022-01-29 CN CN202210112470.1A patent/CN114439703A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116107260A (en) * | 2023-04-13 | 2023-05-12 | 西安中科原子精密制造科技有限公司 | Method for reducing interference by sequential sampling of time sequence control three-channel radar |
| CN116107260B (en) * | 2023-04-13 | 2023-06-23 | 西安中科原子精密制造科技有限公司 | Method for reducing interference by sequential sampling of time sequence control three-channel radar |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113836762B (en) | Wind turbine and wind power plant digital mirror image simulation display system | |
| CN106815771B (en) | Long-term assessment method for wind farm load | |
| CN101016879A (en) | Method, system and computer program product for nacelle wind speed correction | |
| CN105065212B (en) | A kind of wind power plant Wind turbines method of calibration and system | |
| US11680556B2 (en) | Methods and systems of advanced yaw control of a wind turbine | |
| CN112267972B (en) | Intelligent judging method for abnormal power curve of wind turbine generator | |
| EP3771822A1 (en) | A method for computer-implemented determination of a vertical speed wind profile of a wind field | |
| CN107869420B (en) | Method and system for controlling yaw of wind turbine farm | |
| CN111322205A (en) | Wind turbine generator set wind vane zero error identification method and correction method | |
| CN111878322B (en) | Wind power generator device | |
| CN104564542A (en) | Fault diagnosis system and fault diagnosis method based on massive data technology | |
| CN115218801B (en) | Wind driven generator clearance distance measuring method and device based on machine vision | |
| CN114439703A (en) | Wind turbine generator blade imbalance fault monitoring method based on vehicle-mounted device | |
| CN114165392A (en) | Wind turbine generator set power abnormity diagnosis method and device and storage medium | |
| WO2018193802A1 (en) | Wind power generation system | |
| CN104713730B (en) | Method for determining degeneration rate of aircraft engine according to vibration signal | |
| CN113777351B (en) | Fault diagnosis method and device for wind speed sensor of wind power plant | |
| CN115614292A (en) | Vibration monitoring device and method for vertical water pump unit | |
| CN111797545B (en) | Wind turbine generator yaw reduction coefficient calculation method based on measured data | |
| CN110578659B (en) | System and method for processing SCADA data of wind turbine generator | |
| EP4100646B1 (en) | Wake monitoring, wake management and sensory arrangements to such | |
| CN109798226B (en) | Wind turbine generator tower load prediction method and system | |
| CN108474350A (en) | Run the method and system on wind turbine airport | |
| CN114398842A (en) | Method for evaluating generated energy of operating wind power plant | |
| CN115510726A (en) | Digital rapid evaluation method for operating state of power transmission tower |
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 |