Fan blade clearance distance monitoring method, device, equipment and medium

Abstract

The invention discloses a fan blade clearance distance monitoring method, device and equipment and a medium, and belongs to the technical field of fan monitoring, and the method comprises the steps: installing the monitoring equipment at a preset position of a fan tower drum, and obtaining a fan blade motion image; processing the fan blade motion image by adopting a deep learning network, and identifying position coordinates of a blade tip in the fan blade motion image; in combination with the inclinometer angle and Beidou satellite positioning data, converting the blade tip position coordinates into actual space coordinates according to pre-calibrated parameters; calculating the clearance distance from each blade tip to the fan tower; and filtering the clearance distances of the blades, calculating difference values of the clearance distances of the blades, and taking the maximum value in the difference values as the unbalance degree of the impeller. Through the dynamic compensation technology of multi-sensor fusion, the problem that the clearance distance measurement precision is affected by relative shaking of the blades and the tower drum is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of fan monitoring, in particular to a fan blade clearance distance monitoring method, system, device and medium. BACKGROUND

[0002] The fan impeller unbalance degree is a key indicator for measuring whether the fan blade is running normally. The traditional fan impeller unbalance degree detection uses mechanical equipment detection, such as measuring the vibration signal of the impeller, measuring the weight distribution of the impeller, etc. These schemes require high accuracy of sensors, complex installation, are greatly affected by the humidity and temperature of the environment, and the calculation and evaluation method is also complex. Therefore, a method for measuring the clearance distance to measure the unbalance degree has appeared.

[0003] Currently, there are three technical solutions for measuring the unbalance degree by measuring the clearance value: laser radar, millimeter wave radar and visual technology. The laser radar scheme and the millimeter wave radar scheme are commonly installed at the bottom of the cabin, and the laser beam is scanned to the blade tip, and the output result is stable and reliable, but the laser radar is easily scanned to the water droplets or particulate matter in the air in rain, fog, dust weather, resulting in unstable measurement results, in addition, the laser cost is high and needs to be maintained frequently, the calibration method is complex, and the data cannot be intuitively traced back, in addition, the long distance measurement precision of the laser is insufficient.

[0004] The visual scheme is installed below the rudder, and the image of the blade and the tower is collected by a high-definition camera and a light supplementing device at a fixed angle. The edge detection is generally used to identify the coordinates of the blade tip position and the tower center on the image, and then the pixel difference between the two points is calculated, and the three-dimensional distance is calculated through a certain proportion, so as to obtain the clearance distance. This scheme is convenient for tracing back, and the calibration is simple, but the current traditional edge detection visual algorithm is greatly affected by rain, fog and dust, and when the fan blade has a shadow, the accuracy is not high, and a separate algorithm processing unit is needed, which has high computing power cost. SUMMARY

[0005] In view of the above problems, the present application is proposed.

[0006] Therefore, the technical problem solved by the present application is that the camera dynamic blur leads to insufficient blade tip recognition accuracy, and the clearance distance calculation result is unreliable under the shaking condition of the fan.

[0007] To solve the above technical problems, the present application provides the following technical scheme: a fan blade clearance distance monitoring method, comprising, installing a monitoring device at a predetermined position of a fan tower, and obtaining a fan blade motion image; processing the fan blade motion image using a deep learning network to identify the position coordinates of the blade tip in the fan blade motion image; Obtain the inclination angle of the monitoring device and the Beidou satellite positioning data; Convert the blade tip position coordinates into actual spatial coordinates according to the pre-calibrated parameters in combination with the inclination angle and the Beidou satellite positioning data; Calculate the clearance distance of each blade tip to the fan tower drum; Filter the clearance distance of each blade, calculate the difference value of the clearance distance of each blade, and take the maximum value in the difference value as the imbalance degree of the impeller; When the clearance distance or the imbalance degree exceeds the preset threshold, an alarm signal is sent and an alarm video is saved.

[0008] As a preferred scheme of the wind turbine blade clearance distance monitoring method, the step of obtaining the wind turbine blade motion image comprises: collecting environmental illumination information data, establishing an illumination evaluation model through the environmental illumination information data to obtain illumination compensation parameters; and controlling the monitoring device through the illumination compensation parameters and camera exposure parameters to obtain the wind turbine blade motion image.

[0009] The preferred technical scheme has the beneficial effects that: through the combination control of the environmental illumination evaluation model and the high-speed motion shooting strategy, adaptive image acquisition for high-speed motion of the wind turbine blade and complex environmental conditions is realized.

[0010] As a preferred scheme of the wind turbine blade clearance distance monitoring method, the step of obtaining the wind turbine blade motion image comprises: collecting environmental illumination information data, establishing an illumination evaluation model through the environmental illumination information data to obtain illumination compensation parameters; and controlling the monitoring device through the illumination compensation parameters and camera exposure parameters to obtain the wind turbine blade motion image.

[0011] As a preferred scheme of the wind turbine blade clearance distance monitoring method, the step of converting the blade tip position coordinates into actual spatial coordinates comprises: establishing a pixel-to-space mapping model through coordinate conversion relationship information data in the pre-calibrated parameters to obtain static conversion parameters; modifying the static conversion parameters through dynamic coordinate system compensation data to obtain dynamic conversion parameters; and performing three-dimensional space projection calculation through the blade tip position coordinates in combination with the dynamic conversion parameters to obtain the actual spatial coordinates of the blade tip.

[0012] As a preferred scheme of the fan blade clearance distance monitoring method, the step of obtaining the dynamic coordinate system compensation data comprises: calculating the angle deviation value by subtracting the inclination angle from the reference angle data; calculating the position deviation value by subtracting the Beidou satellite positioning data from the reference coordinate data; constructing an angle rotation matrix by using the angle deviation value, constructing a position translation matrix by using the position deviation value, and combining the angle rotation matrix and the position translation matrix to obtain the dynamic coordinate system compensation data.

[0013] The preferred technical scheme has the beneficial effect of the fusion compensation method of constructing a rotation matrix and a translation matrix by using the inclination angle deviation and the Beidou positioning position deviation respectively and then combining them.

[0014] As a preferred scheme of the fan blade clearance distance monitoring method, the step of calculating the clearance distance of each blade tip to the fan tower comprises: calculating the straight-line distance value of the blade tip to the tower center by using the actual spatial coordinates of the blade tip and the tower center spatial coordinates in the pre-labeled parameters, wherein the straight-line distance value is calculated by using the three-dimensional Euclidean distance formula; calculating the preliminary clearance distance value by subtracting the tower radius parameter in the pre-labeled parameters from the straight-line distance value; and calculating the clearance distance of each blade tip to the fan tower by using the preliminary clearance distance value and the dynamic coordinate system compensation data.

[0015] As a preferred scheme of the fan blade clearance distance monitoring method, the step of calculating the blade imbalance degree comprises: grouping the clearance distance of each blade tip to the fan tower according to the blade number to obtain the clearance distance value corresponding to each blade; performing digital filtering processing on the clearance distance value corresponding to each blade to obtain the stable clearance distance value of each blade, wherein the digital filtering processing eliminates measurement noise by performing weighted average calculation on the clearance distance values measured continuously for a single blade multiple times; calculating the inter-blade clearance distance difference data by using the stable clearance distance values of the three blades, and extracting the maximum difference value from the inter-blade clearance distance difference data as the blade imbalance degree.

[0016] The preferred technical scheme has the beneficial effect of the algorithm of calculating the maximum inter-blade clearance distance difference as the imbalance degree by performing digital filtering processing on the three blades, which more accurately reflects the actual imbalance state of the blade.

[0017] The application provides a fan blade clearance distance monitoring device.

[0018] To solve the above technical problems, the present application provides the following technical solutions: a fan blade clearance distance monitoring device, comprising: A protective shell; A high-speed camera arranged inside the protective shell for collecting fan blade motion images; An infrared fill light arranged inside the protective shell for providing fill light; An inclinometer arranged inside the protective shell for measuring real-time cabin sway angle data and initial installation angle; A Beidou satellite positioning system arranged inside the protective shell for obtaining tower cylinder position coordinate data; A temperature sensor arranged inside the protective shell for monitoring device operating environment temperature and starting the heater inside the device; A controller arranged inside the protective shell and connected with the high-speed camera, infrared fill light, inclinometer, Beidou satellite positioning system and temperature sensor for processing image data and sensor data, performing blade tip identification and clearance distance calculation; A communication interface arranged on the protective shell and connected with the controller for outputting clearance distance and unbalance degree data to external devices; A fan blade tip identification module installed inside the protective shell for identifying the position of the blade tip on the image in real time.

[0019] The present application provides a computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor executes the computer program to realize the steps of the fan blade clearance distance monitoring method.

[0020] The present application provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program is executed by a processor to realize the steps of the fan blade clearance distance monitoring method.

[0021] The present application has the following advantages: Through the dynamic compensation technology of multi-sensor fusion, the influence of the relative sway of the blade and the tower cylinder on the measurement accuracy of the clearance distance is solved. The measurement instability phenomenon caused by tower cylinder vibration during the operation of the traditional scheme is fundamentally improved, and the system can obtain inclinometer and Beidou positioning data in real time, and through the construction of dynamic rotation matrix and translation matrix, the coordinate system is compensated to ensure that reliable clearance distance data can be obtained under various operating conditions.

[0022] The adaptive environment sensing technology adopted enhances the working ability of the system in adverse weather conditions. Through the intelligent combination of ambient light evaluation and high-speed motion shooting strategy, stable tip recognition performance can be maintained in complex environmental conditions such as night, light rain, dust, low haze, etc. The collaborative work of the infrared fill light and the AI low-light enhancement function solves the recognition failure problem of the traditional visual scheme in insufficient light conditions, greatly expanding the working environment range of the system.

[0023] The progressive tip recognition method of the present application is specially optimized for the characteristics of the fan tip, which is a high-speed and small target. Through the gradual refinement process of edge feature extraction, blade contour modeling, tip corner detection and accurate positioning, the tip recognition accuracy in complex background is effectively improved, the influence of factors such as blade shadow and background interference on the recognition result is reduced, and more stable and accurate tip position determination is realized. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0025] Fig. 1 A general flowchart of a fan blade clearance distance monitoring method provided by an embodiment of the present application.

[0026] Fig. 2 The monitoring device installation position of a fan blade clearance distance monitoring method provided by an embodiment of the present application.

[0027] Fig. 3 The tip position obtained by a fan blade clearance distance monitoring method provided by an embodiment of the present application. DETAILED DESCRIPTION

[0028] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0029] Embodiment 1, refer to Figs. 1-3 For an embodiment of the present application, the embodiment provides a fan blade clearance distance monitoring method, comprising S1-S7: S1, install the monitoring device at a preset position of the fan tower drum to obtain a fan blade movement image.

[0030] The monitoring device is designed with high integration, and components such as a high-speed camera, an infrared fill light, an AI low-light enhancement module and a temperature sensor are integrated in a protective shell with a size of 192mm*116mm*106mm and a protection level of IP65, and the monitoring device is installed at a preset position below the fan cabin.

[0031] The step of obtaining the fan blade movement image includes S1.1-S1.3. S1.1, collect ambient light information data, and establish a light evaluation model based on the ambient light information data to obtain light compensation parameters; In this embodiment, the integrated light sensor collects the current ambient light intensity value, and the image quality analysis algorithm evaluates the contrast and definition parameters of the image. The light evaluation model classifies the environment according to the light intensity range: when the light intensity is below 500 lux, it is determined as night mode, 500-2000 lux as low light mode, and 2000 lux or above as normal light mode. According to different light modes, the light compensation parameters are calculated, including the opening intensity of the infrared fill light and the enhancement coefficient of the AI low-light enhancement algorithm.

[0032] S1.2, collect blade movement speed information data, and establish a high-speed movement shooting strategy based on the blade movement speed information data to obtain camera exposure parameters.

[0033] By analyzing the fan speed information and the time of the blade passing through the field angle, the blade movement speed is calculated. When the blade linear speed exceeds 80m / s, it is determined as a high-speed movement state. The high-speed movement shooting strategy automatically adjusts the shutter speed and ISO sensitivity parameters of the camera according to the blade speed: for high-speed movement state, the shutter speed is set to 1 / 2000s or above, and the ISO value is appropriately increased to ensure image brightness, ensuring that at least 10 clear blade tip image data can be collected by the camera at the maximum speed of the blade.

[0034] S1.3, combine the light compensation parameters and the camera exposure parameters to control the monitoring device to obtain the fan blade movement image.

[0035] The controller fuses the light compensation parameters and the camera exposure parameters. When detecting rain, fog, dust and other adverse weather conditions, the infrared fill light is automatically started and adjusted to the corresponding brightness, and the AI low-light enhancement function is activated for real-time image processing. Through this adaptive control method, high-quality fan blade movement images can be obtained in complex environmental conditions such as night, light rain, dust and low haze, and the image frame rate is maintained at 25Hz, meeting the requirements of subsequent blade tip identification and clearance distance calculation.

[0036] S2, using a deep learning network to process the fan blade motion image, identifying the position coordinates of the blade tip in the fan blade motion image.

[0037] The progressive blade tip identification algorithm based on deep learning is used to process the fan blade motion image. The deep learning network uses an improved yolov8 architecture, which integrates an attention mechanism in the backbone feature extraction module and is optimized for training high-speed and small target fan blade tips.

[0038] The step of identifying the position coordinates of the blade tip in the fan blade motion image includes S2.1-S2.2: S2.1, using a deep learning network algorithm to reduce the interference of environmental factors through Zero-DCE, and the deep learning network algorithm identifies the blade tip candidate points.

[0039] S2.2, positioning through the blade tip candidate point information data combined with the constraint condition to obtain the blade tip position coordinates.

[0040] Multiple constraint conditions are used to screen and accurately position the blade tip candidate points. The constraint conditions include: geometric constraint (the blade tip must be located at the end of the blade contour), position constraint (the distance from the blade tip to the blade root should be within a reasonable range), and angle constraint (the angle difference between adjacent blades should be close to 120°). By calculating the matching degree of each candidate point with the constraint condition, the point with the highest matching degree is selected as the final blade tip position coordinates. When multiple blades appear in the image at the same time, the blade identity is identified according to the blade motion direction and time sequence relationship to ensure that each blade tip position coordinates can be correctly corresponded to the corresponding blade number. Finally, the accurate blade tip position coordinates data is output, and the coordinate accuracy can reach ±2 pixels.

[0041] S3, obtaining the inclination angle of the monitoring device and the Beidou satellite positioning data.

[0042] In this embodiment, the high-precision inclination angle instrument integrated in the monitoring device is used to collect the shaking angle data of the cabin in real time. The measurement range of the inclination angle instrument is -15° to +15°, the measurement accuracy is 0.01°, and the output frequency is 10Hz. At the same time, the Beidou satellite positioning system obtains the three-dimensional coordinate data of the tower cylinder at a frequency of 1Hz, and the positioning accuracy can reach 2-5 meters. The inclination angle data and Beidou satellite positioning data are transmitted to the controller in real time for subsequent processing.

[0043] S4, combining the inclination angle and Beidou satellite positioning data, and converting the blade tip position coordinates into actual spatial coordinates according to the pre-calibrated parameters.

[0044] The step of converting the blade tip position coordinates into actual spatial coordinates includes S4.1-S4.3: S4.1, establish a pixel-to-space mapping model through coordinate conversion relationship information data in the pre-calibration parameters, and obtain a static conversion parameter.

[0045] A pixel-to-space mapping model is established by using pre-calibration parameters. The pre-calibration parameters include a camera intrinsic parameter matrix K (focal length fx=2000 pixels, fy=2000 pixels, principal point coordinates cx=960 pixels, cy=540 pixels), a distance D=50 meters from a blade tip to the camera, a position of a tower drum center in a camera coordinate system, and the like. Through these calibration parameters, a basic coordinate conversion relationship is established, and a static conversion parameter is obtained, which can convert pixel coordinates into three-dimensional space coordinates in an ideal state.

[0046] S4.2, correct the static conversion parameter through dynamic coordinate system compensation data, and obtain a dynamic conversion parameter.

[0047] The obtaining step of the dynamic coordinate system compensation data includes A1-A3. A1, obtain an angle deviation value through difference calculation of the inclination instrument angle and reference angle data.

[0048] The inclination instrument measures the angle change of the cabin in three axial directions in real time. Real-time angle data of the current inclination instrument is obtained: a pitch angle is 2.3°, a roll angle is 1.8°, and a yaw angle is 0.5°.

[0049] The reference angle data recorded at the time of calibration is read at the same time: a pitch angle is 0°, a roll angle is 0°, and a yaw angle is 0°. The angle deviation values of the respective axial directions are calculated one by one: a pitch angle deviation value , a roll angle deviation value , and a yaw angle deviation value . These angle deviation values reflect the spatial attitude change of the cabin relative to the calibration moment.

[0050] A2, obtain a position deviation value through difference calculation of the Beidou satellite positioning data and reference coordinate data.

[0051] The Beidou satellite positioning system provides real-time three-dimensional coordinates of the tower drum in the geodetic coordinate system. The current Beidou positioning data is obtained: the X coordinate is 1000235.6 m, the Y coordinate is 2000187.3 m, and the Z coordinate is 85.2 m. The reference coordinate data recorded at the time of calibration is read at the same time: the X coordinate is 1000234.0 m, the Y coordinate is 2000186.0 m, and the Z coordinate is 85.0 m.

[0052] The position deviation values of the respective coordinate axes are calculated: an X-axis position deviation value , a Y-axis position deviation value , Z-axis position deviation value These position deviation values represent the spatial displacement of the tower drum relative to the calibration position.

[0053] A3, construct an angle rotation matrix through the angle deviation value, construct a position translation matrix through the position deviation value, and combine the angle rotation matrix and the position translation matrix to obtain dynamic coordinate system compensation data.

[0054] First, construct an angle rotation matrix according to the angle deviation value.

[0055] Convert the angle deviation value from degrees to radians: pitch angle deviation value radians , roll angle deviation value radians , yaw angle deviation value radians . Construct a 3x3 rotation matrix R in the order of ZYX Euler angles.

[0056] Next, construct a position translation matrix according to the position deviation value. The position deviation values ΔX = 1.6 m, ΔY = 1.3 m, and ΔZ = 0.2 m in the X, Y, and Z directions form a 3x1 translation vector T.

[0057] Finally, combine the 3x3 angle rotation matrix R and the 3x1 translation vector T to form a 4x4 homogeneous transformation matrix. The structure of the transformation matrix is: ; This 4x4 transformation matrix contains complete spatial attitude change and position change information, which is used as dynamic coordinate system compensation data for subsequent coordinate conversion correction calculations.

[0058] S4.3, perform three-dimensional space projection calculation by combining the tip position coordinates with the dynamic conversion parameters to obtain the actual spatial coordinates of the tip.

[0059] By performing complete conversion calculation from tip pixel coordinates to actual spatial coordinates. First, the system obtains the tip position coordinates identified in step S2, for example, the pixel coordinates are , is the horizontal direction, is the vertical direction.

[0060] First, perform pixel coordinate to camera coordinate system conversion. Use the inverse matrix of the camera intrinsic parameter matrix to convert the pixel coordinates to normalized coordinates in the camera coordinate system : ; ; wherein , is the principal point coordinate, , is a focal length parameter.

[0061] Next, combined with the depth distance parameter D = 50 meters, the three-dimensional coordinates of the blade tip in the camera coordinate system are calculated: X-axis three-dimensional coordinates: ; Y-axis three-dimensional coordinates: ; Z-axis three-dimensional coordinates: .

[0062] Then, the coordinates are corrected by dynamic coordinate system compensation data. The three-dimensional coordinates in the camera coordinate system are expressed as homogeneous coordinates form: ; The 4x4 transformation matrix obtained in step S4.2 is used to transform the coordinates to obtain the homogeneous coordinate vector of the blade tip in the world coordinate system : .

[0063] The specific calculation process is: perform dot product operation on each row of the transformation matrix and the homogeneous coordinate vector. After matrix operation, the corrected homogeneous coordinates in the world coordinate system are obtained, and the first three components are taken as the actual spatial coordinates of the blade tip.

[0064] Finally, the actual spatial coordinates of the blade tip are output, for example, X = 1000187.3 m, Y = 2000156.8 m, Z = 125.6 m. This coordinate has completed dynamic compensation of cabin shaking and tower displacement, and can accurately reflect the true three-dimensional position of the blade tip in the ground coordinate system. The accuracy of the entire calculation process depends on the accuracy of the camera calibration parameters and the measurement accuracy of the sensor, and the positioning accuracy can usually reach sub-meter level.

[0065] S5, calculate the clearance distance of each blade tip to the wind turbine tower. Specifically, S5.1-S5.3 are included: S5.1, perform three-dimensional spatial distance calculation on the blade tip actual spatial coordinates and the tower center spatial coordinates in the pre-calibrated parameters to obtain a straight line distance value from the blade tip to the tower center, wherein the straight line distance value is calculated by a three-dimensional Euclidean distance formula.

[0066] The three-dimensional distance calculation is performed on the blade tip actual spatial coordinates calculated in step S4 and the pre-calibrated tower center spatial coordinates. Assuming that the blade tip actual spatial coordinates are , and the tower center spatial coordinates are . The three-dimensional Euclidean distance formula is used for calculation: ; get the straight-line distance value from the blade tip to the tower center = 40.7 meters.

[0067] S5.2, calculate the clearance distance correction by subtracting the tower radius parameter in the pre-calibration parameters from the straight-line distance value, and obtain a preliminary clearance distance value.

[0068] Subtract the pre-calibration tower radius parameter from the straight-line distance value to correct the clearance distance. Assuming that the pre-calibration tower radius parameter is , the clearance distance correction calculation is performed, i.e. subtract the tower radius parameter from the straight-line distance value, to obtain 38.2 meters.

[0069] This calculation result represents the shortest distance from the blade tip to the tower surface, i.e. the preliminary clearance distance value is 38.2 meters.

[0070] S5.3, perform three-dimensional space compensation calculation by combining the preliminary clearance distance value with the dynamic coordinate system compensation data, and obtain the clearance distance of each blade tip to the wind turbine tower.

[0071] Combine the preliminary clearance distance value with the dynamic coordinate system compensation data to perform the final three-dimensional space compensation calculation. Considering the possible slight influence of nacelle sway and tower displacement on distance measurement, apply the dynamic compensation information obtained in step A3 for correction. Assuming that the current compensation correction coefficient is 0.98.

[0072] The final clearance distance is the product of the preliminary clearance distance value and the compensation correction coefficient, so the output result is 37.4 meters. Output the final clearance distance of the blade tip to the wind turbine tower as 37.4 meters. This result has considered the influence of dynamic factors such as nacelle sway and tower displacement, and can accurately reflect the actual safety distance between the blade tip and the tower. Repeat the calculation process for each identified blade tip to obtain the clearance distance data of all blade tips.

[0073] S6, filter the clearance distance of each blade, calculate the difference value of the clearance distance of each blade, and take the maximum value in the difference value as the imbalance degree of the impeller.

[0074] The step of calculating the imbalance degree of the impeller includes S6.1-S6.3: S6.1, group the clearance distance of each blade tip to the wind turbine tower according to the blade number, and obtain the clearance distance value corresponding to each blade.

[0075] In this embodiment, the clearance distance data is grouped according to the timing and angular position relationship of the blades appearing in the image. Assuming that the fan has three blades, each blade is assigned a number through blade trajectory and angle recognition: blade 1, blade 2, and blade 3. In a complete rotation cycle, the clearance distance data of each blade tip is collected: the clearance distance value sequence corresponding to blade 1 , the clearance distance value sequence corresponding to blade 2 , and the clearance distance value sequence corresponding to blade 3 . These data are stored and managed separately according to blade number.

[0076] S6.2, the clearance distance values corresponding to each blade are respectively subjected to digital filtering processing to obtain stable clearance distance values for each blade, wherein the digital filtering processing eliminates measurement noise by weighted average calculation of the clearance distance values measured continuously multiple times for a single blade.

[0077] The clearance distance value sequence of each blade is respectively subjected to digital filtering processing. A weighted average filtering algorithm is used, and the weight coefficients from new to old are . Filtering calculation is performed on blade 1: Stable clearance distance value 1 = 37.4 x 0.3 + 37.6 x 0.25 + 37.3 x 0.2 + 37.5 x 0.15 + 37.4 x 0.1 = 37.45 m; Filtering calculation is performed on blade 2: stable clearance distance value 2 = 35.8 x 0.3 + 35.9 x 0.25 + 35.7 x 0.2 + 35.8 x 0.15 + 35.9 x 0.1 = 35.80 m; Filtering calculation is performed on blade 3: stable clearance distance value 3 = 38.1 x 0.3 + 38.3 x 0.25 + 38.0 x 0.2 + 38.2 x 0.15 + 38.1 x 0.1 = 38.13 m.

[0078] Through filtering processing, the stable clearance distance values of the three blades are obtained: 37.45 m, 35.80 m, and 38.13 m.

[0079] S6.3, difference values are calculated between the stable clearance distance values of the three blades to obtain inter-blade clearance distance difference data, and the maximum difference value is extracted from the inter-blade clearance distance difference data as the impeller unbalance degree.

[0080] Difference values are calculated between the stable clearance distance values of the three blades. The inter-blade clearance distance difference is calculated as follows: Difference value ; Difference value ; Difference value ; The maximum difference is extracted from the inter-blade clearance distance difference data The impeller unbalance degree is This value indicates that the current impeller has an unbalance degree of 2.33 meters, reflecting the maximum difference in clearance distance between blade 2 and blade 3.

[0081] S7, when the clearance distance or unbalance degree exceeds the preset threshold, an alarm signal is issued and an alarm video is saved.

[0082] Finally, the calculated clearance distance and unbalance degree are compared with the preset threshold. Assuming that the preset clearance distance safety threshold is 30.0 m and the unbalance degree warning threshold is 2.0 m. By checking the clearance distance of each blade: blade 1 is 37.45 m, blade 2 is 35.80 m, and blade 3 is 38.13 m, all of which are greater than the 30.0 m safety threshold. However, the impeller unbalance degree is 2.33 m, which exceeds the 2.0 m warning threshold.

[0083] An intelligent alarm strategy is adopted: when there are 3 times of unbalance degree exceeding the preset threshold in 5 consecutive detections, an alarm is triggered. Assuming that 4 times of threshold-exceeding situations have been continuously detected, an unbalance degree abnormal alarm signal is immediately sent to the external PLC. At the same time, the video recording function is started, and the monitoring video 30 seconds before and after the current time is saved to the internal storage of the device, facilitating the subsequent analysis of the specific performance and reasons of the impeller unbalance by maintenance personnel.

[0084] Embodiment 2 is an embodiment of the present application, which provides a fan blade clearance distance monitoring device, comprising: a protective shell; a high-speed camera arranged inside the protective shell for collecting fan blade motion images; an infrared fill light arranged inside the protective shell for providing fill light; a tilt meter arranged inside the protective shell for measuring real-time cabin sway angle data and initial installation angle; a Beidou satellite positioning system arranged inside the protective shell for obtaining tower position coordinate data; a temperature sensor arranged inside the protective shell for monitoring the device operating environment temperature and starting the heater inside the device; a controller arranged inside the protective shell and connected with the high-speed camera, infrared fill light, tilt meter, Beidou satellite positioning system and temperature sensor for processing image data and sensor data, performing blade tip recognition and clearance distance calculation; a communication interface arranged on the protective shell and connected with the controller for outputting clearance distance and unbalance degree data to external devices; A fan blade tip recognition module is installed inside the protective shell and used to identify the position of the blade tip on the image in real time.

[0085] Embodiment 3 is an embodiment of the present application, which provides an electronic device suitable for a fan blade clearance distance monitoring method, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions to realize the fan blade clearance distance monitoring method proposed in the above embodiment.

[0086] The embodiment also provides a storage medium having a computer program stored thereon, and the program is executed by a processor to realize the fan blade clearance distance monitoring method proposed in the above embodiment.

[0087] The storage medium proposed in the embodiment and the fan blade clearance distance monitoring method proposed in the above embodiment belong to the same inventive concept, and the technical details not described in the embodiment can be referred to the above embodiment, and the embodiment has the same beneficial effects as the above embodiment.

[0088] From the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary general hardware, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk or an optical disk, etc., including a number of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods of various embodiments of the present application.

[0089] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit the present application, although the present application has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present application can be modified or replaced by equivalent without departing from the spirit and scope of the present application, and all should be covered in the scope of the claims of the present application.

Claims

1. A method of monitoring a tip clearance distance of a wind turbine blade, characterized by, The method comprises the following steps: installing a monitoring device at a preset position of a fan tower drum to obtain a fan blade movement image; processing the fan blade movement image by using a deep learning network to identify the position coordinates of a blade tip in the fan blade movement image; obtaining the angle of an inclinometer and the Beidou satellite positioning data of the monitoring device; combining the angle of the inclinometer and the Beidou satellite positioning data, and converting the position coordinates of the blade tip into actual spatial coordinates according to pre-calibrated parameters; calculating the clearance distance of each blade tip to the fan tower drum; filtering the clearance distance of each blade to calculate the difference value of the clearance distance of each blade, and taking the maximum value in the difference value as the imbalance degree of the impeller; when the clearance distance or the imbalance degree exceeds a preset threshold value, an alarm signal is sent and an alarm video is saved.

2. A method of monitoring a fan blade clearance distance as claimed in claim 1, wherein, The step of obtaining the fan blade movement image comprises the following steps: collecting environmental light information data, establishing a light evaluation model by using the environmental light information data to obtain light compensation parameters; controlling the monitoring device day and night mode and the light compensation lamp switch by using the light compensation parameters to obtain the fan blade movement image.

3. A method of monitoring a fan blade clearance distance as claimed in claim 2, wherein, The step of identifying the position coordinates of the blade tip in the fan blade movement image comprises the following steps: using a deep learning network algorithm to reduce the interference of environmental factors by using Zero-DCE, and identifying blade tip candidate points by using the deep learning network algorithm; positioning by using the blade tip candidate point information data in combination with constraint conditions to obtain the position coordinates of the blade tip.

4. A method of monitoring a fan blade clearance distance as claimed in claim 3, wherein, The step of converting the position coordinates of the blade tip into actual spatial coordinates comprises the following steps: establishing a pixel-to-space mapping model by using the coordinate conversion relationship information data in the pre-calibrated parameters to obtain static conversion parameters; modifying the static conversion parameters by using dynamic coordinate system compensation data to obtain dynamic conversion parameters; performing three-dimensional space projection calculation by using the position coordinates of the blade tip in combination with the dynamic conversion parameters to obtain the actual spatial coordinates of the blade tip.

5. A method of monitoring the clearance distance of a fan blade as claimed in claim 4, wherein, The step of obtaining the dynamic coordinate system compensation data comprises the following steps: calculating the angle deviation value by using the difference between the angle of the inclinometer and the reference angle data; calculating the position deviation value by using the difference between the Beidou satellite positioning data and the reference coordinate data; constructing an angle rotation matrix by using the angle deviation value, constructing a position translation matrix by using the position deviation value, and combining the angle rotation matrix and the position translation matrix to obtain the dynamic coordinate system compensation data.

6. A method of fan blade clearance monitoring as claimed in claim 5, wherein, The step of calculating the clearance distance of each blade tip to the fan tower drum comprises the following steps: performing three-dimensional space distance calculation by using the actual spatial coordinates of the blade tip and the tower center spatial coordinates in the pre-calibrated parameters to obtain a straight line distance value from the blade tip to the tower center, wherein the straight line distance value is calculated by using a three-dimensional Euclidean distance formula; performing clearance distance correction calculation by subtracting the tower radius parameter in the pre-calibrated parameters from the straight line distance value to obtain a preliminary clearance distance value; performing three-dimensional space compensation calculation by using the preliminary clearance distance value in combination with the dynamic coordinate system compensation data to obtain the clearance distance of each blade tip to the fan tower drum.

7. A method of fan blade clearance monitoring as claimed in claim 6, wherein, The step of calculating the imbalance degree of the impeller comprises the following steps: The clearance distance values corresponding to each blade are obtained by grouping the clearance distance from the blade tip to the fan tower tube according to the blade number; The stable clearance distance value of each blade is obtained by performing digital filtering processing on the clearance distance value corresponding to each blade, wherein the digital filtering processing eliminates measurement noise by performing weighted average calculation on the clearance distance values measured continuously for a single blade multiple times; The inter-blade clearance distance difference data is obtained by performing pairwise difference calculation on the stable clearance distance values of the three blades, and the maximum difference value is extracted from the inter-blade clearance distance difference data as the impeller unbalance degree.

8. A wind turbine blade clearance monitoring device applying a wind turbine blade clearance monitoring method according to any one of claims 1 to 7, characterized in that, It comprises: A protective shell; A high-speed camera arranged inside the protective shell for collecting fan blade motion images; An infrared fill light arranged inside the protective shell for providing fill light; An inclinometer arranged inside the protective shell for measuring real-time cabin swing angle data and initial installation angle; A Beidou satellite positioning system arranged inside the protective shell for obtaining tower tube position coordinate data; A temperature sensor arranged inside the protective shell for monitoring the working environment temperature of the device and starting the heater inside the device; A controller arranged inside the protective shell and connected with the high-speed camera, infrared fill light, inclinometer, Beidou satellite positioning system and temperature sensor for processing image data and sensor data, and performing blade tip identification and clearance distance calculation; A communication interface arranged on the protective shell and connected with the controller for outputting clearance distance and unbalance degree data to external devices; A fan blade tip identification algorithm module integrated in the software program inside the controller for real-time identification of the position of the blade tip on the image. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to realize the steps of the fan blade clearance distance monitoring method in any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the fan blade clearance distance monitoring method in any one of claims 1 to 7.

Images